Review of the Private Rented Sector in Wales · 2019-06-17 · 8.1 Absolute and relative size of the private rented sector in Wales: 1986, 1992 and 1997 8.2 Tenure aspirations of

Review of the PrivateRented Sector in WalesWelsh Assembly Government

Housing Research Report HRR 1/02 – Feburary 2002

Llywodraeth Cynulliad CymruWelsh Assembly Government

Further copies of this document can be obtained free of charge from:

Sarah Ballard

Housing Directorate

The National Assembly for Wales

Cathays Park

Cardiff

CF10 3NQ

Tel: (029) 2082 3353

E-mail: [email protected]

Website: http://www.wales.gov.uk/

ISBN 0 7504 2863 5 February 2002 © Crown Copyright 2002

Designed by CartoGraphics Typeset by Text Processing Services

G/331/00-01

Welsh Assembly Government – Review of the Private Rented Stock in Wales

i

Acknowledgments iii

List of figures 1

List of Tables 3

1. Introduction 5

2. Methodology 7

3. Background 11

4. Private Rented Sector Stock 19

5. Private Rented Sector Tenants 33

6. The Role of Housing Benefit in supporting the Private Rented Sector 41

7. Landlords, Housing Grants and Tenancy Types 55

8. Changes in the Private Rented Sector 61

9. Future Research and Monitoring 81

10. Conclusions 87

References 91

Appendix 1 - Data Review 93

Contents


ii

iii

Welsh Assembly Government - Review of the Private Rented Stock in Wales

We would like to thank all members of the steering group for their support, assistance andadvice throughout the duration of the work, in particular Mike Harmer, Sioned Lewis, AlanMorgan, Henry Small and James Watkins. We would also like to thank all those who gavetheir time to be interviewed.

Acknowledgments


iv


1

4.1 Percentage of households private renting by local authority, 1997

4.2 Tenure distribution by local authority, 1997

4.3 Private Rented share of all rented by local authority, 1997

7.1 Type of Private Landlords in Wales, 1997

8.1 Weekly rent, Britain (excl London) and Wales, 1996-2000

8.2 Real Terms weekly rent, Britain (excl London) and Wales, 1996-2000

8.3 Rental yield, Britain (excl London) and Wales, 1996-2000

8.4 Average recorded rent, standardised rent and real standardised rent in Wales, 1970-1999

LIST OF FIGURES

2


3


2.1 Qualitative interviews carried out

4.1 Shared dwellings by tenure, 1991

4.2 Type of dwelling by tenure, 1997

4.3 Date of construction by tenure, 1997

4.4 Number of rooms by tenure, 1997

4.5 Overall condition of property by tenure, 1997

5.1 Household size by tenure, 1997

5.2 Gross household income by tenure, 1997

5.3 Household type by tenure, 1997

5.4 Marital status by tenure, 1997

5.5 Age of residents by tenure, 1997

6.1 The Housing Benefit caseload: private renters in Wales, 1992-2000

6.2 Average weekly eligible rent and Housing Benefit among private renters in Wales, 1996-2000

6.3 Annual household income of private tenants by receipt of HB

6.4 Household savings of private tenants by receipt of HB

6.5 Private sector rents

6.6 Private sector rents by receipt of HB

6.7 Visible property condition by receipt of HB

6.8 Private tenants’ view of the need of repair on their property byreceipt of HB

6.9 Whether private renting households have their name on a social housing waiting list by receipt of HB

8.1 Absolute and relative size of the private rented sector in Wales: 1986, 1992 and 1997

8.2 Tenure aspirations of social and private renters

8.3 Social housing waiting list by tenure

8.4 Proportion of dwellings unfit by tenure, 1993 and 1998

LIST OF TABLES

4

8.5 Average repair cost per dwelling to meet fitness standards by tenure, 1993 and 1998

8.6 Size of privately rented dwellings and all dwellings, 1992 and 1997

8.7 Long-term change in dwelling size, dwellings with central heating and furnished dwellings in the private rented sector in Wales

8.8 Household size in the private rented sector and of all households, 1992 and 1997

8.9 Age of head of household and household type for the private rented sector and all households/individuals, 1992 and 1997

8.10 Average weekly rents, by property type and local authority, 2000 Q4

8.11 Average annual percentage change (discounted) in rent minus inflation by property type and local authority


5

1.1 Introduction

The University of Glasgow wascommissioned by the National Assembly forWales to carry out a Review of the PrivateRented Sector in Wales.

Task Group 2 working on the NationalHousing Strategy for Wales recommendedthat this research be carried out. Thisrecommendation was made against abackground of concern about the housingconditions found in the sector, and asuspicion that the sector is important forkey niche groups such as young vulnerablesingle people, reforming households and soon. In addition, investment in propertiestends to be low, possibly caused by acombination of low returns, and restrictionssince 1996 on the level of Housing Benefitpayable for some tenants and privatelyrented properties.

The over-arching concern of the NationalAssembly for Wales in commissioning thisresearch, however, was that the existingevidence base concerning the privaterented sector in Wales was lacking.Therefore, this report seeks to make amodest contribution to fill that gap.

1.2 Aims and Objectives

The broad aims of the research were toidentify specific problems in the privaterented sector in Wales and gaps inknowledge which could usefully be thesubject of future research. In addition, theNational Assembly for Wales specified thatthe contractor should makerecommendations on which aspects of thesector should be regularly monitored and

how this might be achieved. Data sourceswere also to be reviewed with a view tohelping the research team draw conclusionsas to what research could realistically becarried out in the future.

The specific objectives of the research wereto assess:

1 the scale, spatial distribution andcharacteristics of privately rentedproperties,

2 the characteristics of tenants,

3 the role of Housing Benefit insupporting the private rented sector,

4 the characteristics of landlordscatering for different segments of themarket, their reasons for involvementand views on how investment couldbe increased,

5 the role that housing grants tolandlords play in sustaining the sector,

6 the relative importance of differenttypes of tenancy agreement, and

7 how the private rental sector ischanging and has changed overrecent years.

1.3 Approach

Descriptive analysis of the private rentedsector using some of the quantitative datasources which were reviewed has beencarried out. This draws mainly, but notexclusively, on: the 1991 Census ofPopulation; 1992-1999 DSS/DWP HousingBenefit data; 1970-1999 FamilyExpenditure Survey; 1996-2000JRF/University of York Indices of Private


Chapter One Introduction

Rents and Yields; and the 1998 WelshHouse Condition Survey.

Some of this analysis is for all of Wales,while some has been carried out at finergeographies. The scale and spatialdistribution of the private rented sector hasbeen analysed at local authority level, ashave data relating to Housing Benefit andinformation on private rents. The scale anddistribution of shared dwellings using the1991 Census has been analysed at wardlevel.

Qualitative interviews were carried outacross Wales with people in the public andprivate sectors with a knowledge andinterest in the future of the private rentedsector. A limited number of previousstudies into particular aspects of the privaterented sector in Wales were also identifiedand reviewed.

Twenty-six qualitative interviews werecarried out with a number ofEnvironmental Health Officers, HousingAdvice Centres, Housing Strategy/PolicyOfficers, landlord representativeorganisations, letting agents, Rent Officersand University Accommodation Officers.Locally based interviews were carried out inAnglesey, Merthyr Tydfil, Powys andSwansea.

1.4 Structure of this Report

After explaining the methods used (chaptertwo), this report goes on to provide abackground to the private rented sectorand the Housing Benefit system in the UK

and Wales (chapter three). Considerationis given to how this research relates to thedevelopment of the National HousingStrategy for Wales and the on-goingdebate in Wales regarding the future ofprivate rental housing.

The chapters containing the main empiricalfindings (chapters four to eight) eachconform to the following broad structure:

• Data Review – outlining the relevantinformation contained in variousdatasets

• Data Analysis – presenting the dataanalysis carried out on selecteddatasets

• Qualitative Analysis – presenting thefindings of the interviews carried out

• Future Research – presenting ideas forfuture research as suggested byinterviewees and/or designed by theresearch team.

Chapter nine recaps the ideas for futureresearch before making specificrecommendations regarding theforthcoming Welsh Household andDwelling Survey. Lastly, a means tomonitor the private rented sector in Walesby centralising the data collected by theRent Officer Service is outlined.


6

2.1 Introduction

This chapter outlines the approaches takenby the research team to tackle the sevenobjectives of this research. The researchteam’s approach consists of four mainstrands of work, which are reflected in thestructure of the remainder of this chapter:

• a literature review;

• a data review and analysis;

• the selection of case study localauthorities; and

• qualitative interviews and analysis.

2.2 Literature Review

Three main methods were used to identifyprevious research into the private rentedsector in Wales. Firstly, electronic searcheswere performed of three databases.Secondly, enquiries were made by letter ofthree key Welsh organisations with aninterest in housing. Finally, telephoneenquiries were made of the Department ofCity and Regional Planning at CardiffUniversity.

The following three bibliographicaldatabases were electronically searched: theSocial Science Citation Index; the PlanningExchange; and Glasgow University Librarycatalogue. The Social Science CitationIndex (accessed through the Web ofScience, formerly through BIDS) is abibliographical database of academicarticles dating from 1981 and is updatedweekly. The Planning Exchange is abibliographical database of academicarticles, discussion papers, consultants’reports and various types of policy

documents. The Glasgow UniversityLibrary catalogue searches all materiallocated in all libraries of the University ofGlasgow.

Enquiries by letter were made of three keyWelsh organisations with an interest inhousing. These were: the CharteredInstitute of Housing in Wales (CIH Cymru);Shelter Cymru; and the Institute of WelshAffairs (which is not an exclusively housingoriented organisation, but carried outresearch into housing in Wales in 1997).

Finally, enquiries by telephone were madeof the Department of City and RegionalPlanning at Cardiff University. Thisenabled access to two post-graduatedissertations from the masters course inhousing which relate to aspects of theprivate rented sector in Wales.

2.3 Data Review and Analysis

The purpose of the data review wasthreefold. First, to provide an overview ofthe available data relating to the privaterented sector. An account of the contentsof a number of datasets relating to theprivate rented sector can be found inappendix one.

Second, to gain access to a number ofdatasets in order to provide a description ofthe private rented sector in Wales. Theresults of this are presented in the mainbody of the report.

The final purpose of the data review was toallow the research team to makepracticable suggestions for future researchand monitoring. These are presented


7

Chapter TwoMethodology

8

towards the end of each of chapters fourto eight.

The following datasets were reviewed:

1 Census of Population, Local BaseStatistics & Small Area Statistics, 1991

2 DSS/DWP Housing Benefit data,annually from 1992

3 Family Expenditure Survey, annuallyfrom 1961

4 General Household Survey, annuallyfrom 1971

5 JRF/Uni of York Index of Private Rents& Yields, available quarterly from1996Q1

6 Student Accommodation data,1998/99

7 Welsh Health Survey, 1998

8 Welsh House Condition Survey, 1998

9 Welsh Index of Multiple Deprivation,2000

2.4 Selection of Case Study LocalAuthorities

Four case study local authority areas wereselected in which to carry out the locallybased qualitative interviews (for example,interviews with Environmental HealthOfficers, Housing Advice Centres, HousingStrategy/Policy Officers, etc.). It wasdecided to select case study areas on twoprincipal grounds. Firstly, it grounds thequestions in a particular setting to ensurerealism. Secondly, selecting contrastingcase study areas ensures that a range ofexperiences and circumstances encounteredacross Wales are heard.

The four ‘case study’ areas are notintended to provide a detailed picture ofthe private rented sector in each. Rather,they are intended to provide a focus forthe interviews.

The criteria used to select case study areaswere population density and the proportionof the population in receipt of statebenefits. These criteria were chosen tohelp ensure that the four case study areasrepresented a range of different types ofarea found across Wales. Local authorities’population density was plotted against thepercentage of the population in each areain receipt of key social security benefits. Indiscussion with the National Assembly forWales, the following four local authorityareas were chosen in which to baseinterviews:

• Anglesey,

• Merthyr Tydfil,

• Powys, and

• Swansea.

2.5 Qualitative Interviews

Seven different interview topic guides weredeveloped for the following interviewees:

Local area interviewees

1 Environmental Health Officers

2 Housing Advice Centres

3 Housing Strategy/Policy Officers

4 Letting agents

5 Rent Officers

6 University Accommodation Officers


9

National interviewees

7 Landlord representative organisations

8 Large letting agents (the same topicguide as for local letting agents withminor adjustments to wording)

Semi-structured qualitative interviews werecarried out covering the nature of theprivate rented sector, changes to the sectorover the previous five years, problems withthe sector including tenancy and repairissues, the operation of the Housing Benefitsystem, issues relating to Houses inMultiple Occupation, and policies whichcould be amended or introduced to helpimprove or expand the sector. Face-to-faceinterviews were arranged over thetelephone from Glasgow prior to travellingto Wales. Initial contact was made withthe names provided by the NationalAssembly for Wales. In some cases, thesepeople themselves were interviewed, whilein others they provided the name of analternative or additional person qualified toanswer the questions. Most of the localletting agents were identified while in thefield via the local press and by utilising thelocal knowledge of other interviewees.

Table 2.1 summarises the interviews whichwere carried out. Both the ‘national’letting agents were part of chains which

have, or had, offices across Wales, but inpractice it was not possible to identify aperson with all-Wales knowledge, thereforethese interviews predominantly reflect theexperience of the Cardiff offices. It wasattempted to interview the Welsh regionalrepresentative of the UK-wide Associationof Residential Letting Agents (ARLA), butthis position was vacant at the timeinterviews were carried out. One of thelandlord representative organisationsinterviewed had members across NorthWales, and the other had members basedmainly in and around Cardiff.

One of the ‘Powys’ letting agents wasactually based in Aberystwyth, althoughmanaged some properties in Powys. TheAnglesey letting agent and one of theSwansea letting agents completed theinterview topic guide in writing andreturned by post. All other interviews wereconducted face-to-face.

Interviewees where possible were sent acopy of the topic guide prior to theinterview in order to consider the issuesbefore hand and so be able to providemore comprehensive responses. Interviewswere tape recorded and detailed notes andquotations written up subsequently. Thesenotes were then thematically coded.


Table 2.1: Qualitative interview

s carried out

10

Interviewee Area

National Anglesey Merthyr Tydfil Powys Swansea Total

Environmental Health Officer NA 1 1 1 - 3

Housing Advice Centre NA 1 1 1 1 4

Housing Strat/Pol Officer NA 1 1 1 1 4

Letting agent 2 1 1 2 2 8

Rent Officer NA 1 1 1 1 4

University Acc. Officer NA NA NA NA 1 1

Landlord representative 2 NA NA NA NA 2

TOTAL 4 5 5 6 6 26

Welsh A

ssembly G

overnment – Review

of the Private Rented Stock in Wales

11

3.1 Introduction

This chapter provides contextualbackground to the private rented sector inthe UK and outlines the context in Wales inwhich this review of the private rentedsector was commissioned. Closelyassociated areas of public policy are alsooutlined, namely Housing Benefit, tenancylegislation and housing grants. Finally, fourspecific studies relating to particular aspectsof the private rented sector in Wales whichwere identified during our literaturesearches are summarised.

3.2 The Private Rented Sector in theUK

The private rented sector has animportance for housing policy that faroutweighs its small numerical size. It isgenerally recognised that the sector has animportant role in providing accommodationfor a number of important niche groupswithin the housing market. As well as adeclining group of elderly people, thesector accommodates young peopleincluding students living away from theparental home, and people living in tiedaccommodation. In addition to providingthe first home for many newly formedhouseholds, the private rented sector alsoaccommodates households who are unable,at least for the time being, to gain accessto social housing, many of whom aredependent upon Housing Benefit to affordtheir rent. The sector also eases theoperation of the owner occupied sector byallowing movers to sell without necessarilyhaving found a dwelling suitable forpurchase, thus helping to prevent theaccumulation of housing chains. A large

and efficient private rented sector cantherefore have significant economicbenefits by enhancing labour mobility,allowing workers to move swiftly inresponse to changing labour marketconditions (Maclennan and Pryce, 1996).

In addition, the importance of the sectormay be set to grow as the number ofhouseholds rises over the next twodecades. This anticipated demographicchange has raised questions over whethermany of the new smaller households willbe able to gain access to owner occupancygiven that house prices in many areas haverisen to the point where only householdswith more than one earner can afford thenecessary mortgage payments, althoughthis is generally less true in Wales than inthe rest of the UK. Combined withsignificant shifts in the labour markettowards more flexible (and less stable)forms of employment contract, thesustainability of a tenure structuredominated by owner occupancy andmortgage finance has diminished, raisingthe economic importance of the privaterented sector (Maclennan and Pryce, 1996;Bramley, Pryce and Satsangi 1999).

However, notwithstanding these importantroles, the private rented sector is beset by anumber of serious problems. In particular,the sector features some of the worsthousing conditions in the UK and in Wales.

In addition to poor house conditions, thereare concerns about the quality ofmanagement by some private landlords.The establishment and growth of theAssociation of Residential Letting Agents


Chapter ThreeBackground

12

(ARLA) has helped to raise standards andmake the management of the sector moreprofessional. The Buy-to-Let schemesponsored by ARLA has also helped in thisregard. However, many landlords managetheir property themselves, often in theirspare time rather than as a full-timebusiness. While some of these sidelinelandlords have the requisite skills tomanage their properties well, others do not(Bevan, Kemp & Rhodes, 1995; Crook,Hughes & Kemp, 1995). Although recentgovernments has sought to modernise theownership of the private rented sector, todate the financial institutions have beenreluctant to invest in this market.

House conditions and managementstandards are particularly problematic inhouses in multiple occupation (HMOs).The introduction of compulsory licensing,which has been proposed in the 2000Housing Green Paper but has yet to begiven parliamentary time, should help toimprove physical and managementstandards within this small but importantpart of the privately rented market.However, there are also concerns that theintroduction of licensing may prompt someHMO landlords to withdraw from themarket. This may increase homelessnessamong the more vulnerable tenants on themargins of the housing market. Moreover,the Westminster proposals for licensingcover only HMOs, yet the problems ofpoor standards and inadequatemanagement are by no means confined tothis part of the private rented sector (Crooket al, 2000). The National Assembly forWales have, however, recently issued aconsultation paper on selective licensing ofnon-HMO properties in the private rentedsector.

The student sub-sector of the privaterented housing market is also a cause forconcern in some respects. The changes instudent finance (the replacement of grantsby loans, the introduction of tuition fees,and the withdrawal in September 1990 ofeligibility for Housing Benefit) have raisedconcerns about student poverty and debt.The evidence suggests that students livingin privately rented accommodation have tospend a disproportionate share of theirresources on rent. There are relatedconcerns about overcrowding, about thecleanliness and condition of student rentalhousing, about anti-social behaviour bystudent households, and about the impactof student lettings on local propertymarkets (Rugg et al, 2000).

In recognition of the important roles thatprivate renting performs, but mindful ofthe problems that characterise much of thesector, recent governments, and now theNational Assembly, have sought to furthermodernise this part of the housing market,building upon existing initiatives, not leastthe past and present role of housingimprovement grants in Wales. However, asthe National Assembly for Wales Task Force(2000b) report noted, the development ofeffective policy has been hindered by therelative lack on an adequate evidence baseon the private rented sector in Wales.Although there has been a considerableamount of research on the private rentedsector in recent years, most of this workhas been focused on Britain as a whole,England, or Scotland, rather thanspecifically on Wales.

3.3 The Role of Housing Benefit

Concerns exist about the difficulties thatpeople in receipt of social security benefitsface in gaining access to, and being able to


13

afford the rents charged for, privatelyrented dwellings. The difficultiessurrounding the administration of HousingBenefit and the impact of restrictions oneligible rents appear to have made someprivate landlords reluctant to let theiraccommodation to tenants in receipt ofHousing Benefit. These restrictions haveled to shortfalls between the rentsconsidered eligible for Housing Benefit andthe rents charged by private landlords(Kemp & McLaverty, 1995; Kemp & Rugg,1998; Murphy, 1998).

In some cases, these restrictions in eligiblerents, not to mention delays in theprocessing of Housing Benefit claims, havemade it more difficult for ‘Access (Bond)Schemes’ to persuade private landlords toaccept homeless people as tenants (Rugg,1997). In addition, as research by NACAB(1998) and others has shown, thedifficulties that some private tenants face ingetting some of the less scrupulouslandlords to return bonds has also been aconcern for policy makers. Theeffectiveness of various tenant depositschemes around the UK remains to beseen.

Since 1996, restrictions have been in placeon Housing Benefit payment in threeregards. Firstly, Housing Benefit will onlybe paid up to the going market rent for aparticular property, as valued by the RentOfficer. This is designed to preventlandlords and/or tenants making a profitout of inflated claims for Housing Benefit.Secondly, Housing Benefit is only payablefor the number of rooms that a householdrequires. Thirdly, Housing Benefit is onlypayable up to a maximum ceiling of theaverage local rent for a given size and typeof property. This is designed to preventHousing Benefit recipients living in luxury

accommodation. This ceiling for differentsizes and types of property is know as theLocal Reference Rent and is set by the RentOfficer with regard to market evidencebased on private lets to people not inreceipt of Housing Benefit.

More stringent restrictions apply to peopleunder the age of 25 years. Theseindividuals are subject to the Single RoomRent which restricts Housing Benefitentitlement to only cover a ‘single room’rather than a whole self-containeddwelling. This means that people under 25years on Housing Benefit are restricted tobed-sit type properties. However, as fromJuly 2001 the definition of accommodationthat is not self-contained was expanded toinclude cases where there is a sharedlounge area, i.e. will include ‘house share’arrangements. This means the SingleRoom Rent set by Rent Officers should riseto mean people under 25 years on HousingBenefit could afford to live in a ‘houseshare’.

Prospective tenants looking foraccommodation are entitled to ask for aPre-Tenancy Determination. This isperformed by the Rent Officer and tells theprospective tenant what the maximumeligible rent for a particular property will befor Housing Benefit purposes. This isintended to help people decide if they canlive in a particular property before enteringinto a contract with the landlord.

3.4 Tenancy Types

Before 1989, rent levels in the privaterented sector were regulated and tenantshad greater security of tenure than they donow. The 1988 Housing Act introducedassured shorthold tenancies which wereaimed at the private rented sector. Since


then, almost all private lets have beenassured shorthold tenancies. Indeed,without signing a contract, they are thedefault legal position.

An assured shorthold tenancy locks boththe landlord and tenant into a minimumperiod of six months. Thereafter, thelandlord is required to serve two months’notice to the tenant to vacate the property(the two months notice can be served fourmonths into the tenancy so that theminimum of six months residency will stillbe achieved). The tenant is required toserve one month’s notice to the landlord oftheir intention to vacate a property(similarly, this can be served five monthsinto a tenancy). After the initial six monthperiod, the tenancy is said to be ‘periodic’.

3.5 Housing Grants

The Local Government and Housing Act of1989 saw increased public sectorinvestment in private sector residentialproperties in the form of grants. Waleswas a particular beneficiary of this policydue to the greater incidence of unfitnessand disrepair found there than in the restof the UK. Between 1986 and 1996, threetimes the number of grants were providedper 1,000 private dwellings (owneroccupied as well as privately rented) inWales than in England. In addition, theaverage grant paid was nearly twice ashigh in Wales (Bourton and Leather, 2000).

However, the mechanisms for allocatingfunds for housing renewal grants havechanged since devolution in Wales, and theabolition of mandatory renovation grantsby the Housing Grants, Construction andRegeneration Act of 1996. Since then,local authorities have the responsibility to

develop a strategy to deal with unfitnessand disrepair in their own areas. GeneralCapital Funding in Wales continues to rise,but local authorities may choose to allocateless funds to housing grants, reflectingcompeting local priorities.

3.6 The Welsh Context

A notable feature of housing (of alltenures) in Wales is the wide range ofgeographical variation in housingconditions and housing markets foundacross Wales. This makes it difficult togeneralise about housing in Wales, andunderlines the need for reliable data atsub-Wales level.

Another feature of housing in Walescompared to the rest of the UK is the ageof the housing stock. The 1998 WelshHouse Condition Survey shows that justunder a third of dwellings in Wales wereconstructed before 1919 compared to justunder a quarter in England. Privatelyrented properties are over-representedamong older properties.

Another feature is the relatively high levelof owner occupation in Wales which is inpart a result of the ‘building clubs’prevalent in the late nineteenth century,particularly in the Valleys (Fisk, 2000).More recently, low residential propertyprices in Wales have helped sustain this(Williams, 2000a). However, increasinghouse prices in some parts of Wales nowmean that an increasing number of peoplecannot afford to enter owner occupationand therefore choose the private rentedsector in its place (Holmans, 1996). Thisraises a question of meeting housing needwith accommodation of a decent standard,as the better quality properties in the

14


15

private rented sector tend not to accepttenants in receipt of Housing Benefit(Institute of Welsh Affairs, 1997).

Overall, 8.6% of dwellings in Wales werein the private rented sector in 1999, lowerthan 11.1% of dwellings in England buthigher than the 6.8% of dwellings inScotland (Williams, 2000a). The proportionof dwellings in the private rented sectorwas in long term decline in Wales as in therest of the UK until it rose slightly in theearly 1990s but has stabilised throughoutthe late 1990s (Williams, 2000b).However, because the number ofhouseholds and dwellings in Wales is rising,this constant proportion masks a continuedgrowth in the absolute number of privaterented properties. According to 1998Welsh House Condition Survey, there were88,535 privately rented dwellings in Wales.

The Welsh and English House ConditionSurveys showed that in Wales 8.5% ofhousing in 1998 was unfit for humanhabitation compared to 6.5% in England in1996. This is the result of the age of thestock, the wet climate in Wales, lowproperty values reducing the incentive toinvest, and because low incomes in Walescombine with a relatively high rate ofowner occupation to mean that owners areless likely to be able to afford repairs(Bourton and Leather, 2000). The rate ofunfitness found in the private rented sectorin Wales was 18.4% in 1998 compared to7.6% found in the owner occupier sector.

The devolution of additional decisionmaking functions and the power to passsecondary legislation to the NationalAssembly for Wales established in 1999poses new housing challenges as well aspresenting opportunities for change. The

increased accountability brought bydevolution has raised the public’sexpectations of change and improvement,particularly in housing (Gallent andTewdwr-Jones, 2000).

The establishment by Win Griffiths MP in1998 of the National Consultative Forumon Housing has led to the development ofthe National Housing Strategy for Wales bythe Forum, the Welsh Office and now theNational Assembly for Wales. Task GroupTwo which worked on developing thestrategy in relation to private housing(owner occupation and private renting)made five recommendations in relation tothe private rented sector in Wales. Thesewere that:

1 The National Assembly be given thepower to introduce its own model oflicensing of HMOs within any primarylegislation proposed by Westminster.[However, licensing of HMOs was notin the 2001 Queen’s Speech].

2 The National Assembly initiateresearch on the private rented sectorin Wales. [i.e. this review is the firstpiece of such research].

3 The Assembly should ensure thatthere is equal access to bond schemesacross Wales.

4 Research be carried out by theAssembly to establish best practice ondeposit management schemes.

5 The Assembly investigates possiblemeans to help bring about theabolition of the Single Room Rentwhich restricts Housing Benefiteligibility to people under the age of25 years.


The subsequent consultation paper issuedby the National Assembly in December2000 broadly accepted 1), 2) and 3). Inaddition, the National Assembly made acommitment to consider the viability of alandlord accreditation scheme in the lightof DETR/DTLR research into this. Inrelation to 5), note that from July 2001,the definition of a ‘single room’ has beenbroadened to include rooms in houses witha shared living room, thus increasing therange of accommodation available topeople in receipt of Housing Benefit underthe age of 25 years.

The subsequent National Housing Strategyfor Wales document reported that therewas strong support from consultees of:mandatory licensing of HMOs but in amanner that would not restrict supply; alandlord accreditation scheme; an increasein access to bond schemes; and of the rolethat the private rented sector has ingeneral in meeting housing need. Thisdocument was debated by the NationalAssembly on 12th July 2001 and is nowformal policy. In addition, as noted insection 3.2, the National Assembly haverecently issued a consultation paper onselective licensing of non-HMO propertiesin the private rented sector.

3.7 Previous Studies of the PrivateRented Sector in Wales

There is a dearth of research into theprivate rented sector in Wales. Fourspecific pieces of work were found, two ofwhich are dissertations carried out as partof the post-graduate course in housingoffered by the Department of City andRegional Planning at Cardiff University.One other is a study commissioned by the

Housing Association Charitable Trust ofolder private tenants in Carmarthenshireand Pembrokeshire. The final studyidentified by the research team is a surveyof advertisements for private lets in Alynand Deeside in 1992 and 1993 carried outby Shelter Cymru. These four studies arenow summarised in turn.

3.7.1 Gaining and maintaining privaterented accommodation for young singlehomeless people

This housing dissertation by Daniel (1999)examined the problems that young singlehomeless people encounter in securingprivately rented accommodation as well asremaining in that accommodation.Interviews were carried out with BondBoards, ‘support workers’ and ‘landlorddevelopment workers’ in Cardiff, Vale ofGlamorgan and Rhondda Cynon Taff. Inaddition, a survey of tenants and landlordsusing the Bond Boards was carried out.

This study found evidence of people in Valeof Glamorgan under 25 years on HousingBenefit having to share rooms because ofthe restrictions of the Single Room Rent.This resulted because of a shortage of bed-sit type accommodation in this area, soyoung people end up having to share aroom in a house together.

The reasons most commonly cited byyoung homeless people for not being ableto secure private rented accommodationwere: ‘unable to pay deposit and rent inadvance’, ‘prejudice towards homelesspeople’, ‘lack of accommodation inpreferred area’, and ‘could not affordfurniture’.


16

3.7.2 Access to the private rented sectorfor low income people

This housing dissertation by Thomas (1995)is an evaluation of the Cardiff Bond Board.The main methods used were: interviewswith a housing advice provider, a lettingagent, landlords, Social Fund Officers, andpeople involved with the Bond Board; asurvey of all Citizens Advice Bureaux inSouth East Wales; and a survey of bondees.

The main research findings were: only asmall proportion of clients were ex-hosteldwellers (i.e. a major target group); hostelswere not fulfilling their role as participatingagencies; the Housing Help Centre was themain source of information about the BondBoard for tenants; ALMA, Cardiff Flatshopand the local press were the main sourcesof information for landlords; rent arrearsamong Bond Board clients were low andthese were mainly caused by delays in thepayment of Housing Benefit; and the BondBoard encouraged good management bylandlords generally.

3.7.3 Older private tenants and landlordsin Carmarthenshire and Pembrokeshire

This was a survey of 13 landlords, 36tenants and 73 owner occupiers, all overthe age of 60. The survey was carried outbetween 1998 and 2000. The mainpurpose of the study was to establish whatproblems older landlords and tenants hadin maintaining their properties compared toowner occupiers.

The survey found that half of tenants hadlived in their current home for more than20 years compared to two-thirds ofowners. This suggests that older ownersare more settled than older private tenants.

Both tenants and owners underestimatedthe extent of work required to be done totheir homes to bring them up to a standardof good repair. However, tenantsunderestimated this to a greater extent.

Eighty-nine per cent of private tenants saidit was ‘very important’ to them to stay putwhere they lived. The area and neighbourswere more important than the house inwhich they lived. Many older tenants werecautious about asking their landlord tocarry out repairs for fear of eviction, evenin cases where this risk was perceived to besmall.

Only 20% of private tenants were awareof Care and Repair Cymru services,compared to 40% of owners. In addition,renters were less likely to seek assistancefrom Care and Repair due to the need todiscuss the process with their landlord.

Most of the landlords had not heard ofCare and Repair, but half said that theythought their older tenants would benefitfrom using Care and Repair. The same halfalso said they might contribute to the costof the service.

3.7.4 The availability of the private rentedsector to families in Alyn and Deeside

This work reports the results of a surveycarried out by Shelter Cymru whichcollected information on advertised privatelets in Alyn and Deeside. The survey wasinitially carried out in September 1992 andwas repeated in September 1993. Overthe two years, 691 lets were sampled. Thesurvey was mainly concerned with theaccessibility of the private rented sector tofamilies. The key findings were:


17


18

• shortholds increased from 88% ofnew lets for family sizeaccommodation in 1992 to 94% in1993,

• almost all single person lets wereshortholds,

• 97% of new lets surveyed in 1993required a deposit compared to 95%in 1992,

• 84% of new lets in 1993 stated ‘noDSS’ compared to 80% the previousyear, and

• median asking rents for two bedroomhouses increased from £289 permonth to £324 per month, a rise of12%.


19

4.1 Introduction

This chapter is organised into two parts.The first covers the scale and spatialdistribution of the private rented sector inWales. The second covers thecharacteristics of properties.

Each part of this chapter and chapters fiveto eight are structured under the headingsoutlined in section 1.4. In summary, theseare: data review; data analysis; qualitativeanalysis; and future research. Each sectionor chapter finishes with a summary offindings.

4.2 Scale and Spatial Distribution

4.2.1 Data Review

4.2.1.1 1991 Census of Population

Given near 100% coverage of thepopulation, this can provide information onthe precise spatial distribution ofhouseholds and persons in the privaterented sector in Wales. This can be doneat the level of all Wales, former counties,former districts, wards or EnumerationDistricts (EDs) using the Small AreaStatistics (SAS) and the Local Base Statistics(LBS).

Most Census tabulations refer tohouseholds or persons, but a small numberof tables in the LBS refer to dwellings. Forexample, LBS Table 64 provides informationon the number of shared and non-shareddwellings, disaggregated by tenure.

4.2.1.2 1998 Welsh House Condition Survey

The 1998 Welsh House Condition Survey(WHCS) is formed of two sub-surveys.First, a household interview survey whichwas carried out in 1997 with just over40,000 households. Second, a physicaldwelling inspection by a surveyor whichwas carried out in 1998 on around 12,000of the dwellings included in the householdsurvey.

The household survey includes informationon the geographical location of dwellings,households and persons at various spatialscales. These include various regionalclassifications within Wales (e.g. TEC areas,Health Authority areas, local authorities,parliamentary constituencies, postcodeareas e.g. CF and postcode districts e.g.CF1). Note, however, that the WHCS wasdesigned only to provide robustinformation down to local authority levelfor most variables. Therefore, a reliableestimate of the housing tenure structure ofdifferent local authority areas can be made,but further disaggregations within theprivate rented sector will result in smallnumbers of observations in some cells.

4.2.1.3 1998 Welsh Health Survey

The 1998 Welsh Health Survey (WHS)provides information on a random selectionof 30,000 individuals. Information iscollected regarding housing tenure. TheWHS and the WHCS could, in theory, becombined to provide a larger dataset butthe weights used in each survey to factor-up the results to reflect the age, gender

Chapter FourPrivate Rented Sector Stock

and geographical distribution of populationin Wales would need to be merged.

4.2.2 Data Analysis

The results of the 1998 Welsh HouseCondition Survey indicate that 88,535households, representing 7.7% of allhouseholds in Wales, rented theiraccommodation from a private landlord.This compares to 21.6% of householdsrenting from a ‘social’ landlord (mainlylocal authorities and housing associations),and 70.8% who own their home (includingthose still repaying a mortgage or otherloan).

However, the prevalence of the privaterented sector (PRS) varies across localauthority areas from 2.3% of households inTorfaen to 13.0% in Powys. Figure 4.1shows the percentage of households ineach local authority area in Wales living inthe PRS. There is a clear geographicalpattern to the prevalence of the PRS inWales. The local authorities of mid-Wales,south-west Wales and the north-west (withthe exception of Carmarthenshire) have anabove average prevalence of the PRS.

None of these has less than 10% ofhouseholds living in the PRS. In contrast,the local authorities with below averagerepresentation of the PRS cover the SouthWales Valleys and industrial towns such asPort Talbot, Bridgend and Newport. Localauthorities with around the Welsh averageproportion of households living in the PRSare Cardiff, Swansea, the Vale ofGlamorgan, Carmarthenshire andMonmouthshire. Flintshire and Wrexhamstand out from this broad North/Southpattern, both with below averagerepresentation of the PRS, both with 5.8%of households in the PRS.

Part of the reason for commissioning thisresearch was to investigate whether thePRS could be expanded. A key question,therefore, is why is the PRS more prevalentin some locations than others? If someinsight to this question can be provided, itmay help identify the conditions in Waleswhich would enable the PRS to expand. Itis therefore useful to consider which otherhousing tenure is most likely to currentlyhouse people who may have, given slightlydifferent circumstances, chosen the PRS.


20 Source: WHCS, 1998

Figure 4.1 Percentage of households private renting by local authority 1977

(Wales = 7.7% PRS)

0%2%4%6%8%

10%12%14%

Powys

Gwyned

d

Ceredig

ion

Denbig

hshir

e

Isle o

f Ang

lesey

Pembr

oksh

ire

Conwy

The V

ale of

Glam

orga

n

Cardiff

Carmart

hens

hire

Mon

mouth

shire

Swan

sea

Newpo

rt

Rhond

da C

ynon

Taff

Flint

shire

Wrex

ham

Bridg

end

Mert

hyr T

ydfil

Caerp

hilly

Blaen

au G

went

Neath

Port

Talbo

t

Torfa

en

% o

f ho

useh

olds

priv

ate

rent

ing

The distribution of households across thethree main housing tenures by localauthority is shown in figure 4.2. In Walesas a whole in 1997, 70.4% of householdslived in owner occupied properties, 21.9%lived in the social rented sector, and 7.7%in privately rented accommodation. Localauthorities in figure 4.2 are ranked indescending order by the proportion ofhouseholds in the private rented sector, asis figure 4.1.

Figure 4.2 shows a moderate cross-sectional (local authority level) inverseassociation between the proportion ofhouseholds in the social rented sector (SRS)and the proportion in the PRS. This maymean that one acts as a substitute for theother, although many factors lie behind theproduction of this statistical association.

Local authorities with a high prevalence ofowner occupation generally tend to have aslightly lower than average prevalence ofsocial renting. This is consistent with thesetenures being substitutes for one another,i.e. if earnings and employment rates werehigher, more people would buy their ownhome, but being unable to do so, opt forthe social rented sector.

Figure 4.3 shows the proportion of allrenting households that rent from a privatelandlord. It ranges from around 7% inTorfaen to nearly 50% in Ceredigion.Comparing figures 4.1 and 4.3, there is astatistical association between the privaterented sector’s share of the rented sectorand its share of all housing in differentareas.


21

Figure 4.2 Tenure distribution by local lauthority, 1997

0%10%20%30%40%50%60%70%80%

OO %

SR %

PR %

% o

f household

s

Pow

ys

Gw

ynedd

Cere

dig

ion

Denbig

hsh

ire

Isle

of A

ngle

sey

Pem

bro

kesh

ire

Conw

y

The V

ale

of G

lam

org

an

Card

iff

Carm

art

hensh

ire

Monm

outh

shire

Sw

anse

aN

ew

port

Rhondda C

ynon T

aff

Flints

hire

Wre

xham

Bridgend

Mert

hyr Tydfil

Caerp

hilly

Bla

enau G

went

Neath

Port

Talb

ot

Torf

aen

Houses in Multiple Occupation (HMOs)form a small but important part of theprivate rented sector. Some of the worstconditions and practices of the privaterented sector are found in HMOs and someof the most vulnerable people in societyare housed in HMOs. These factors, andbecause HMOs are subject to statutorycontrols, make HMOs of profound interest

to policy makers. Therefore, the scale anddistribution of shared dwellings acrossWales have been analysed using the 1991Census of Population.

Table 4.1 shows the distribution of sharedand unshared dwellings between differenthousing tenures for Wales as a whole.


22

Figure 4.3 Private rented share of all rented by local authority, 1997

Tenure All dwellings Shared Shared MeanDwellings dwellings per households

1000 dwellings per shareddwelling

% %

Owner occupied 71.1 24.4 0.8 2.8

Social Rented 21.5 8.4 0.9 4.6

Private Rented 7.3 67.2 20.4 3.9

TOTAL** 100.0 100.0 2.2 3.7

Table 4.1 Shared dwellings* by tenure, 1991

Source: 1991 Census of Population

* The Census definition of a household is one person living alone or a group of people who share either a living room or at least one meal a

day. The Census definition of a dwelling is a self-contained property with a common front door. A shared dwelling is a dwelling occupied

by more than one household.

** Percentages throughout this report may not sum due to rounding

(Wales = 26.1%)

0%10%20%30%40%50%60%

Cered

igion

Denbig

hshir

eCon

wyPo

wys

The

Vale o

f Glam

orga

n

Isle

of A

ngles

eyGwyn

edd

Pem

brok

eshir

eCar

diff

Mon

mou

thsh

ire

Carm

arth

ensh

ire

Rhon

dda C

ynon

Taf

fBr

idgen

dFli

ntsh

ireSw

anse

aNew

port

Wre

xham

Caerp

hilly

Mer

thyr

Tyd

fil

Blaen

au G

went

Neath

Por

t Talb

otTo

rfaen

Priv

ate

rent

ed a

s %

of

all r

ente

d

Over two-thirds of shared dwellings in1991 were in the private rented sector,despite the privately rented dwellings onlyaccounting for 7.3% of all dwellings inWales. Overall, 2.2 per 1,000 dwellings inWales were shared in 1991. However,nearly ten times this proportion of privatelyrented dwellings were shared, at 20.4 per1,000. The mean occupancy (householdsper shared dwelling) of shared dwellings in1991 was 3.9 households in the privaterented sector compared to only 2.8 in theowner occupied sector. The figure forsocial rented accommodation is 4.6. Theaverage social rented figure is likely to behigh due to large hostels for homeless andother vulnerable people run by some localauthorities.

A dataset of shared dwellings at ward levelfor all housing tenures has been derivedfrom the 1991 Census of Population LocalBase Statistics Table 64 by the researchteam. This has been supplied to theNational Assembly for Wales in electronicformat. It contains the followinginformation for all 945 wards in Wales:

• number of dwellings,

• number of unshared dwellings,

• number of shared dwellings,

• number of household spaces,

• shared dwellings per 1,000 dwellings,and

• mean number of households pershared dwelling.

The above variables are providedseparately for:

• all properties,

• owner occupied properties ownedoutright,

• owner occupied properties purchasedwith a mortgage,

• all owner occupied properties,

• local authority properties,

• housing association properties,

• all social rented properties,

• unfurnished privately rentedproperties,

• furnished privately rented properties,

• privately rented properties tied toemployment, and

• all privately rented properties.

4.2.3 Qualitative Analysis

A number of the groups of peopleinterviewed in the course of this researchwere asked about the private rental marketin their local area. These were mainlyletting agents, housing officers and rentofficers. A number of responses shed lighton the reasons for the private rental markethaving a greater share of all housing insome areas compared to others.

Inward investment and general economicbuoyancy were mentioned by a number ofinterviewees, particularly in the context ofCardiff, as factors which encouraged theexpansion of the private rented sector inorder to cater for in-coming workers. Thisboosted the middle and upper ends of theprivate rental market in particular bycatering largely for professional workersand, in some cases, managers in executivepositions who may seek luxuryaccommodation in the private rentedsector. In the case of large inwardinvestors, companies may purchase a


23

property or properties in which to housenewly appointed staff.

Low incomes in an area can restrict thenumber of households who can afford toenter owner occupation and who maychoose to live in privately rentedaccommodation as an alternative. This wascited as a factor in explaining the high rateof private renting found in Powys. Inaddition, limited availability of social rentedaccommodation in Powys can add to therole of the private rented sector in meetinghousing need.

The quality of social rented accommodationavailable can also contribute to the privaterented sector increasingly fulfilling the roleof meeting housing need. A number ofinterviewees in all four case study districtsmentioned that they had noticed anincrease in the number of people already inthe social rented sector moving to a privaterented property. There was uncertaintyregarding the reasons for this, but drugproblems, anti-social behaviour and poorenvironmental quality on some councilestates were suggested as possible factors.In addition, the social rented sector’s abilityto meet housing need was reported asbeing compromised by demographicchanges meaning that properties in thesocial rented sector tend to be too large forthe increasing numbers of smallerhouseholds in housing need.

Historical influences on the housing marketwere also mentioned by interviewees. Forexample, in rural areas, properties formerlytied to agricultural employment or farmtenancies tend to remain in the privaterented sector, even if the occupants nolonger work the land. Seaside towns alongthe North Wales coast where the touristindustry has collapsed in recent decades

retain an inheritance of large former B&Bproperties, many of which have now beenconverted to bed-sits.

In rural parts of North Wales,accommodation used for holiday lets in thesummer often are rented out for a sixmonth period during the winter off-season.This was not considered by the housingstrategy officers interviewed to be asatisfactory arrangement since this bydefinition can only provide short termtemporary accommodation.

The presence of a university enhances thesize of the private rental market in a givenlocation, since a significant proportion ofstudents living away from their parentalhome are accommodated in privatelyrented accommodation. This may helpaccount for the fact that both Swansea andCardiff have a greater share of theirhousing stocks in the private rented sectorthan the rest of South Wales.

The broader housing market also has aninfluence on the size of the private rentedsector in a given location. For example, ifhouse prices are particularly low, then thisoften increased the yields to be obtainedby landlords as rent levels appear to hold-up relatively well for landlords in low houseprice areas. This was the view of a lettingagent and a housing officer in Powys, aletting agent in Swansea, a letting agent inMerthyr Tydfil and a rent officer inAnglesey. In addition, a letting agent andan Environmental Health Officer in MerthyrTydfil made the point that some peoplewho had left the area to find betteremployment had difficulty selling theirhouse in such a depressed local housingmarket, therefore chose to place it on theprivate rental market.


24

In terms of the potential to expand theprivate rented sector as a whole, thelandlord representatives, letting agents andsome of the local authority HousingOfficers expressed concern at excessiveand inconsistent regulation of the sector,particularly with regard to HMOs. Anumber of interviewees expressed the viewthat only a minority of properties in theprivate rented sector are problematic, butthat the whole sector suffers from over-regulation as a result and that this detersprospective landlords from entering thesector and existing landlords fromexpanding their portfolio. Inconsistenciesbetween local authorities in how stringentEnvironmental Health Officers were inapplying fire regulations in HMOs wasraised as a source of confusion andfrustration to landlords who operated inmore than one local authority.

4.2.4 Future Research

As noted above, a number of intervieweesmentioned that they had noticed anincrease in the number of householdsmoving from the social rented sector to theprivate rented sector. This is of moreconcern to the social rented sector than theprivate rented. Nevertheless, this appearsto be an increasing trend which socialhousing providers may wish to address,therefore it needs to be better understood.Indeed, a number of those interviewed,mainly housing officers, suggested that aninvestigation of the reasons for this wouldbe a useful piece of research to be carriedout in the future.

The 2001 Census of Population presents anopportunity to provide an up to datedetailed account of the scale and spatialdistribution of different types of persons,households and dwellings in the private

rented sector, including shared dwellings.In addition, it will be possible to assess thechange since 1991 at a fine grainedgeography.

The 2001 Census of Population alsoprovides the opportunity for localauthorities to compare the number ofHMOs they have on any database orregistration scheme against the actualnumber of properties in their area(although, of course, the identity ofparticular properties cannot be identifiedfrom the Census). The uncertaintysurrounding the proportion of HMOs thatEnvironmental Health departments haveknowledge of was a source of concern toall the Environmental Health Officersinterviewed. Indeed, there is a need formore general research into the nature ofthe problems associated with the HMOsub-sector.

There is also scope for the Rent OfficerService to make similar use of the 2001Census of Population. Again, the Censuscan act as a benchmark against which toassess the proportion of the private rentalmarket for which market evidence hasbeen obtained for the purposes of settingthe Local Reference Rent and the SingleRoom Rent.

4.2.5 Summary of findings

• 7.7% of households in Wales in 1997were in the private rented sector,varying from 2.3% in Torfaen to13.0% in Powys.

• There is generally a greaterprevalence of privately rentedproperties in rural areas, North, Midand West Wales.


25

• The private rented sector tends to bemore prevalent in local authorityareas where a lower proportion ofhouseholds live in the social rentedsector.

• Interviewees thought that the maininfluences on demand for privaterented accommodation in a particulararea are: economic buoyancy andinward investment; the presence ofstudents; the quality and quantity ofsocial rented accommodation; andthe affordability of entry to owneroccupation.

• Interviewees thought that the maininfluences on the supply of privaterented accommodation in a particulararea are: profitability; and thehistorical inheritance of excessproperty in an area, e.g. off seasonholiday lets, seaside B&B propertiesand tied agricultural cottages.

4.3 Dwelling Characteristics

4.3.1 Data Review

4.3.1.1 1991 Census of Population

The 1991 Census provides information byhousing tenure on the number of rooms indwellings, dwelling type (detached house,terraced house, flat, etc.) and amenities(e.g. central heating and shared WCfacilities). Housing tenure in the Censusdifferentiates between furnished andunfurnished private rented properties andaccommodation that is rented from thetenant’s employer. These data refer tohouseholds rather than dwellings.

4.3.1.2 1998 Welsh House Condition Survey

The 1998 Welsh House Condition Surveyphysical inspection is well documented bythe National Assembly for Wales and localauthorities across Wales. Therefore, it wasnot reviewed in detail in this study. Rather,the research team concentrated on the1997 household interview component..

The 1997 household survey includesinformation on: the date of construction,number of rooms, type of dwelling, level ofinsulation, and the presence of centralheating and basic amenities. Therespondents’ and the interviewers’assessments of the condition of propertiesare also recorded.

The WHCS was designed to provide robustinformation down to local authority levelfor most variables. Therefore, a reliableestimate of the housing tenure structure ofdifferent local authority areas can be made,but further disaggregations within theprivate rented sector will result in smallnumbers of observations in some cells.

4.3.1.3 General Household Survey

The General Household Survey (GHS) isavailable annually from 1971 and providesdetailed information on housing, householdand individual personal characteristics. Inany year, it has information onapproximately 8,000 households across theUK, which generates around 40 privaterented sector observations in Wales.Therefore, analysis could only be carriedout at all-Wales level, and adjacent yearswould need to be combined to generaterobust sample sizes. The geographicalindicators contained in this dataset areStandard Regions.


26

The GHS also contains information on thetype of landlord people rent from. Thiscould be used to provide an insight into thelong term changing type of landlordoperating in the private rented sector, forexample employers offeringaccommodation tied to a job, otherinstitutional landlords and privateindividuals. Long term change in the sizeand type of properties in the private rentedsector could also be assessed.

4.3.1.4 Family Expenditure Survey

The Family Expenditure Survey (FES) isanother UK wide on-going survey whichcollects information on approximately8,000 households each year, the same sizeas the GHS. The FES is available from1961. Since the focus of the FES is onexpenditure, it includes levels of rent andrent rebates (presently Housing Benefit).The FES also provides information on thesize of properties, whether they havecentral heating or not, whether rentedproperties are furnished or not, and whereproperties are located. The geographicalindicators contained in this dataset areStandard Regions.

4.3.1.5 1998 Welsh Health Survey

Information in this survey on housingcharacteristics is limited to tenure and theage of dwelling.

4.3.2 Data Analysis

This section describes the physicalcharacteristics of the properties in whichhouseholds living in the PRS areaccommodated. In order to providecontext to this, the PRS is comparedagainst the Social Rented Sector (SRS) andOwner Occupation (OO). This section

describes these three housing tenures interms of type of accommodation, age ofaccommodation and the size of properties.It then goes on to briefly describe the levelof insulation found in different housingtenures and amenities inside the propertiessuch as central heating and smoke alarms.Finally, the state of repair of properties indifferent tenures is described and then theadequacy of the size of dwellings for thefamilies living in them is outlined.

Table 4.2 shows the dwelling typecomposition of each of the three tenuregroups. Over 20% of private rentinghouseholds live in flats/maisonettescompared to only 2% of owner occupiers.Social renters are also substantially over-represented in flats/maisonettes, to aneven greater extent than private renters,with over a quarter of all social rentersliving in this type of dwelling.

Terraced housing is split in approximatelythe same tenure proportions as nationalhousing tenure breakdown would predict,although private renting is marginally over-represented while owner occupation ismarginally under-represented. Privaterenting is substantially under-representedamongst semi-detached housing with only17% of private renters living in semi-detached dwellings compared to 34% ofowner occupiers and 37% of social renters.The healthy representation of semi-detached housing among social renters isdue to significant quantities of this type ofhousing built by the public sector, mainlyduring the inter-war period. However, bearin mind that of semi-detached housing, lessthan a quarter of it is in the social rentedsector while nearly three-quarters of it is inowner occupation, owing to the overallgreater prevalence of owner occupationcompared to social renting.


27

Type of dwelling Tenure Total

Owner Occ Social Rented Private Rented% % % %

Detachedhouse/bungalow 31 2 20 24

Semi-datchedhouse/bungalow 34 37 17 33

Terracedhouse/bungalow 32 35 38 33

Flat/maisonette 2 26 20 9

Other 1 1 5 1

Total 100 100 100 100

Source: WHCS, 1998

Table 4.2 Type of dwelling by tenure, 1997


28

Curiously, private renting breaks thepattern of being over-represented by small,flatted dwellings when it comes todetached properties, with nearly one-in-five PRS households living in a detachedproperty, not substantially below the Welshaverage of 24%. However, nearly one-in-three owner occupiers lives in a detachedproperty compared to 2% of social renters.The higher than expected proportion ofprivate renters in detached properties islikely to be due to a combination of tied,and formerly tied, agricultural propertieswithin the PRS and second and holidayhomes in remoter locations being rented

out by the owners while they are not inresidence and in the off season.

Table 4.3 shows the date of construction bytenure. The PRS is dominated by olderdwellings, with 60% of private rentersliving in dwellings built prior to 1919compared to only of 6% social renters and35% of owner occupiers. Similarly, only17% of private renters live in propertiesconstructed since 1964 compared to 43%of social renters and 32% of owneroccupiers.

Table 4.4 shows the size of households’accommodation by tenure. Overall, thePRS tends to be over-represented amongstsmaller dwellings, although this is also trueof the SRS. The PRS is particularly over-represented amongst one room dwellings(bed-sits) although these only form 0.3%of all households. The PRS is also over-

represented amongst two and three roomdwellings (generally one bedroom flatswith and without separate lounge/kitchensrespectively). The PRS is under-represented amongst all dwellings largerthan three rooms, although only marginallyso for dwellings of seven or more rooms.


29

Type of construction Tenure Total


Pre 1919 35 6 60 31

1919-1944 15 14 14 15

1945-1964 17 38 10 21

Post 1964 32 43 17 33

Total 100 100 100 100

Source: WHCS, 1998

Table 4.3 Date of construction by tenure, 1997

Number of rooms Tenure Totalavailable to household


One 0 1 2 0

Two 1 17 10 5

Three 13 28 25 17

Four 31 30 26 30

Five 33 22 21 30

Six 13 2 9 11

Seven or more 9 1 7 7

Total 100 100 100 100Source: WHCS, 1998

Table 4.4 Number of rooms by tenure, 1997

The properties in which private rentinghouseholds live generally have lessamenities, insulation and central heatingcompared to other tenures, although inabsolute terms the private rented sector isnot too badly equipped. For example,96% of households in the PRS have use ofall basic amenities (gas/electricity andplumbed WC, bath/shower, wash-handbasin and sink) compared to 99% in boththe SRS and owner occupation. However,owing to the small overall size of the PRS,of the households without use of all basicamenities, nearly two-thirds are actually inowner occupation.

Of the households interviewed in the 1998WHCS, 2% of the private renters said theirdwelling had no insulation compared toless than 1% of owner occupiers and anegligible proportion of social renters. Themore significant insulation problems in thePRS are with roofs/lofts (17% with noroof/loft insulation compared to 10% ofowner occupiers); double glazing (over halfwith no windows double glazed comparedto less than one-in-four of owneroccupiers); and hot water tanks (13% withno insulation compared to 7% of owneroccupiers).

Three-quarters of households in the PRShave central heating compared to aroundnine out of ten owner occupiers and socialrenters. Of the private renters who do

have central heating, it is more likely to bepowered by electricity than other tenureswhich is less cost effective to run.

Just under four-in-five of all households inWales have smoke alarms. However, onlytwo-thirds of households in the PRS havesmoke alarms compared to over three-quarters of owner occupiers and 92% ofthose in the SRS. Of those householdswithout smoke alarms, 13% of in the PRS,79% are in owner occupation and 9% arein the SRS.

Table 4.5 shows the overall condition ofdwellings as reported by the householdinterviewer. The overall condition ofdwellings is rated ‘good’ if no parts of thedwelling appear in disrepair; ‘bad’ if one ofroof, walls or doors/windows appear indisrepair; ‘very bad’ if two or more of roof,walls or doors/windows appear in disrepairor any one is in major disrepair.

Eighty-five per cent of owner occupierhouseholds’ accommodation was classifiedas in ‘good’ condition compared to onlytwo-thirds of privately renting households’and just over three-quarters of that in thesocial rented sector. Similarly, 16% of thePRS households’ dwellings were classifiedas in ‘very bad’ condition compared to 6%of owner occupiers’ and 9% of socialrenters’.


30

The household interviewers’ assessment ofthe state of repair of properties isconfirmed by residents’ own assessment.Fifty-three per cent of private rentersthought their accommodation was in goodrepair compared to 77% of owneroccupiers. Nine per cent of private rentersthought their accommodation was in needof major repair compared to 3% of owneroccupiers.

These figures are confirmed by the resultsof the physical inspections of properties bysurveyors in the 1998 Welsh HouseCondition Survey. These showed that 18%of properties in the private rented sectorwere classified as unfit for humanhabitation compared to only 8% of owneroccupied properties and 8% of socialrented accommodation. Similarly, theestimated average cost of repairs perdwelling to meet minimum fitnessstandards were £1,883 in the privaterented sector compared to only £951 in theowner occupied sector and £654 in thesocial rented sector.

Finally, in terms of the adequacy of the sizeof dwellings for the households living inthem, there was no significant problem in

relation to the ‘bedroom standard’ withany particular housing tenure. The‘bedroom standard’ assumes one bedroomis required per single adult (21 years orover), adult partners (whether 21 year ornot), two children sharing (but one roomeach for children over ten years if ofdifferent genders). Only 6% of bothprivate and social renting households werebelow their bedroom standard and 2% ofowner occupiers were below their bedroomstandard. However, 81% of owneroccupiers were above their bedroomstandard compared to 60% of privaterenters and 50% of social renters.

4.3.3 Qualitative Analysis

A number of interviewees were askedabout the nature and condition ofproperties prevalent in the private rentedsector. Landlord representatives and lettingagents were asked about landlords’attitudes towards repair and maintenance.

The landlord representatives and the lettingagents had a strongly held view that themajority of landlords who are in the sectorfor the purposes of investment have agood record on repairs and maintenance.


31

Overall condition of Tenure Totalproperty


Good 85 76 66 82

Bad 7 11 13 9

Very Bad 6 9 16 7

Not Visible 2 5 5 3

Total 100 100 100 100

Source: WHCS, 1998

Table 4.5 Overall condition of property by tenure, 1997

They pointed out that many investmentlandlords want to sell the property whenthey retire therefore want to maintain it ingood condition. Other landlords may wishto live in the property in question at somepoint in the future, therefore these‘accidental’ landlords also wish to maintaintheir property to a high standard.

The restrictions on the level of HousingBenefit payable introduced in 1996 is afactor cited by landlord representatives andletting agents as limiting some landlords’ability to carry out repairs andmaintenance. This was particularly true inareas where a large proportion of privatetenants are in receipt of Housing Benefit.

Letting agents are often given a highdegree of responsibility for carrying outrepairs and maintenance themselves.Letting agents said that it is not in theirinterests to be managing properties whichare in disrepair, as they are difficult to letand cause complaints from tenants. Someletting agents said they would decline tomanage a property if it was in a poor stateof repair.

Environmental Health Officers reportedthat there is a high incidence of dampness,disrepair and unfitness in the private rentedsector. This is especially true of HMOs,consistent with findings of the 1998 WelshHouse Condition Survey physicalinspections.

4.3.4 Future Research

Much is known in general from the 1991Census of Population and the 1998 WHCS,therefore the most fruitful areas for futureresearch are those which focus on specificdisrepair issues and how they may beremedied. For example, the HMO sub-sector and the impact of Housing Benefitrestrictions on repairs and maintenance indifferent private rental market contextscould be usefully investigated further.

4.3.5 Summary of findings

• Flats/maisonettes; older propertiesesp. pre 1919; smaller dwellings esp.one room; below average conditionproperties; and properties without allbasic amenities are over-representedin the private rented sector comparedto other tenures.

• Most investment landlords and lettingagents have high maintenancestandards.

• DSS sector landlords have less goodmaintenance records, especially thosenot using a letting agent in areas oflow market demand.

• Landlord representatives and lettingagents say that Housing Benefitrestrictions have limited landlords’ability to carry out repairs andmaintenance.

• Environmental Health Officers reportmuch dampness, disrepair andunfitness in the private rented sector,especially HMOs.


32

5.1 Data Review

5.1.1 1991 Census of Population

Information on household size is availableby tenure in the 1991 Census ofPopulation. In addition, counts of thenumbers of households with dependentchildren, households with pensioners andhouseholds with any dependants areavailable disaggregated by housing tenure.Information on overcrowding and carownership is also provided by housingtenure.

5.1.2 1998 Welsh House Condition Survey

The 1998 WHCS contains information onthe characteristics of households andindividuals. Broadly, data refer tohousehold size, household income,individuals’ age, individuals’ socio-economic status, and individuals’ receipt ofvarious state benefits. Whether or notHousing Benefit is paid to the household isalso recorded, as is the amount of rentpaid.

5.1.3 General Household Survey

The General Household Survey (GHS)provides detailed information on householdand individual personal characteristics at allWales level.

5.1.4 1998 Welsh Health Survey

The only non-health related informationthis survey contains that is not containedwithin the WHCS is ethnic group.

Obviously, detailed information iscontained within this dataset regardingindividuals’ ailments, state of health andfrequency of visits to GPs and hospital.This type of health data could usefully beanalysed against housing conditions, as ithas long been recognised that poorhousing can result in poor health, yetrecent empirical evidence to support this issometimes lacking.

5.2 Data Analysis

This section describes the characteristics ofthe households and individuals who live inthe PRS, the SRS and OO. Thecharacteristics of the residents of thesethree housing tenures are described interms of household size, household type,household income, marital status, age,socio-economic status, ethnic group andstate of health.

Table 5.1 shows household size by tenure.Single person households are substantiallyover-represented in both the PRS and theSRS, with 32% of households in the PRSconsisting of one person and 36% of thosein the SRS compared to 21% of those inowner occupation. There are notparticularly marked deviations from theWelsh aggregate tenure distribution forhouseholds of more than one person,although larger households of six or morepeople are more likely to be housed in theSRS and, to a slightly lesser extent, thePRS.


33

Chapter FivePrivate Rented Sector Tenants

Table 5.2 shows the income distribution ofhouseholds in each of the three housingtenures. Private renting households are onlower than average income but not as lowas social renters. Households in the SRSare the most skewed towards lowerhousehold incomes, with just over half ongross household incomes of less than£7,000 per annum compared to 15% ofowner occupiers and 36% of privaterenters. Private renters are also stronglyskewed towards lower incomes, althoughless so than social renters. At the upperend of the income scale, 8% of privaterenting households have gross annualincomes over £20,000 compared to 18%of owner occupiers and only 1% of socialrenters.

However, gross household incomeunderestimates the actual wealthexperienced by individuals in the PRSbecause private renting households are onaverage much smaller than other tenures,so a given household income goes further.The PRS, as outlined above, has a particularconcentration of single person households.An ‘equivalent income’ adjustment wouldneed to be made to accurately reflect how‘well-off’ different household types andsizes actually are. Nevertheless, thedifferences in household income betweenhousing tenure groups shown in table 5.2are sufficiently marked so that the generalpattern is unlikely to be altered if using anequivalent income scale.


34

Number of people Tenure Totalin household


One 21 36 32 25

Two 37 27 31 34

Three 18 16 16 17

Four 17 12 12 16

Five 6 6 6 6

Six or more 2 3 3 2

Total 100 100 100 100

Source: WHCS, 1998

Table 5.1 Household size by tenure, 1997

35

Another indicator of wealth is householdsavings. In particular, savings makepurchasing a property with a mortgagemore manageable. However, 37% ofthose in the PRS have no savings and 54%have less than £3,000 (a realistic minimumamount required to put down a 5%deposit and pay legal, survey and otherset-up costs when buying a home). Thiscompares to only 13% of owner occupiershaving no savings and 34% with less than£3,000. Fifty per cent of social rentershave no savings and 68% have less than£3,000.

Table 5.3 shows the distribution of differenthousehold types found within each housingtenure. The importance of the PRS inhousing non-pensioner single persons isstriking, with 21% of households in thePRS falling into this category compared toonly 7% of owner occupiers and 11% ofsocial renters. Single pensioners, however,are under-represented in the PRS, with11% of private renting households beingsingle pensioners compared to 25% ofsocial renters and 13% of owner occupiers.Lone parents are also substantially over-represented in the PRS, but also in the SRS


Gross household Tenure Totalannual income


Up to £4,000 5 22 17 9

£4,001-£7,000 11 29 19 15

£7,001-£10,000 9 13 10 10

£10,001-£15,000 10 6 8 9

£15,001-£20,000 9 3 5 8

£20,001-£30,000 10 1 5 8

Above £30,000 8 0 3 6

Refused 24 14 17 21

Don’t Know 15 13 16 15

Total 100 100 100 100

Average income (£) 16,196 6,351 9,762 13,281(based on mid-points;upper band £40,000)

Source: WHCS, 1998

Table 5.2 Gross household annual income by tenure, 1997

Welsh Assembly Government - A Housing Research Audit for Wales

36

This section so far has described thecharacteristics of households. It nowconsiders the characteristics of theindividuals, again disaggregated by housingtenure. Table 5.4 shows individuals’ maritalstatus by housing tenure. The ‘single nevermarried’ category is substantially over-represented in the PRS, illustrating the roleof the tenure as a first stop after youngadults leave their parental home. Fifty-three per cent of individuals in owneroccupation are married compared to only27% of those in privately rentedaccommodation and 30% of those in theSRS. Just under 3% of private renters aremarried but separated compared to justover 1% overall and under 1% of owneroccupiers. This illustrates the role of thePRS in providing accommodation forpeople immediately after they separate

from their marriage partner. Divorcees arealso over-represented in the PRS but lessso, reflecting a gradual shift back intoowner occupation (but not into socialrenting) after couples’ divorces comethrough.

Table 5.5 shows the age profile of theindividuals living in each of the three mainhousing tenures. The PRS contains itsproportionate share of dependent childrenwhile the SRS contains more than its shareand owner occupation slightly less. Youngadults (age 18-24) are heavily over-represented in the PRS, reflecting thetenure’s important role for young peopleafter leaving home and prior to enteringowner occupation and, to a lesser extent,prior to entering the SRS.

Household type Tenure Total

Owner Occ Social Rented Private Rented(h/h = household) % % % %

Single pensioner 13 25 11 16

Other single pensioner 8 11 21 9

Lone parent h/h 3 16 13 6

Other h/hs withdependent children 28 22 18 26

Families without dependent children 49 27 37 43

Total 100 100 100 100

Source: WHCS, 1998

Table 5.3 Household type by tenure, 1997

37


Household type Tenure Total


Single, never married 36 50 59 41

Married, living with spouse 53 30 27 46

Married, separated 1 2 3 1

Divorced 3 8 6 5

Widowed 7 11 5 7

DNA 0 0 1 0

Total 100 100 100 100

Source: WHCS, 1998

Table 5.4 Martial Status by housing tenure, 1997

Age Group Tenure Total


0-15 (children) 19 30 22 22

16-17 (adolescents) 3 3 2 3

18-24 (young adults) 7 7 22 8

25-59/64 (adults ofworking age) 51 39 44 48

60/65+ (pensioners) 20 21 10 20

Total 100 100 100 100

Source: WHCS, 1998

Table 5.5 Age of residents by tenure, 1997

The private rented sector also contains adisproportionate share of householdsheaded by an individual under the age of30 years. In 1997, 34% of households inthe private rented sector were headed bysomeone under the age of 30, comparedto only 14% in the social rented sector and6% in owner occupied accommodation.

Unemployment is high among people inthe private rented sector, although not ashigh as among those in the social rentedsector. Of economically active privaterented sector tenants of working ageinterviewed in the 1998 WHCS, 12% wereunemployed compared to 19% of socialrented sector tenants and only 3% ofowner occupiers. The Welshunemployment rate according to the 1998WHCS was 5%. These figures relate to the‘ILO’ definition of unemployment, whichrequires an individual to have looked forwork in previous four weeks and beavailable to start within a fortnight.

Members of ethnic minority groups aresubstantially over-represented in the PRS.Just over 3% of individuals living in thePRS are members of ethnic minority groupscompared to just over 1% of both owneroccupiers and social renters.

The general state of individuals’ health ismarginally, but not dramatically, better inthe PRS than in Wales as a whole. 18% ofpeople living in the PRS describe theirhealth as ‘excellent’ compared to 15% ofthose in owner occupation. Only 8% ofsocial renters, however, describe theirhealth as ‘excellent’. Conversely, 5% ofboth private renters and owner occupiersdescribed their state of health as ‘poor’compared to 13% of social renters. Thesefigures are not surprising since privaterenters tend to be younger than people in

other housing tenures. Indeed, given thatthe tenure differences are not substantial, ifage were controlled for, it may be the casethat private renters actually have poorerhealth than their counterparts in othertenures.

5.3 Qualitative Analysis

Market segments commonly identified bymost of the groups of interviewees werebroadly:

• professionals and other mobileworkers,

• young single people and couples,

• students, and

• ‘DSS lets’ (i.e. recipients of HousingBenefit).

These four broad groups formed distinctmarket segments in the minds of lettingsagents and investment landlords. Withineach of these three broad groupings,however, distinctions were made betweensub-groups. The ‘professionals and mobileworkers’ group included itinerant workers,for example in construction or tourism,contract staff at universities, young juniorprofessionals and company executives whomay move into family accommodation atthe luxury end of the private rented sector,mainly in Cardiff.

The ‘young single people and couples’category is thought to include people whocannot yet afford to enter owneroccupation and couples in the early stagesof their relationship who are not yet readyto commit to a joint mortgage. There isoverlap between this and the formergroup, as many young people seekingentry to owner occupation may also beunsure of how long they may stay in their


38

39

current job. In Powys and Anglesey whereemployment is more seasonal than in SouthWales and the private rented sector moreprevalent, this group of young rentersextends to older couples with children.Indeed, in these areas, letting agentsreported a small proportion of familiesmake private renting their long term tenureof choice.

‘Students’ form a fairly homogenous group,although letting agents differentiatedbetween undergraduates and post-graduates. Undergraduates oftenapproached landlords or letting agents in agroup, while post-graduate students weremore likely to live alone. Post-graduatestudents were also more likely to sharewith non-students and as such had aspectsin common with the mobile workercategory.

Two distinct groups within the ‘DSS lets’category were long-term Housing Benefitrecipients and a more transient group. Theformer were most commonly lone parentsor people also in receipt of IncapacityBenefit. The latter group were more oftenin receipt of Job Seekers Allowance.

This latter group of shorter term, moretransient, tenants in receipt of HousingBenefit were reported by landlordrepresentatives and letting agents to be thesource of some problematic behaviour fromlandlords’ point of view. Rent arrears andvacating properties without serving noticeto quit, and in some cases not returningkeys, were the most commonly reportedproblems. These problems appeared to bethe most severe in the seaside towns ofNorth Wales where ‘DSS lets’ form a largeproportion of the private rental market dueto the conversion of former B&B propertiesto bed-sit type accommodation and a

particularly high turnover of tenants. InMerthyr Tydfil, there was some informalpolicing of ‘problem’ tenants through wordof mouth between landlords.

Vacating properties in a poor state ofcleanliness was reported as a commonproblem and that this was more commonamong a ‘certain type’ of Housing Benefitrecipient. Damage to property,neighbourhood nuisance and drug takingwere mentioned by some letting agentsand housing officers as problems in thebottom end of the private rented sector,but these were not generally considered tobe particularly significant in extent orseverity in this tenure.

5.4 Future Research

As with dwelling characteristics, much isalready known from the Census and the1998 WHCS with regards to thecharacteristics of private tenants. What isless well understood is the preciseexperiences and aspirations of privatetenants, as little qualitative work has beendone with them. Questions to ask mightinclude: How long do people intend to livein a particular property? How long dopeople intend to live in the private rentedsector? What problems do they encounterin finding suitable accommodation? Dopeople chose different housing tenuresbecause of the inherent nature of thetenure or because of the quality, type,location and size of properties available indifferent housing tenures?

The apparent problems with rent arrearsand the vacation of properties by tenantswithout serving notice to quit couldusefully be examined from tenants’ pointof view. For example, their ‘side of thestory’ with regards to repairs, rent arrears,


the Housing Benefit system andrelationship with landlords could be sought.As with much previous work into theprivate rented sector in the UK, this reviewof the sector in Wales has not gatheredinformation on tenants’ perspectives, buthas rather relied on landlords’ and lettingagents’ opinions.

5.5 Summary of Findings

• private tenants are substantiallyyounger and lower income than thepopulation as a whole,

• single people, both pre- and post-marriage, are more likely to be privatetenants (apart from pensioners),

• private tenants are more likely to beunemployed and members of ethnicminority groups compared to thepopulation as a whole,

• landlords and letting agents generallyrecognise the following marketsegments:

- professionals and other mobileworkers,

- young single people andcouples, including families inrural areas,

- students, and

- ‘DSS lets’ (i.e. recipients ofHousing Benefit), differentiatingbetween long-term and moretransient sub-groups.

• landlords representatives and lettingagents reported problems at thebottom end of the ‘DSS’ market ofrent arrears and tenants vacatingproperties without serving notice toquit.


40

41

6.1 Introduction

Housing Benefit plays an important rolewithin the housing market by helping low-income tenants to afford their rent. It isespecially important in the private rentedsector, where rents are generally muchhigher than in social housing.

This chapter considers the role of HousingBenefit (HB) within the private rentedsector in Wales. It draws on administrativedata supplied by the Department for Workand Pensions, on the socio-economic dataon private tenants collected as part of thehousehold interviews in the Welsh HouseCondition Survey, and on qualitativeinformation collected from interviews withprivate landlords, managing agents, andRent Officers.

6.2 Data Review

6.2.1 Administrative Data

The Department for Work and Pensions’(DWP) Housing Benefit ManagementInformation System is based on returns bylocal authorities on 100% of their HBcaseload. It contains annual data on thenumber of regulated and deregulatedprivate tenants in receipt of HB in Walesand in each local authority from 1992 to2000. It also contains, for Wales and foreach local authority, the average eligiblerent and average HB entitlement ofregulated and deregulated private tenantsfrom 1996 to 2000.


The household interview survey carried outas part of the 1998 Welsh House ConditionSurvey includes questions on gross annualrent and on whether or not anyone in thehousehold receives Housing Benefit. Thismakes it possible to examine thecharacteristics of households in receipt ofHB and compare them with those of othertenants in the private rented sector.

6.3 Data Analysis

The first part of this section summarises theDWP administrative data on HB claims byprivate rented sector (not housingassociation) tenants in Wales. The secondpart of this section then draws on thequantitative data from the householdsurvey in the 1998 WHCS to compare thecircumstances of households on HousingBenefit with those of non-recipients in theprivate rented sector in Wales.

6.3.1 Housing Benefit Claims

Because Housing Benefit is a demand-ledsocial security benefit, the caseload tendsto fluctuate with the economic cycle, otherthings being equal. In fact, the HBcaseload for private tenants in Walesincreased in the early to mid-1990s andthen fell, broadly in line with the rise andfall of unemployment over this period(Table 6.1).


Chapter SixThe Role of Housing Benefit in Supporting the PrivateRented Sector

Year Regulated tenancies Deregulated tenancies Total

1992 16,000 35,000 51,000

1993 14,000 41,000 55,000

1994 12,000 46,000 58,000

1995 10,000 48,000 58,000

1996 9,000 49,000 58,000

1997 7,000 47,000 54,000

1998 6,000 46,000 52,000

1999 5,000 44,000 49,000

2000 5,000 42,000 47,000

However, the change in the caseloadcannot be wholly reduced to a function ofthe change in unemployment because thisperiod also witnessed a long-run decline inthe number of regulated tenants. Most ofthe increase in caseload in the early to mid-1990s was due to a rise in the number ofderegulated tenants on income-basedJobseeker’s Allowance (JSA) and IncomeSupport (IS) claiming HB. But the fall inthis component of the caseload over thesecond half of the decade was matched bya fall in the number of regulated tenantson HB. In the deregulated sub-sector, thefluctuation in the Income Support/JSAcaseload was much greater over thedecade than among other HB claimants.

Table 6.2 shows average weekly eligiblerent and average weekly HB entitlementamong private tenants in Wales from 1996to 2000. The eligible rent may not be thesame as the contractual rent that thetenant has agreed to pay the landlord, forexample if it is net of ineligible servicecharges that are included in the contractual

rent. In addition, the eligible rent may belower than the contractual rent if it isaffected by one or more of the variousrestrictions within the HB scheme. Forexample, the rent may be restricted for thepurpose of calculating HB because it isover-expensive compared with the rentsomeone not on benefit would pay orbecause the accommodation is deemed tobe over-large for the tenant’s needs. Theeligible rent will also be capped if itexceeds the local reference rent or (in thecase of single claimants under 25) thesingle room rent.

As would be expected, the average eligiblerent of deregulated tenants on HousingBenefit is roughly £17 to £18 per weekhigher than that of regulated tenants (Table6.2). Not surprisingly, the average amountof HB is also higher.

The average amount of HB is higher fortenants on Income Support or income-based Jobseeker’s Allowance (JSA (IB))than it is for other claimants. This is


42

Source: Department for Work and Pensions

Table 6.1 The Housing Benefit caseload: private renters in Wales, 1992-2000

43

because the former generally get all oftheir rent covered by Housing Benefit,whereas other claimants may only havepart of their rent covered by HB because ofthe ‘taper’ whereby benefit falls as incomerises.

Hence, the average amount of HB is notfar short of the average eligible rent in the

case of tenants who are also on IS or JSA(IB). For regulated tenants who are not onIS or JSA (IB), there is a difference ofapproximately £8 to £9 per week betweenthe average eligible rent and the averageHB entitlement. The equivalent differencefor deregulated tenants is approximately£12 per week (Table 6.2).


Year Regulated tenancies Deregulated tenancies

IS/JSA Non IS/JSA IS/JSA Non IS/JSA

Rent HB Rent HB Rent HB Rent HB

1996 £34.40 £34.00 £35.10 £27.30 £52.30 £52.00 £53.20 £41.40

1997 £36.40 £36.00 £36.90 £28.80 £53.30 £53.00 £53.70 £42.10

1998 £37.90 £37.50 £39.00 £30.10 £54.50 £54.20 £55.40 £42.90

1999 £39.10 £38.70 £40.50 £32.20 £56.10 £55.80 £56.30 £43.80

2000 £40.50 £39.70 £41.50 £34.50 £53.80 £57.00 £57.90 £43.50

Note: ‘Rent’ is average weekly eligible rent; ‘HB’ is average weekly Housing Benefit. ‘IS/JSA’ is Income Support and Jobseeker’s Allowance

(income-based) cases; ‘Non-IS/JSA’ is non-Income Support and non-Jobseeker’s Allowance (income-based) cases.

Source: Department for Work and Pensions

Table 6.2 Average weekly eligible rent and Housing Benefit among private renters in Wales,1996-2000

6.3.2 Private Tenants and Housing Benefit

This section draws on the quantitative datafrom the household survey in the 1998WHCS to compare the circumstances ofhouseholds on Housing Benefit with thoseof non-recipients in the private rentedsector in Wales.

There are significant differences betweenthe family characteristics of HB recipientsand other tenants in the private rentedsector in Wales. The main difference is thata much higher proportion of HB recipientsare lone parents. Thus, 27% of HBrecipients are lone parents compared withonly 3% of non-recipients. In fact, 87% oflone parents living in the private rentedsector in Wales are on HB. By comparison,only 33% of couples with children, 24% ofcouples without children and 35% of singlepeople are on HB.

As Housing Benefit is a means-testedbenefit, it is not surprising that recipientshave lower incomes than other privatetenants. Table 6.3 shows the grosshousehold income of private tenants inWales in 1997. The data exclude the largeproportion of respondents who did notknow what their household income wasand those who refused to say. Amongthose who were willing and able to providethe information, eight out of ten recipients,compared with only a quarter of non-recipients, had a total gross householdannual income of less than £7,000.Approximately three-quarters ofhouseholds with an income below thisamount were receiving HB. HousingBenefit receipt tails off quite quickly asannual income rise above £7,000.


44

Annual household income HB recipients Non-recipents

% %

Up to £4,000 41 11

£4,001-£7,000 43 15

£7,001-£10,000 12 17

£10,001-£15,000 3 20

£15,001-£20,000 1 14

£20,001-£30,000 1 14

Over £30,000 - 10

Total 100 100

Source: WHCS, 1998

Table 6.3 Annual household income of private tenants by receipt of HB

45

For much the same reasons, HB recipientsin the private rented sector also have few,if any, savings. Among those whoresponded to this question in the 1998WHCS, eight out of ten HB recipients hadno savings at all, a figure that was doublethat for other private tenants (Table 6.4).

About one in ten recipients, compared withone in six non-recipients, had savings of upto £500. Hardly any (1.5%) recipientshad savings of over £3,000 – the level atwhich savings start to reduce HBentitlement - compared with one in fivenon-recipients.


Gross annual household HB recipients Non-recipentsincome

% %

None 82 43

Up to £500 11 17

£501 to £1,000 3 9

£1,001 to £3,000 3 10

£3,001 to £10,000 1 12

£10,000 to £16,000 - 4

Over £16,000 - 6

Total 100 100

Source: WHCS, 1998

Table 6.4 Household savings of prviate tenants by receipt of HB

The lack of savings among HB recipients isnot surprising given their low income andthe fact that savings above £3,000 reducethe amount of benefit to which people areentitled. But it can have significantimplications for those who wish or have tomove – and turnover within the PRS is veryhigh - because private landlords commonlyrequire prospective tenants to pay rent inadvance and a deposit.

Table 6.5 shows gross annual rent (beforeHB) in the private rented sector in 1997.Two out of five private tenants were beingcharged a rent that was between £2,500and

£4,999, while a further fifth were chargedbetween £1,000 and £2,499. One in 20had a rent of £7,500 or more per year (i.e.,£625 or more per month).

Private sector rents are generally higherthan social housing rents. The spread ofrents is also much greater in the privatesector. This reflects the fact that theprivate sector caters for a much widerrange of income groups than the socialhousing sector, with a correspondinglygreater range in housing size and quality.In addition, around one in ten privatetenants, but hardly any social housingtenants, live in rent-free accommodation.Social housing rents are subsidised, whichkeeps them below market levels. Finally,private rents in the deregulated sectoradjust to the demand and supplyconditions found in different sub-markets,unlike social housing rents.

The extent of rent-free accommodation inthe private rented sector reflects a number

of factors. Firstly, 6% of households in1997 were renting accommodation fromtheir employer and half of them were notcharged rent. In the agricultural sector inparticular, the provision of rent-free ‘tiedhousing’ is often associated with lowwages. Second, 12% of private tenantswere renting their accommodation from arelative or friend and two-fifths of themwere not being charged rent.

HB recipients generally have lower rentsthan other tenants in the private sector. In1997, 95% of HB recipients were beingcharged a rent that was below £5,000 perannum, compared with 78% of tenants noton HB (Table 6.6). Seven out of ten HBrecipients had rents between £2,500 and£4,999 per annum, equivalent to between£52 and £96 per week.


46

Gross Annual Rent %

Rent Free 9

Up to £499 1

£500-£999 2

£1,000-£2,499 22

£2,500-£4,999 40

£5,000-£7,499 7

£7,500 or more 5

Don’t know 9

Refused 4

Not asked 2

Total 100

Source: WHCS, 1998

Table 6.5 Private sector rents

47

A very substantial minority of privatetenants in Wales receives financialassistance from the Housing Benefitscheme to help them afford their rent. In1997, 45% of private tenants who werebeing charged rent were in receipt of HB.This is a higher proportion than in eitherEngland or Scotland, where currentlyaround a quarter and a third of privatetenants respectively are on HousingBenefit. However, it is much lower thanthe social housing sector in Wales, where72% were receiving HB in 1997. Becauseit is much smaller than the social housingsector and a smaller proportion of tenantsis in receipt of HB, the private rented sectoraccounts for only about one in six HBrecipients in Wales.

Nevertheless, it is clear that a significantproportion of private landlords is willing (orhas little choice) to let theiraccommodation to HB claimants. Ofcourse, some of the private tenants inreceipt of HB may not have requiredassistance to pay their rent when they werefirst offered a tenancy, but may have

subsequently experienced a fall in income.If a landlord is otherwise satisfied with thetenant, then they may be willing to acceptrent payment wholly or partly through HB.

Turning now to the homes in which privatetenants live, there is no apparentassociation between the date ofconstruction of the dwelling and receipt ofHB. For example, recipients are just aslikely as non-recipients to be living in theoldest (i.e. pre-1919) dwellings and in themost modern (i.e. post-1964) ones. Thereis, however, a difference between thesetwo groups of tenants in terms of the typeof dwelling they rent from their landlord.HB recipients are more likely to be living inflats, while non-recipients are more likelyto be living in detached homes orbungalows. They are equally likely to beliving in semi-detached or terraced houses.In general, HB recipients tend to live insmaller dwellings (measured in terms of thenumber of rooms) than non-recipients.

There are some significant differences insome aspects of the quality of dwellings in


Gross annual rent HB recipients Non-recipents

% %

Up to £499 1 2

£500-£999 2 2

£1,000-£2,499 22 33

£2,500-£4,999 70 42

£5,000-£7,499 5 11

£7,500 or more - 11

Total 100 100

Source: WHCS, 1998

Table 6.6 Private sector rents by receipt of HB

which HB recipients live compared withthose in which non-recipients live. Forinstance, HB recipients are less likely thannon-recipients to be rentingaccommodation that has the benefit ofcentral heating (72% compared with79%). This may have an impact ontenants’ ability to keep their home warm orthe amount of money have to pay in fuelcosts, which may have implications for fuelpoverty given the lower income ofrecipients compared with non-recipients.

Housing Benefit recipients in Wales are alsoless likely than non-recipients to be living inproperties that are in good physicalcondition. The WHCS shows that 66% ofnon-recipients were living in properties thatwere assessed by the survey interviewer asbeing in ‘good’ condition, compared with61% of HB recipients (Table 6.7).Conversely, HB recipients were more likelyto be renting properties assessed as beingin ‘bad’ condition. However, they wereslightly less likely to be living in ‘very bad’properties, which suggests that therelationship between property conditionand benefit receipt is not a straightforwardone.


48

Condition of property HB recipients Non-recipents

% %

Good 61 67

Bad 17 12

Very bad 17 19

Not visible 5 3

Total 100 100

* as assessed by the survey interviewer

Source: WHCS, 1998

Table 6.7 Visible property condition* by receipt of HB

There are also some differences betweenhow HB recipients perceive the repaircondition on their property as comparedwith other private tenants. In fact, thosewho are on HB are slightly more likely to

say their property is in good repair than arenon-recipients. But they are also slightlymore likely to say that their property needsmajor repair work (Table 6.8).

49

Private tenants on HB are less likely to berenting their home from their employerthan are non-recipients. This is notsurprising given that many tied tenants donot pay rent and most are in paid work (aminority is retired). The other maindifference is that HB recipients are morelikely to be renting from an individual thanfrom an organisation.

Housing Benefit recipients are just as likelyas other private tenants to prefer living inthe PRS. In the WHCS 1997, 27% ofrecipients and 28% of non-recipients saidthey would prefer to rent from a privatelandlord if they had the choice. In otherwords, while little more than a quarter ofprivate tenants is in their preferred tenure,HB recipients are as likely to be in theirpreferred tenure as non-recipients. Where

these groups differed most of all was inrelation to social rented housing. Privatetenants on HB were three times as likely asnon-recipients to say they would prefer torent from a social housing landlord. Non-recipients were a little more likely to saythey would prefer to buy their home.These differences are likely to reflect theirassessment of their long-term incomeprospects, among other things.

Not surprisingly given their respectivepreferences for social housing, HBrecipients are much more likely to be on alocal authority or housing associationwaiting list than are other private tenants.In the 1998 WHCS, one in five HBrecipients and one in 20 non-recipients inthe private rented sector had their name ona social housing waiting list (Table 6.9).


On a waiting list? HB recipients Non-recipents

% %

Yes 21 5

No 79 95

Total 100 100

Source: WHCS, 1998

Table 6.9 Whether private renting households have their name on a social housing waitinglist by receipt of HB

Need for repair HB recipients Non-recipents

% %

In good repair 52 49

Needs minor repair 37 42

Needs major repair 11 9

Total 100 100

Source: WHCS, 1998

Table 6.8 Private tenants’ view of the need of repair on their property by receipt of HB


This section reports on the responses toquestions on Housing Benefit included inthe semi-structured, qualitative interviewswith landlords’ representatives, managingagents and Rent Officers.

6.4.1 Landlords and agents

The proportion of tenants who were HBclaimants was thought to vary significantly,depending upon the landlord or the area.Thus, one agent had very few tenants onHB, while another had more than 50% inthis category. In one seaside area, ‘DSS lets’were said to account for most of themarket, especially among HMOs.

Three of the eight landlords and agentssaid there were no advantages at all inletting to people on HB. However, in someareas they did not always have the choiceabout whether to let accommodation tosomeone on HB or not. Somedistinguished between long-term tenantsand transient ones, liking the former butnot the latter. Several said one advantageof letting to HB recipients was that directpayments meant that they were assured ofgetting the rent.

However, the quid pro quo of directpayments was that the local authoritywould claw money back from the landlordif the tenant had been overpaid HB, aproblem that could be exacerbated wherethere were delays in the initial payment ofbenefit. As a result, one letting agent (whohad few claimants on their books) said theywere now reluctant to accept directpayments and insisted instead that HBclaimants paid their rent themselves via astanding order.

Three respondents complained about theslow speed with which HB claims werebeing processed by the local authority.One authority was said to have a 6 monthsbacklog of claims.

Two respondents said that anotherdisadvantage of letting to HB recipientswas that the benefit was paid in arrearsinstead of in advance, while anothermentioned the fact that it is paid 4-weeklybut they charge rent monthly. Finally, HBshortfalls were said to be a disadvantage ofletting property to claimants.

There was a range of responses aboutwhether tenants ever tried to negotiateover the rent when they viewed a property.Some landlords and agents said that thishappened sometimes, one said it did nothappen at their office but it did at theirbranches elsewhere in Wales, and one saidit never happened. However, one agentsaid that 3 out of ten tenants negotiatedthe rent and this reflected the fact that themarket was over-supplied and rents hadfallen over the previous 5 years. Tworespondents said that HB recipients wereless likely to negotiate the rent whenviewing a property. But two others saidthey were more likely to do so than tenantspaying out of their own pocket, especiallyif they had had a pre-tenancydetermination (PTD) that was less than theasking rent. The PTD could thus provide alever for people on HB.

The knowledge of the rent restriction rulesamong the landlord representatives andmanaging agents varied considerably andwas not always accurate. In general,agents and landlords that let a significantproportion of their properties to HBrecipients had considerable knowledge ofthe rules, whereas those who had few such


50


51

tenants tended to have quite poorknowledge.

Three of the eight landlords and agentshad not heard of the local reference rent(LRR); two of these had also not heard ofthe single room rent (SRR), while the otherhad only heard of it because his mother letbedsits. Most of the landlords and agentswho were aware of the LRR and SRR didnot know how either was calculated, butespecially the latter.

Several landlords and agents were verycritical of the level of LRR set by the RentOfficer Service, claiming that it wasunrealistically low, had driven rent levelsdown, or made them less willing to letaccommodation to HB recipients. On theother hand, one landlord felt that the LRRwas set at about the right level.

All of the landlords and agents thoughtthat shortfalls (resulting from the gapbetween the contractual rent and the renton which the local authority had calculatedHB) were the norm rather than theexception. One agent tried to ensure thatthey only placed HB recipients in propertieswhere their experience suggested thatthere would be no shortfall or where thetenant could provide a guarantor for therent. Most of the landlords and agentssaid that they expected the tenant to makeup the shortfall and would not lower therent, though one said they sometimes didre-negotiate the rent. However,expectation was not the same as realityand tenants often fell into arrears. Sometenants moved to somewhere cheaper, butthis was not always possible and, if theydid move, the process simply repeated itselfwith the next landlord.

Responses were mixed about whether HBtenants were more or less concerned aboutthe rent than other tenants. Severalthought that, because of the rentrestriction rules and the consequentpossibility of experiencing a shortfall, HBrecipients were more concerned about therent than other tenants. One landlordclaimed that longer-term tenants on HBwere just as concerned as other tenants,but the transient tenant was not concernedabout the rent (and tended not to make upany shortfall). One agent believed thatsome tenants had a relaxed attitude to therent if they were not paying themselvesand the HB was often late. Another (whohad cut back on lettings to HB tenants)claimed that tenants do not pay shortfalls.

6.4.2 Rent Officers

The proportion of HB referrals inspected byRent Officers varied from just under aquarter to two-fifths. All four Rent Officerssaid they aimed to inspect all propertiesbeing referred to them for the first timeand carried out re-inspections either everyfour or five years.

Finding ‘market evidence’ - that is,information on rents being paid by tenantsout of their own pocket - was a constantaspect of the Rent Officer’s job. All saidthat there were areas or types of propertyfor which it was difficult to find any marketevidence. The scattered nature of smallrural villages was a particular problem andit often meant that a property was beingcompared with the rent on a property inanother village that might be 30 milesaway. One Rent Officer said the localitieswhich they used for setting LRRs often hadto be extended in order to find enoughmarket evidence. However, this meant thatthe LRR was almost always likely to ‘bite’,

52


and thus restrict choice to HB recipients, inthe more expensive pockets of propertywithin these larger localities. One RentOfficer noted that because few propertiescould be inspected internally, they had toassume that they were in averagecondition.

Rent Officers believed that many HBclaimants did not understand the rentrestriction rules. This was thought to beespecially true of first-time claimants andyoung people affected by the SRR. Repeator long-term claimants had a much bettergrasp of the system. It was said that somelandlords had a very good workingknowledge of HB but that others hadrelatively little idea. Rent Officers felt thesize criteria were less well understood thanthe LRR or SRR. Some landlords werethought to target HB claimants, but otherssteered clear of them if they could. Theformer knew the rules and set their rentsaccording to the LRR or PTD valuation.

Three of the four Rent Officers said that itwas very difficult to find market evidencefor properties that fitted in with the pre-July 2001 definition of the SRR. In onelocal authority there was said to be almostno accommodation that fell within therather restrictive definition. However, allthree noted that the revised definition wasmuch more realistic and that it should beeasier to find market evidence. So far asbenefits of the SRR are concerned, oneRent Officer could not think of any.Another said it had helped to get rid of theworst HMOs in the area, though in someother areas it may have helped to maintaindemand for such accommodation as itforced people into bedsits.

It was claimed that some landlords initiallyasked claimants for a high rent, but then

agreed whatever rent the local authorityaccepted for HB purposes. It was alsoclaimed that other landlords would acceptthe LRR and not chase up shortfalls if theywere happy with the tenant (but would ifthey were not). Some landlords tried toavoid HB claimants because of the LRR, butthe success of this strategy depended onthe characteristics of the local housingmarket. In other words, if supply exceedsdemand in a particular area, then landlordscannot avoid letting to recipients of HB.

Pre-tenancy determinations whereby theRent Officer, via the local authority,provides claimants and their prospectivelandlord with a valuation of the rent for HBpurposes on individual properties, weregenerally thought to be a good thing. Inone area, PTD applications were falling,while in another they were increasing. Aswell as providing claimants withinformation about the rent that would beaccepted by the local authority for HBpurposes, PTDs helped to speed up HBclaims because the Rent Officer hadalready inspected the property and made avaluation. They also put Rent Officers intomore direct contact with landlords, whichhelped them to collect market evidence.However, the procedure of joint applicationby landlord and tenant could be a problemas some tenants did not want theirprospective landlord to know they wouldbe claiming HB in case this stopped themgetting the tenancy.

6.5 Future Research

Most of the research to date on the impactof HB on the private rented sector hasbeen focused on England or covers onlyCardiff rather than the rest of Wales. Theremay therefore be merit in commissioningresearch on the role of HB in the PRS in

53


Wales, covering a range of housing marketcircumstances. It would be useful to get abetter insight into the impact of HBadministration on tenants and theirlandlords, although it may be prudent todelay this until after proposed reforms toHB administration set out in the HousingGreen Paper. It would also be useful togain further insights into the operation andimpact of pre-tenancy determinations.Finally, the impact of rent restrictions onclaimants and landlords deserves furtherstudy in Wales.

6.6 Summary of Findings

• The number of private tenantsreceiving HB in Wales increasedduring the first half of the 1990s andfell during the second half of thedecade.

• In 2000, the number of privatetenants on HB was 47,000 (comparedwith 51,000 in 1992), of whom nineout of ten had deregulated tenancies.

• There are significant differencesbetween private tenants on HB andthose who are not. Differences existin terms of income and savings,household composition, and someaspects of housing quality such as thepresence of central heating in thedwelling.

• Knowledge of HB among managingagents and private landlords variesaccording to their experience withletting to this part of the market

• Attitudes and perceptions ofmanaging agents and privatelandlords to tenants on HB variesbetween different housing marketareas

• Most private landlords and managingagents prefer not to let to tenants onHB, though many do in practice and aminority prefer them

• Poor administration of HB, recoveryof overpayments where the HB ispaid direct, and the impact of rentrestrictions are the main reasons whylandlords and managing agents arenegative towards HB

• The main advantage of HB is that therent is guaranteed once HB is inpayment and paid direct to thelandlord

• Rent Officers experience difficulties infinding market evidence in someareas and for certain types ofproperty; this is especially true of thepre-July 2001 definition of the singleroom rent

• Landlords and managing agents saythat the response of tenants to rentrestrictions varies (re-negotiating therent, making up the shortfall, gettinginto rent arrears, moving to anotherproperty)

• Pre-tenancy determinations canprovide a lever for HB recipients tonegotiate a lower rent

54


55


7.1 Data Review


The General Household Survey (GHS)contains information on the type oflandlord people rent from. This could beused to provide an insight into the longterm changing type of landlord operatingin the private rented sector, for exampleemployers offering accommodation tied toa job, other institutional landlords andprivate individuals. Because of the size ofthe GHS, this could only be done at the all-Wales level.


The 1998 Welsh House Condition Surveycontains information regarding type ofprivate landlord according to the followingcategories:

1) employer (organisation) of householdmember,

2) another organisation,

3) relative or friend of householdmember,

4) employer (individual) of householdmember,

5) other individual landlord.

The 1998 Welsh House Condition Surveyalso contains a question on whether or nota housing grant has been received toimprove the property.

7.2 Data Analysis

Figure 7.1 shows the breakdown of privatelandlords into the five different categorieslisted above. Overall, 6% of households inthe PRS rent from their employer. Asignificant minority, 12%, rent from afriend or relative. However, the majority(74%) rent from individuals who are notan employer or a friend/relative.

Six per cent of private renting householdsin the 1998 WHCS reported that theirdwelling had been improved with ahousing grant compared to 7% of owneroccupiers. However, 36% of privaterenters said they did not know if thedwelling in which they live had beenimproved with a housing grant. If the‘don’t knows’ are excluded, 9% of theremaining private renters said theirdwelling had been improved with a grant,actually higher than the 8% of owneroccupiers (excluding the ‘don’t know’responses from owner occupiers).

Chapter Seven Landlords, Housing Grants and Tenancy Types


Landlords and letting agents reported thatmost landlords they deal with fall into oneof two categories: ‘investment’ landlordswho have made the decision to become alandlord for financial reasons; and‘accidental’ landlords who no longerrequire to live in their owned property andtherefore decide to place it on the rentalmarket. This latter group includes peoplewho move abroad and wish to maintain abase in the UK, as well as people whomove within the UK for employmentreasons. Often people are unable orunwilling to sell their home in certain partsof Wales where house prices areparticularly low, notably Merthyr Tydfil ofthe four case study areas.

Many investment landlords have very smallproperty portfolios, in the majority of casesonly one or two properties. This type ofsmall scale landlordism has beenencouraged by ARLA supported Buy-to-Let

mortgages throughout the 1990s and morerecently by other mortgage lenders whonow offer increasingly competitive buy-to-let mortgages. Many lenders now offerstandard mortgage lending rates for buy-to-let. However, the maximum loan tovalue ratio is generally 75% and areference from a letting agent, notnecessarily a member of ARLA, is requiredstating that the property in question can beexpected to generate a certain rentalincome without significant periods ofvacancy. Also, some lenders have loosenedthe restrictions on the number of buy-to-letmortgages an individual may hold. Manyindividual investment landlords see theirproperty as a ‘nest egg’ which they intendto sell at retirement.

While the majority of landlords are smallscale investment landlords who are in paidemployment and generally use lettingagents, some landlords with a number ofproperties in an area manage the propertiesthemselves and make landlordism their full-

56


Figure 7.1 Type of Private Landlord

Source: WHCS, 1998

Employer (organisation) of h/h memberAnother organisation

Relative or friend of h/h memberEmployer (individual) of h/h member

5%

7%

12%

2%

74%

Other individual landlord

57


time business. Some are also qualifiedbuilders and are involved with propertyimprovements.

Students’ parents are increasinglybecoming landlords. This increase inSwansea was concentrated in a period ofrapid expansion in student numbers inSwansea during the early 1990s and hassince stabilised. However, this trend ofincreasing landlordism among students’parents continues in Aberystwyth. It wasthought by a student accommodationofficer and a local letting agent that thistrend can be accounted for by theintroduction of student loans andincreasing house price differentials betweensouthern England and Wales makingproperty purchase in Wales appear moreaffordable. In many cases students’parents do not sell the property when theirsons and daughters graduate but rathercontinue to rent the property out, thusthere is a degree of cumulative increase insupply in university towns.

Over-supply of private rentedaccommodation in certain parts of Wales is,or threatens to become, a problem. Thepopularity of buy-to-let, the perceivedbuoyancy of the Cardiff economy in excessof its actual buoyancy, and students’parents becoming landlords were cited by anumber of different types of interviewee ascontributing towards over-supply.

Many interviewees indicated that thebalance between demand and supplydiffered with regards to different types ofproperty. In most locations, demand fortwo and three bedroom flats outstrippedsupply. One letting agent suggested thatthis was because some first time landlordsbought a property that they themselveswould like to live in, without consulting a

letting agent or considering the types ofpeople who are most likely to rent.Consequently, there is over supply in somelocations of larger properties and non-flatted accommodation in particular. This isconsistent with lower yields for these typesof properties for the UK as a whole(University of York, 2000).

Housing Advice Centres, and to a lesserextent Environmental Health Officers andlocal authority Housing Officers, reportedproblematic behaviour from somelandlords. The main problems highlightedwere a lack of investment in repair andmaintenance resulting in dampness, thenon-return of deposits and landlordsserving less than the two months statutorynotice to quit on tenants under assuredshorthold tenancies or, in a smaller numberof cases, serving notice to leave within theassured six month period. Other problemsincluded landlords expecting access toproperties while tenants are resident andnon-flexibility over late payment of rentwhile Housing Benefit claims are beingprocessed.

There was a general consensus across allinterview groups that it would be desirableto increase the level of grant assistanceavailable to private landlords for repairs.However, some Housing Officers andEnvironmental Health Officers reported adegree of unease, particularly from electedmembers, at the prospect of offering grantsto for-profit landlords. In the climate ofreduced availability of resources to localauthorities for housing grants generally,Basic Credit Approval is usually used tofund improvement grants for owneroccupiers. Some of the local authoritiesinterviewed had secured SupplementaryCredit Approval but this is in most casesrestricted to fund improvements of HMOs

58


in order to meet fire regulations. Both thelandlord representatives interviewed saidthat restrictions on Housing Benefit hadlimited landlords’ ability to carry out repairsand maintenance, while some of the lettingagents interviewed said that this had hadeither no or only marginal impact. This canmost likely be accounted for by the factthat some of the letting agents interviewedwere operating towards the middle andupper end of the private rental market,while the two landlord representativesinterviewed had many members operatingtowards the bottom end of the market.

Housing Advice Centres, Housing Officers,and Environmental Health Officers werestrongly in favour of Bond Boards inprinciple. However, there was a degree ofconcern that they are not attractive tolandlords, a view borne out by the landlordrepresentatives and letting agents who saidthere was no benefit or incentive fromlandlords’ point of view to let to a tenantusing a Bond Board rather than one payinga cash deposit. One letting agent reportedthat a local landlord had used the localBond Board and when the tenant left theproperty in a poor state of cleanliness theBond Board would not pay to have theproperty cleaned. Word of this anecdotespread between landlords in the area andmay have had a detrimental impact of thesuccess of the Bond Board overall. Thishighlights the importance of Bond Boardsin being seen to be fair and to deliver abenefit to the landlord as well as the tenantin order for landlords to have faith in usingBond Boards.

One means around these problems wouldbe to create a compulsory custodial bondscheme whereby an independent body heldall tenants’ deposits and assessed theproportion to be returned to the tenant at

the termination of a lease. However, anumber of interviewees questioned thepracticability of this, particularly withregard to the speed with which such anorganisation could realistically inspectproperties at the beginning and end oftenancies. In addition, one letting agentwas of the view that most letting agentsrun ‘client accounts’ to hold deposits andperform this brokerage function overdeposits between tenants and landlord in asatisfactory manner, therefore public sectorinvolvement in this area would duplicateexisting efforts and may crowd out goodpractice in the private sector.

Finally, in terms of attitudes towardsdifferent types of tenancies, there appearsto be a high degree of satisfaction withshorthold tenancies among both landlordsand tenants. Letting agents reported thatfew tenants wanted longer leases as manytenants are uncertain of where they will beworking in six months or a year’s time.Some tenants, however, preferred thesecurity of renewing a shorthold at the endof the assured six months in order toobtain a further assured six months ratherthan allowing the tenancy to go ‘periodic’and only have two assured months. Inaddition, some tenants at the bottom endof the private rental market were reportedby Housing Advice Centres and landlordrepresentatives to be vulnerable to unfaireviction from a minority of landlords.

Generally, landlords were reported to behappy with the existing tenancyarrangements as they could regainpossession of properties relatively quickly.However, at the bottom end of the privaterental market, some landlords werereported to be of the view that they wereunsupported by the legal system as inpractice it was not cost or time effective to

pursue tenants through the courts for rentowed during the assured six months of ashorthold tenancy if the tenant hadvacated during that period. A betterstrategy was to refill the property asquickly as possible.

However, these findings relate only tolandlords and tenants currently in theprivate rented sector. It does not provideinsights regarding prospective landlordsand tenants who may have beendiscouraged from entering the privaterented sector because of the currenttenancy arrangements. A number ofinterviewees were of the opinion thatassured shorthold tenancies are notattractive to people seeking a long termhome, particularly older people andfamilies with children. From landlords’perspective, the view was also expressedby one letting agent that a private rentalmarket with lower turnover of tenantswould be more attractive to institutionalinvestors who would therefore perhapsprefer to offer assured tenancies.

7.4 Future Research

Given the excess supply of certain types ofproperties for private rent in certainlocations, and excess demand in others, itwould be beneficial to policy makers tobetter understand the reasons peoplebecome landlords and how they decidewhat type and location of property topurchase. Promoting buy-to-let mortgagesis effective in expanding the private rentedsector generally, but this is a bluntinstrument in persuading landlords to lookclosely at the market segments in shortestsupply. Landlords need to be persuaded toexamine economic indicators such as rents,house prices and yields prior to purchasingin order to deliver the most appropriate

types of accommodation in the mostappropriate locations. There is a potentialto increase the involvement of estate andletting agents in assisting prospectivelandlords in this regard.

A review of housing grants, particularly thebenefits derived from grants received byprivate landlords versus tenants, would beof interest. If tenants resident in privatelyrented properties which were improvedwith the use of a local authority housinggrant were shown to benefit from thatimprovement, and that the improvementwould not have occurred without thegrant, then there may be more politicaland financial support for housing grants toprivate landlords in the future.

Given the high degree of scepticism amonglandlords and letting agents about BondBoards and their apparent low usage ofthem, a review of Bond Boards acrossWales would be of great benefit inensuring their success in the future. Theresearch team notes that the NationalHousing Strategy for Wales proposes tocarry out such a review in 2001/02. Wewould urge this review to includelandlords’ and letting agents’ perspectivesas well as those of tenants.

In general, there appeared from theinterviews to be a low level of trust andco-operation between private landlordsand local authorities. A high degree ofcynicism and frustration was reported to befelt by landlords towards local authorities,particularly with regard to reportedinconsistencies in enforcement of HMOregulations and delays in processingHousing Benefit claims. Equally, localauthorities and Housing Advice Centres,who are subject to the worse conditionsand practices found in the private rented

59


60


sector, appeared to be of the opinion thatlandlords were uncooperative andsometimes unscrupulous. No localauthorities knew of any local privatelandlord forums. However, in Angleseyand Powys, Environmental Health andHousing Departments reported that theyhad good relationships with selected locallandlords and referred people looking forhousing to those landlords. However,means to increase the co-operation andtrust between private landlords and localauthorities could be usefully investigated.This is an important area to foster, asprivate landlords play an important role inmeeting housing need.

7.5 Summary of findings

• Eighty-seven per cent of tenants’landlords are individuals, of which12% are a friend or relative of thetenant,

• seven per cent of private tenants rentfrom their employer,

• seven per cent of tenants’ landlordsare organisation (other than theiremployer),

• approximately 9% of PRS and 8% ofowned properties have beenimproved with a housing grant,

• the majority of landlords are smallscale either ‘investment’ or‘accidental’ landlords,

• buy-to-let mortgages haveencouraged small scale investmentlandlordism throughout the 1990s,

• there is over-supply of largerproperties in some locations,

• there is under-supply of two andthree bedroom flats in some locations,

• a small minority of landlords andtenants are unscrupulous,

• reduced availability of housing grantsover recent years and Housing Benefitrestrictions are blamed by landlordrepresentatives and some lettingagents for low levels of repair in theprivate rented sector,

• Bond Boards were supported inprinciple by most interviewees, butthere was some concern expressedover their attractiveness to landlords,

• a high degree of satisfaction withassured shorthold tenancies wasreported regarding landlords andtenants, notwithstanding someunscrupulous behaviour at thebottom end of the market from bothlandlords and tenants,

• however, assured shorthold tenanciesare unlikely to attract many tenantslooking for a long-term home,

• individual landlords are unlikely to bepersuaded by assured or longer termshortholds, but institutional landlordsmay not be deterred by such types oftenancy.

8.1 Data Review

8.1.1 1993 and 1998 Welsh HouseCondition Surveys

The household interview components ofthe 1993 and the 1998 Welsh HouseCondition Surveys can be used to assesschanges in the characteristics of propertiesand tenants in the private rented sector,and these changes can be compared tochanges in housing as a whole. Thephysical dwelling inspections can be usedto assess changes in the level of unfitnessand the cost of repairs required to meetfitness requirements. This can bedisaggregated, of course, by tenure.


The GHS is available annually from 1971and provides detailed information onhousing, household and individual personalcharacteristics. In any year, it hasinformation on approximately 8,000households across the UK, which generatesaround 40 private rented sectorobservations in Wales. Therefore, analysiscould only be carried out at all-Wales level,and adjacent years would need to becombined to generate robust sample sizes.The geographical indicators contained inthis dataset are Standard Regions.

The GHS contains information on propertytype and size as well as demographic andsocio-economic information relating tohouseholds and individuals. Long termchange in the size and type of propertiesand the characteristics of tenants in theprivate rented sector could therefore beassessed.

8.1.3 Family Expenditure Survey

The Family Expenditure Survey (FES) isanother UK wide on-going survey whichcollects information on approximately8,000 households each year, the same sizeas the GHS. The FES is available from1961. The FES, being a survey ofexpenditure, has information on the levelof rent paid and rebate received (currentlyHousing Benefit). It also providesinformation on the size of properties,whether they have central heating or not,whether rented properties are furnished ornot, and where properties are located. Thegeographical indicators contained in thisdataset are Standard Regions. The FES alsoprovides demographic and socio-economicinformation regarding individuals andhouseholds. As with the GHS, the FEScould be used to provide a picture of longterm change in the characteristics ofproperties and tenants at the all-Waleslevel, although the information in thisregard is less detailed than that containedin the GHS.

8.1.4 1991 and 2001 Censuses ofPopulation

Given near 100% coverage of thepopulation, this can provide information onthe precise spatial distribution ofhouseholds and persons in the privaterented sector in Wales. This can be doneat the level of all Wales, former counties,former districts, wards or EnumerationDistricts (EDs) using the Small AreaStatistics (SAS) and the Local Base Statistics(LBS).

61


Chapter Eight Changes in the Private Rented Sector

The 1991 Census provides information byhousing tenure on the number of rooms indwellings, dwelling type (detached house,terraced house, flat, etc.) and amenities(e.g. central heating and shared WCfacilities). Information on household size,household structure, overcrowding and carownership are also provided by housingtenure. Housing tenure in the Censusdifferentiates between furnished andunfurnished private rented properties andaccommodation that is rented from thetenant’s employer. These data refer tohouseholds rather than dwellings.

The 2001 Census of Population haschanged relatively little compared to the1991 Census. This means that the abovecharacteristics could be compared between1991 and 2001 at the level of all Wales,wards or Enumeration Districts (EDs). 1991Census data could be converted into thenew local authority boundaries forpurposes of comparison at this geographyby aggregating 1991 data from therelevant wards or EDs. Detailed 2001Census data will not be released for sometime.

8.1.5 Joseph Rowntree Foundation/University of York Index of Private Rents andYields

The University of York Index of PrivateRents and Yields (formerly the JosephRowntree Index of Private Rents andYields) consists of two indices. The first isthe Valuations Index which is based onRent Officers’ valuations of the typical rentfor six different types of common privaterented properties, each with a prescribedset of attributes. This is done at the localauthority district level and is standardisedto take account of the type of propertiesthat are more likely to change tenants

therefore be included in a Rent Officer’svaluation which is based on evidence fromrecent transactions. An adjustment is alsomade for the property type mix in eachdistrict.

The second index produced by theUniversity of York, the Transactions Index,is based, as the name suggests, on actualtransactions in the private rented sector.Data on these are provided by a sample ofmembers of the ARLA and other principleletting agents. Note that this is biasedtowards the middle and upper end of theprivate rental market because landlords atthe bottom end of the private rentalmarket are less likely to use letting agents.The Transactions Index is available atdistrict level, and for selected smallerpostally based units, usually the main UKcities, when a sufficient number oftransactions are present in order to providea reliable estimate of market rent. Thisindex is standardised through time to allowfor the fact that different segments of themarket will have a higher proportion of alltransactions at different points in time e.g.students taking up private tenancies inSeptember/October, and longer scalevariations with the economic cycle. Aswith the Valuations Index, an adjustment isalso made for the property type mix ineach district.

The York Indices provide time series datasince the first quarter of 1996 on the rentcharged for different types of property andthe yield returned to landlords. Yield isannual rent as a proportion of the openmarket vacant repossession value of theproperty. In the Valuations Index, this isbased on valuations of the same propertiesas the rental valuations have been madeon. However, in the Transactions Index,property values are taken from Halifax Plc

62


63


sale prices, therefore rental yields in theTransactions Index are estimates.

8.2 Data Analysis

8.2.1 Change in scale

Table 8.1 below shows for 1986, 1992 and1997 the number of households living inthe private rented sector and the totalnumber of households in Wales. The finalcolumn in the table shows the percentageof households in the private rented sector.

The private rented sector expanded in sizeconsiderably between 1986 and 1992,

from 81,789 to 88,300 households, anincrease of 8%. However, this was againsta background of a significant increase inthe total number of households in Wales,reflecting strong growth in the number ofhouseholds across the UK as a whole,meaning that the private rented sector’sshare of all households in Wales actuallyfell slightly from 7.9% to 7.8% of allhouseholds. In contrast, between 1992and 1997 the private rented sectorexpanded only marginally against abackground of continued increases in thenumber of households overall.

Year Privately renting % of householdsHouseholds All households privately renting

1986 81,789 1,035,000 7.9

1992 88,300 1,126,100 7.8

1997 88,536 1,157,280 7.7

Source:s: 1986, 1993 and 1998 Welsh House Condition Surveys

Table 8.1 Absolute and relative size of the private rented sector in Wales: 1986, 1992 and1997

An important question to consider inregard to the possible expansion of the PRSis the future housing tenure aspirations ofpeople currently in different housingtenures. Social and private renters in theWelsh House Condition Survey were askedwhat their preferred housing tenure is (thiswas not asked of owner occupiers). Theresults are shown in table 8.2 below.

Over half of social renters are currently intheir tenure of first preference, comparedto only a quarter of private renters. Overhalf of private renters aspire to owneroccupation and 40% of social renters.

64


Table 8.3 shows the proportion ofhouseholds who are on a social landlord’swaiting list, as an indication of the level ofneed for people to move out of privaterenting. Under 2% of owner occupiers are

on a social housing waiting list comparedto 12% of private renters. Significantlymore private renters are on a socialhousing waiting list than say that socialhousing is their preferred tenure.

Preferred tenure Current Tenure Total

Social rented Private Rented% % %

Own occupation 40 52 43

Social Rented 52 7 40

Private Rented 2 26 8

Other 0 4 1

Don’t mind 4 7 5

Don’t know 3 5 4

Total 100 100 100

Source: WHCS, 1998

Table 8.2 Tenure aspirations of social and private renters

On social housing Tenure Totalwaiting or transfer list


Yes 2 11 12 4

No 98 90 89 96

Total 100 100 100 100

Source: WHCS, 1998

Table 8.3 Social housing waiting list by tenure

65


8.2.2 Change in dwelling characteristics

Chapter four showed that dwellings in theprivate rented sector had a higherincidence of unfitness compared to other

dwellings in Wales. Is this situation gettingbetter or worse over time? Table 8.4below shows the percentage of dwellingsin different tenures which were unfit in1993 and in 1998.

Tenure 1993 1998 Change in share ofdwellings unfit

% share % share % change

Own occupation 12 8 -36%

Social Rented 16 8 -48%

Private Rented 26 18 -28%

All dwellings 13 9 -37%

Source: 1993 and 1998 Welsh House Condition Surveys.

Note: The fitness standard for human habitation used in both the 1993 and 1998 Welsh House Condition Surveys was that set out in

Section 604 of the Housing Act 1985 as amended by Schedule 9 of the Local Government and Housing Act 1989

Table 8.4 Proportion of dwellings unfit by tenure, 1993 and 1998

Overall, the level of unfitness of housing inWales fell dramatically between 1993 and1998, from 13% of all dwellings to 9% ofall dwellings, a reduction by some 37%.However, the reduction in unfitness in theprivate rented sector fell by 28% (from26% to 18% of dwellings), a smallerpercentage fall than in other tenures.Therefore, although the rate of unfitnessfound in the private rented sector hasimproved considerably, its position relativeto other tenures has worsened.

Table 8.5 shows the average cost of repairsper dwelling required to meet standards forhuman habitation for the same two years.The reduction in the average repair costhas been more marked than the reduction

in the proportion of dwellings unfit, withan overall reduction in the average repaircost for housing in Wales by 64% in realterms. This indicates that the properties inthe worst condition have been upgraded,consistent with the ‘worst first’ strategy inadministering housing grants that wasreported by the local authoritiesinterviewed in the course of this research.However, once again, the improvement inthe private rented sector lags behind theimprovement in other housing tenures,with a 50% reduction in the private rentedsector average repair cost. This left theaverage repair cost for a privately rentedproperty at a not inconsiderable £1,883 in1998.


66

Tenure 1993 1998 Change in averageAverage repair Average repair repair cost(£, 1998 prices) cost (£) (%)

Own occupation 2,611 951 -64%

Social Rented 1,961 654 -67%

Private Rented 3,739 1,883 -50%

All dwellings 2,611 953 -64%


Table 8.5 Average repair cost per dwelling to meet fitness standards by tenure, 1993 and1998

There was a reduction in the prevalence offlats in the private rented sector between1992 and 1997. A quarter of householdsin the private rented sector lived in flats in1992, but this had fallen to a fifth by 1997.This is a reduction in share by 21%,compared to a reduction by only 9% in theshare of all households living in flats inWales. Since flats are over represented inthe private rented sector, this shows thatthe private rented sector is becoming moresimilar to the rest of the Welsh housingstock through time. Unfortunately, figuresfor other types of dwellings cannot bemeaningfully compared between 1992 and1997 because of changes to the categoriesof dwelling types reported.

Table 8.6 shows changes in the size ofdwellings found in the private rented sectorcompared to housing as a whole. Thisagain shows a general convergence

between the private rented sector andhousing as a whole in Wales. In 1992,one, two, three and four room dwellingswere over represented in the PRS.Between 1992 and 1997 the PRS growthwas greater, or decline lesser, than thehousing stock as a whole in these sizes ofdwelling. Similarly, dwellings with five, sixor seven or more rooms were underrepresented in the PRS in 1992 and thesesize categories experienced lesser declinethan the housing stock as a whole orgrowth in the face of slight decline in thecase of five room dwellings. Broadlyspeaking, the proportions of both PRSdwellings and dwellings as a whole aredeclining in the cases of very small andvery large dwellings and growing in themid-ranges. This reflects the pattern ofdemographic change in household sizewhich is shown in the following sub-sectionof this chapter.

67


Dwelling size 1992 1997 Change in % share(no rooms) PRS All PRS All PRS All

(% share) (% share) (% share) (% share) (% share) (% share)

One 5 1 2 0 -62 -50

Two 14 2 10 5 -25 148

Three 15 7 25 17 60 162

Four 22 18 26 30 21 63

Five 19 30 21 30 13 -0

Six 16 28 9 11 -43 -63

Seven or more 10 15 7 7 -30 -52

Total 100 100 100 100 - -


Table 8.6 Size of privately rented dwellings and all dwellings, 1992 and 1997

The foregoing analysis has assessedchanges in dwelling characteristics between1992 and 1997. How is the private rentedsector in Wales changing in the longer

term? Selected characteristics for selectedyears from 1970s are shown in table 8.7below.

Year No. Observations Average number % PRS Dwellings % PRS Dwellingsof Rooms per with Central that arePRS dwelling Heating Furnished

1970 49 5.0 8 24

1975 45 5.2 11 18

1980 19 5.3 5 21

1989 21 4.9 48 48

1995 30 5.1 60 47

1999 21 4.2 76 29

Source: Family Expenditure Survey

Table 8.7 Long-term change in dwelling size, dwellings with central heating and furnisheddwellings in the private rented sector in Wales

There appears to be no long term trend inthe average number of rooms in privatelyrented dwellings in Wales, with the averageapparently stable at around five rooms perdwelling (although bear in mind that thesefigures are based on small samplenumbers). Unsurprisingly, the proportionof privately rented dwellings in Wales withcentral heating is rising dramatically overtime. Finally, the proportion of dwellingsthat are let furnished rose steeplythroughout the 1970s, reflecting thedecline of the private rented sector inproviding permanent homes (which aregenerally let unfurnished) during that time.The drop between 1995 and 1999 may bea statistical aberration.

8.2.3 Change in tenant characteristics

Table 8.8 shows that, overall, between1992 and 1997 in Wales the proportion ofsingle person households remained stablewhile the proportions of two and threeperson households grew and theproportions of larger households fell. Thepattern of change in the size distribution ofhouseholds in the private rented sector,however, is considerably different. Here,the proportion of single person householdsfell by 28.7% and the proportion of allother household sizes grew. The growthwas particularly strong in the case of largerhouseholds, with an increase of over 60%in the share of PRS households of six ormore people compared to a decline ofnearly 10% in the share of all householdsin this size category. As with the size ofdwellings, this shows a convergencebetween the size of households in theprivate rented sector and householdsgenerally.


68

Dwelling size 1992 1997 Change in % share(no rooms) PRS All PRS All PRS All


One 45 25 32 25 -29 -0

Two 26 34 31 34 22 2

Three 13 17 16 17 23 0

Four 11 16 12 16 12 -1

Five 4 6 6 6 36 -3

Six 2 2 3 2 61 -9

Total 100 100 100 100 - -


Table 8.8 Household size in the PRS and of all households, 1992 and 1997

Characteristic 1992 1997 Change in % sharePRS All PRS All PRS All


Age of HoH

Under 30 years 40 12 34 10 -15 -15

Household type

Single pensioner 12 16 11 16 -7 1

Single non-pensioner 32 9 21 9 -34 -1

Table 8.9 shows the age of householdheads and selected household types for1992 and 1997, both for the privaterented sector and for all housing tenures.(Only ‘single pensioner’ and ‘other single

person’ household type categories werecommon to both the 1993 Welsh HouseCondition Survey and the 1998 WelshHouse Condition Survey).

69



Table 8.9 Age of head of household (HoH) and household type for the PRS and allhouseholds/individuals, 1992 and 1997

The proportion of household in the privaterented sector which are headed by anindividual under the age of 30 years fellfrom 40% in 1992 to 34% in 1997,although this remains substantially greaterthan the 10% of household heads as awhole in this age group in 1997. Althoughthe proportion of household heads in theprivate rented sector under the age of 30years is falling, the reduction is broadly inline with the reduction in the proportion ofall household heads in this age group.However, the fall in the proportion ofhouseholds headed by an individual underthe age of 30 years was slightly greateramong private renters than among allhouseholds - the reductions between 1992and 1997 were just over 15% for privaterenters and 14.5% for all households.

Since young people are over represented inthe private rented sector, this shows aslight tendency towards convergence withother housing tenures but it is not marked.

The private rented sector lost its share ofsingle pensioner households relative to thepopulation as a whole, with a reductionfrom 12% in 1992 to 11% in 1997,compared to a stable figure of just over15% of all households. The proportion ofprivate renting households in the ‘singlenon-pensioner’ category fell by over athird from 32% in 1992 to 21% in 1997,while this proportion remained stable atjust over 9% of all households. This onceagain shows that the characteristics of theprivate rented sector are converging withother housing tenures over time.

8.2.4 Change in rents and yield, 1996-2000

Figure 8.1 shows weekly rent (standardisedfor changes in the mix of property typesover time) in Wales and Britain (excludingLondon) for each quarter from the firstquarter of 1996. The first thing to notefrom figure 8.1 is that standardised rentsare generally lower in Wales than in therest of Britain. Both Welsh and Britishrents are on a broadly similar upwardtrajectory, although rents in Walesstagnated throughout 1998 while theycontinued to rise in the rest of Britain, andsince the second quarter of 2000 rents inWales have fallen while they haveincreased sharply in Britain, although it isnot clear from the data if this divergencewill continue.

Standardised rents in Wales and Britain(excluding London) were stable in realterms (1996 prices) at around £78 perweek and £87 per week respectively fromthe start of 1996 until the first quarter of1998, as shown in figure 8.2. Since thefirst quarter of 1998, however,standardised rents in Wales have beenfalling in real terms with a period ofstability in early 1999, and with sharp fallsbetween the second and fourth quarters of2000. Standardised real rent fell in Britainafter the first quarter of 1998 in line withWales, but stabilised and since the end of1998 has been rising at an increasing rate

with particularly steep rises since thesecond quarter of 2000, the same timeWelsh rents started to fall steeply.

Despite lower rents in Wales than in therest of Britain, yields are higher as shownin figure 8.3 owing to lower house pricesin Wales. Indeed, Wales’ position relativeto the rest of Britain appears to beimproving with regard to rental yieldachieved in the private rented sector. Inthe early part of 1996, yields were actuallyslightly higher in Britain than in Wales butthis position switched after the thirdquarter of 1996. Yields in Wales andBritain then both fell sharply until thesecond quarter of 1997 when Welsh yieldsrose again steeply and have sincefluctuated around a slight downwardtrend, while British yields continued todecline and have only recovered slightlysince the second quarter of 2000.

The relatively unstable trends in rentalyields compared to the relatively smoothtrend in rents illustrates that changes inrental yields are in large part determinedby changes in house prices. Furthermore,increases in rent are not keeping pace withincreases in house prices leading todeclining rental yields. This is likely to actas a deterrent to new landlords enteringthe private rental market, but should notadversely affect existing landlords who willenjoy capital appreciation on theirproperties.


70

71


70

75

80

85

90

95

100

105

Weekly

Rent (

£)

Britain (excl London)Wales

96Q

1

96Q

2

96Q

3

96Q

4

97Q

1

97Q

2

97Q

3

97Q

4

98Q

1

98Q

2

98Q

3

98Q

4

98Q

1

99Q

1

98Q

1

99Q

2

99Q

3

98Q

1

99Q

4

00Q

1

00Q

2

00Q

3

00Q

4

72.0

74.0

76.0

78.0

80.0

82.0

84.0

86.0

88.0

90.0

92.0


96Q

1

96Q

2

96Q

3

96Q

4

97Q

1

97Q

2

97Q

3

97Q

4

98Q

2

98Q

3

98Q

4

99Q

1

99Q

2

99Q

3

98Q

1

99Q

4

00Q

1

00Q

2

00Q

3

00Q

4

Weekly

Rent, 1996 p

ric

es (

£)

8.0

8.1

8.2

8.3

8.4

8.5

8.6

8.7

8.8

8.9

9.0


96Q

1

96Q

2

96Q

3

96Q

4

97Q

1

97Q

2

97Q

3

97Q

4

98Q

2

98Q

3

98Q

4

99Q

1

99Q

2

99Q

3

98Q

1

99Q

4

00Q

1

00Q

2

00Q

3

00Q

4

Yie

ld (

%)

Source: JRF/University of York Index of Private Rents (Valuations Index)

Figure 8.1 Weekly rent, Britain (excluding London) and Wales, 1006-2000

Figure 8.2 Real terms weekly rent, Britain (excluding London) and Wales, 1996-2000

Source: Derived from JRF/University of York Index of Private Rents (Valuations Index) and RPLX

Figure 8.3 Rental yield, Britain (excl London) and Wales, 1996-2000


This section so far has considered changesthrough time in rents. However, there isvariation in rent, and variation in the rateof change in rent, both geographicallywithin Wales and for different types ofproperties. Weekly rents in Walesdisaggregated by local authority and bytype of property for the fourth quarter of2000 are shown in table 8.10. Blank cellsare due to insufficient or missing data. Thehighest rents for most property types canbe found in Anglesey, Bridgend, Caerphilly,Cardiff, Ceredigion, Conwy, Gwynedd,Monmouthshire, Newport, Swansea andthe Vale of Glamorgan. These areas aregenerally, but not exclusively, areas with ahigher than average share of dwellings inthe private rented sector, as shown inchapter four. This is perhaps due to highdemand increasing both supply ofproperties and rent levels. The lowest rentsfor most property types are generally foundin Blaenau Gwent, Merthyr Tydfil, Powysand Rhondda Cynon Taff. These aregenerally the less economically buoyantareas in Wales, and low rents are likely tobe a reflection of low incomes and lowhouse prices.

Table 8.11 shows the average annualpercentage change (discounted) in real rentbetween the fourth quarter of 1997 andthe fourth quarter of 2000, againdisaggregated by local authority and bytype of property. Note that inflation hasbeen deducted so the changes are in realterms. There is a wide range of changes inrents, both increases and decreases, butwith decreases slightly out-numberingincreases, reflecting the overall slight fall inrent in real terms in aggregate across Walesduring this period as shown in figure 8.2.There is no obvious pattern by type ofdwelling, although there appears to begreater variation (i.e. larger increases andlarger decreases) between local authoritiesfor detached and semi-detached propertiescompared to other property types. Falls inreal rent for terraced properties appear tobe the most consistent between localauthorities, with many of the falls in theregion of 2.0-4.5% per annum.


72

73

Welsh A

ssembly G

overnment - Review


Local Authority Unfurnished Furnished

D’ched Semi Terrcd Flat Flat D’ched Semi Terrcd Flat Flat Bedsits2 bed 1 bed 2 bed 1 bed

Anglesey 105 85 80 67 55 110 90 80 70 60 38

Blaenau Gwent 84 70 58 58 50 87 76 60 58 53 40

Bridgend 100 85 75 70 65 100 85 75 70 65 50

Caerphillly 110 81 69 62 55 120 87 74 665 58 48

Cardiff 185 127 98 90 75 40

Ceredigion 100 75 75 98 115 85 80 78 45

Carmarthenshire 85 80 70 70 60 85 80 70 70 60 45

Conwy 115 92 85 75 67 127 100 90 80 70 40

Denbighshire 95 81 75 65 60 95 81 75 65 60 38

Flintshire 98 83 75 70 63 98 83 75 70 63 45

Gwynedd 110 90 85 70 55 120 100 85 75 60 38

Merthyr Tydfil 75 67 63 30 55 80 67 63 60 55 40

Monmouthshire 133 95 80 81 73 138 98 84 84 76 46

Neath Port Talbot 87 70 60 65 55 100 75 70 68 60 38

Newport 115 87 76 74 62 115 92 81 78 65 42

Pembrokeshire 85 80 75 70 65 85 83 75 70 65 40

Powys 62 72 62 64 56 65 72 65 64 58 44

Rhn’da, Cynon Taff 80 65 60 60 55 85 68 62 62 58 40

Swansea 115 80 65 70 55 150 80 70 80 65 35

Torfaen 98 81 62 60 54 104 87 65 62 56 39

Vale of Glamorgan 150 110 86 83 70 39

Wrexham 95 81 65 63 55 95 81 65 63 55 40


Table 8.10 Average w

eekly rents, by property type and local authority, 2000 Q4

Welsh A

ssembly G

overnment - Review


74

Local Authority Unfurnished Furnished

D’ched Semi Terrcd Flat Flat D’ched Semi Terrcd Flat Flat Bedsits2 bed 1 bed 2 bed 1 bed

Anglesey 1.5% -3.4% 2.8% -2.8% -2.3% 3.1% -1.5% 2.8% -1.3% 0.7% -1.4%

Blaenau Gwent -1.9% -1.3% -2.3% -2.3% -2.3% -0.7% 1.5% -1.2% -2.3% 0.3% -2.3%

Bridgend -0.6% -0.2% -2.3% -3.7% -3.8% -2.3% -2.3% -4.4% -4.6% -4.7% -0.9%

Caerphillly 0.3% 0.3% -0.3% -2.3% -2.3% 3.3% 2.8% 2.1% -0.7% -0.5% 2.3%

Cardiff 11.1% 2.9% -1.2% -1.2% -1.4% 0.3%

Ceredigion 50% -4.3% 6.2% -2.3% -2.7% 2.4% 1.7%

Carmarthenshire -2.3% -2.3% 0.2% 0.7% -4.6% -2.3% 5.4%

Conwy 0.8% -1.6% -2.3% -3.6% -0.2% 2.6% -0.6% -2.3% -2.3% 0.2% -2.3%

Denbighshire -4.0% -2.3% 0.0% -2.3% -0.6% -4.9% -2.3% 0.0% -2.3% 0.6% -2.3%

Flintshire -1.2% 0.2% -2.3% 0.2% -0.7% -3.0% -1.1% -2.3% 0.3% -0.7% 1.7%

Gwynedd 2.7% -0.4% 0.6% -1.3% -2.3% 4.0% 1.3 -0.2% 0.0% 0.7% -2.3%

Merthyr Tydfil -6.4% -3.7% -2.3% -0.6% -0.4% -4.3% -3.7% -3.3% -2.3% -2.3% 0.3%

Monmouthshire 2.7% 0.7% -0.6% 1.7% 1.1% 4.0% 1.8% 1.1% 1.6% 1.0% -1.6%

Neath Port Talbot -3.4% -6.6% -9.0% -0.7% -2.3% -2.3% -4.4% -4.1% 0.8% 0.7% 0.5%

Newport -2.3% 0.1% 1.0% 1.6% 0.6% -2.3% 2.0% 3.2% 1.4% 0.4% -2.3%

Pembrokeshire 0.3% 2.6% 0.2% -2.3% 0.7% 1.6% 2.6% 0.2% -2.3%

Powys -8.4% 0.1% 1.8% 2.9% -0.4% -9.0% -1.3% 0.4% -0.1% -0.5% -3.0%

Rhn’da, Cynon Taff -9.5% -4.7% -2.3% -0.6% -1.0% -11.9% -5.5% -1.2% -1.2% -0.5% -0.6%

Swansea -2.9% -1.4% -6.9% -6.6% -7.7% 0.0% -1.4% -4.6% -2.3% -2.3% -3.2%

Torfaen -2.3% -0.1% -1.7% 0.7% -0.4% -0.3% 2.3% -1.2% 0.0% -0.4% -1.4%

Vale of Glamorgan 3.7% 0.9% -1.9% 0.2% 0.2% 1.4%

Wrexham -4.0% 1.2% -2.3% -1.8% -4.0% -4.9% 0.3% -3.3% -1.8% -4.0% -2.3%

Source: Derived from

JRF/University of York Index of Private Rents (Valuations Index)

Table 8.11 Average annual percentage change (discounted) in rent, m

inus inflation byproperty type and local authority, 1997 Q

4 to 2000 Q4

75

8.2.5 Constructing a long-term rent index,1970-1999

One of the factors that has constrained theanalysis of private renting in Wales, andindeed in the rest of the UK, has been theabsence of time series data on rents.Although this omission has been rectified inpart by the establishment of the York rentindex, two major data shortfalls remain:(1) no data series precedes 1996 and sovery little is known about the pattern ofrents in most of the post-war period; (2)no attempt has been made to construct along run index that controls (or‘standardises’) for variation in dwellingattributes. This standardisation issue iscrucial in calculating accurately theunderlying change in rent after removingthe effect of, for example, a greaternumber of large than small propertiesentering the private rented sector overtime. Not to take account of this could belikened to saying that selling oil in largerbarrels would result in the price of oilrising.

Our approach was to select six years ofdata from the Family Expenditure Survey(FES) - 1970, 1975, 1980, 1989, 1995,1999. We calculated average rent in Walesfor these years, standardising for changesin the following attributes: size of

properties in the PRS, and the prevalenceof both central heating and furnishedproperties. A ‘standard’ property refers toa property with the average values of theseattributes across all years. It was ourintention to use every fifth year from 1970,but inconsistencies in the structure of theFES made this difficult for 1985 and 1990.1989 was substituted for 1990, and 1999was substituted for 2000, the data forwhich have not yet been released.

Standardised rent is compared with theaverage recorded rents in each year, asshown in figure 8.4. As one might expect,figure 8.4 shows rising nominal rent bothfor the standardised and recorded rents.However, standardised rent has grown lessrapidly, reflecting the change in the PRSover time towards a higher proportion ofdwellings with central heating and beingfurnished. Standardised rent in real terms(1999 prices) in Wales was considerablyhigher in 1970 than at any other timesince. The steep falls in standardised rentin real terms throughout the 1970s is areflection of high price inflation during thisperiod in the face of stable nominal rentlevels. Real standardised rent thenrecovered throughout the 1980s andstabilised throughout the 1990s at just over£30 per week (1999 prices).


0

10

20

30

40

50

60

70

1970 1975 1980 1989 1995 1999

Average RecordedRent

Standardised Rent

Standardised Rent(1999 prices)

Average W

eekly

Rent (

%)

Figure 8.4 Average recorded rent, standardised rent and real standardised rent in Wales,1970-1999

Source: Derived from the Family Expenditure Survey

The change over time in the impact ofdwelling characteristics cannot be verified,however, using the FES because of itslimited description of dwellingcharacteristics. Thus a further possibleavenue for future research would be tocompare regressions based on moredetailed cross sectional data (such as fromRent Officer records) from different pointsin time to ascertain which of theseexplanations is most valid.


Many interviewees said that they thoughtthat the quality of privately rentedproperties was improving and that thesector as a whole was increasing inprofessionalism. A number of intervieweessaid that this was in part driven byincreasingly confident and choosy tenantswho will not accept poor standards,particularly of cleanliness. Certainly inCardiff, it was thought that privately rentedaccommodation is no longer perceived tobe a second preference housing tenure.However, this is not borne out by thestated tenure preferences by social andprivate renters in the 1998 Welsh HouseCondition Survey, as reported earlier in thischapter.

There were some clear differences ofopinion between local authorityinterviewees, particularly EnvironmentalHealth Officers, and letting agents andlandlords representatives. The majority oflocal authority interviewees thought thatstate of repair and conditions generally inthe private rented sector are deterioratingwhile letting agents and landlordrepresentatives thought they areimproving. Section 8.2.2 of this chapterreported the results of the 1993 and 1998Welsh House Condition Surveys which

show that physical conditions in the privaterented sector are actually improving.However, this aggregate improvementacross the sector in Wales as a whole maymask falling standards towards the lowerend of the market which are more thanoffset by rising standards in the upper endof the private rental market. Consistentwith this interpretation, some letting agentsand the landlord representatives reportedthat landlords’ ability to carry out repairswas being curtailed by the restrictions onthe level of Housing Benefit payable whichwere introduced in 1996. However,counter to this, some letting agentsreported that landlords would find itdifficult to let properties if they did notkeep them well maintained, althoughgenerally the letting agents who said thiswere those who were not letting to manyrecipients of Housing Benefit. Thedecreasing availability of housing grants,particularly to private landlords, was alsocited by some letting agents and thelandlord representatives for contributingtowards the poor repair and maintenancerecords of some landlords.

Buy-to-let mortgages have beeninstrumental in contributing towardsincreased supply of properties for privaterent. However, other factors have alsobeen important in certain locations, notablystudents’ parents purchasing properties torent and out-migrants from economicallydepressed areas being unable to sell theirhome therefore decide to place it on theprivate rental market. In Merthyr Tydfil, itwas reported in this context that someformer local authority properties which hadbeen purchased through the Right-To-Buyscheme had now found their way onto theprivate rental market.


76

77


A phenomenon mentioned by several localauthority interviewees is the increasingnumber of people choosing to move fromthe social rented sector to the privaterented sector. Thus, not only are someproperties moving from the social rentedsector to the private rented sector, but thisis also true of some tenants. It wasthought by interviewees that this is likely tobe due to low environmental quality anddrug problems in certain housing estatesdominated by social rentedaccommodation. This phenomenon mayalso be due in part to increasing standardsof quality found in much of the privaterented sector, as shown by the results ofthe 1993 and 1998 Welsh House ConditionSurveys.

Overall, it was thought by a number ofinterviewees that the rapid expansion ofthe PRS in the early 90s was nowstabilising, perceptions borne out by theestimates of the number of privately renteddwellings from the 1986, 1992 and 1997surveys of households carried out in Walesand reported in section 8.2.1 of thischapter. However, this was not universallytrue across Wales, for example in Cardiffthe private rented sector continues toexpand. Part of the reason for theexpansion of the private rented sector inthe early 1990s was thought to be due tonegative equity in the housing market insome locations. It was also thought thatthe prospects for significant futureexpansion of the private rented sector werelimited without the introduction of assuredtype tenancies, as without these, the sectorwill not appeal to older ‘settled’ people andfamilies.

8.4 Future Research

The Welsh Household and Dwelling Surveyis due to be carried out within the nextyear or two. Once the results of this areavailable, obvious research to carry outwould be to assess changes since 1998 infitness, repair costs, and dwelling andtenant characteristics in the private rentedsector as well as housing as a whole. Togain detailed information from both thehousehold interviews and the physicalinspections regarding particular property orhousehold types within the private rentedsector, it may be necessary to increase thenumber of households and properties inthe private rented sector sampled in theforthcoming Welsh Household andDwelling Survey.

Similarly, the results of the 2001 Census ofPopulation when available will allow longterm change since 1991 in the privaterented sector to be assessed. The Censuswould be particularly useful to examine thechange in the scale and distribution of theprivate rented sector, including shareddwellings.

The findings reported in section 8.2.5 ofthis chapter suggest that meaningful resultscould be obtained from applying hedonictechniques to the Family ExpenditureSurvey in order to derive a long term rentindex for Wales and other regions in theUK. If all the years of the FES wereutilized, it would be advisable to combineadjacent years in the construction ofdummies in order to ensure that theestimates of the regional effects werebased on an adequate number ofobservations. Further research could alsoexplore in more detail how the shape andexplanatory power of the hedonic rentschedule has changed over time.


78

8.5 Summary of findings

• The private rented sector in Wales, asin the UK as a whole, was in longterm decline until the late 1980s/early 1990s when there was someexpansion, but the sector hasstabilised in size since the mid 1990s.However, the sector’s share of alldwellings and households in Walescontinues to decline slightly becausethe numbers of dwellings andhouseholds in Wales are rising.

• The private rented sector is overallnot a popular choice of housingtenure, with it being the firstpreference for only a quarter of itstenants. Furthermore, less than 2%of social renters aspire towardsprivate renting.

• The level of unfitness and averagerepair cost per dwelling fellsubstantially in the PRS between1993 and 1998, but this laggedsignificantly behind even moredramatic improvements in otherhousing tenures.

• The size and type of dwellings foundin the PRS converged towards thosefound in housing as a whole in Walesbetween 1992 and 1997, butimportant differences remain.

• The size and type of households inthe PRS also showed someconvergence with households as awhole, but the PRS’s below averageshare of single pensioners fell furtherrelative to the population as a whole.

• Private rents in Wales are lower thanthe rest of Britain, and they divergedfurther throughout the late 1990s,and particularly so during 2000. Thisis due to stagnant nominal rent in

Wales from the end of 1997 and fallsin 2000, in the face of increaseselsewhere.

• Yields in Wales, however, are higherthan the rest of Britain due to lowerhouse prices. Yields in Wales andBritain were both on downwardtrends over the period 1996-2000.

• Recent changes in rent levels showgreat variation within Walesaccording to the type of property andgeographical location. No particularlystrong patterns emerged from thisanalysis.

• In the long run, average nominalrents in Wales were stagnant in the1970s, rose steeply in the 1980s androse at a slower rate in the 1990s.However, around half of this increasecan be accounted for by the fact thatthe properties are becoming largerand have better amenities rather thanbeing due to any underlying changein market conditions. Once the effectof inflation is also removed, realstandardised rents in Wales fellsharply in the 1970s, recoveredslightly in the 1980s and stabilisedthroughout the 1990s. This trend inreal standardised rent mirrors thetrend in the total number of privatelyrented properties in Wales over thisperiod.

• There is an increasingprofessionalisation of the sector.

• Local authority officers, particularlyEnvironmental Health Officers,perceive the PRS to be deterioratingin terms of unfitness and disrepair,while letting agents and landlordrepresentatives perceive it to beimproving. While the Welsh House

Condition Surveys objectively showedthat the sector is dramaticallyimproving in aggregate, this maymask important deviations from thistrend at the bottom end of themarket.

• Buy-to-let mortgages have beeninstrumental in facilitating theexpansion of the private rented sectorthroughout the early 1990s, butmany parts of Wales have reachedmarket saturation. Therefore, furthermeasures to expand the supply ofproperties for private rent may needto be entered into with caution.

• Although the majority of existingtenants are reported to be happy withassured shorthold tenancies, toexpand the demand for private

renting in the future from groups atpresent under represented in thetenure, greater security of tenurewould be required.

• Part of the reason for increasedsupply of properties for rent in theearly 1990s was negative equity inthe mortgaged owner occupied sectormeaning home owners were unwillingor unable to sell when they vacatedtheir home, for example to move toanother part of the country.

• Although the numbers of tenantsinvolved are relatively small, manyinterviewees noted an increasingtrend in people choosing to movefrom social rented accommodation tothe private rented sector.

79



80

81


Chapter Nine Future Research and Monitoring

9.1 Introduction

A key output specified by the NationalAssembly for Wales was recommendationson future research and on "which aspectsof the sector should be regularly monitoredat local authority and all-Wales level andhow this might be achieved" (NAW, 2001,p.3). This chapter therefore drawstogether the ideas for future researchwhich have been outlined at the end ofeach empirical chapter. Specificsuggestions for the forthcoming WelshHousehold and Dwelling Survey are thenmade. Finally, some thoughts on how theprivate rented sector in Wales could bemonitored are presented.

9.2 Broad Themes for FutureResearch

Two broad themes for future research intothe private rented sector in Wales unitemany of the specific issues for futureresearch suggested in the foregoingempirical chapters. These two broadthemes are outlined here.

9.2.1 Tenure choice in Wales

Wales has experienced rapid growth inowner occupancy over the past twodecades, and simultaneous shifts awayfrom both private and social renting. If theprivate rented sector is to be revived, policymakers need to understand the key factorsthat drive tenure choice and the sensitivityof the home ownership decision to thesefactors. On a more detailed level, thefactors which influence the choice betweenprivate and social renting could also beconsidered.

Quantifying this process requires anapplication of tenure choice analysis: astatistical examination of tenure decisiongrounded in economic theory. Although alarge number of such studies have beencarried out in the US, relatively few havebeen estimated in the UK context and nonespecifically for Wales. It may be, however,that peculiarities of the Welsh housingmarket and its socio-economiccomposition, limit the applicability ofstudies carried out elsewhere, and point tothe need for a tenure choice analysisspecifically for Wales. A possible datasource for such an analysis is the BritishHousehold Panel Study, which now has abooster sample for Wales and Scotland.The forthcoming Welsh Household andDwelling Survey offers an opportunity togather information on people’s previoushousing and residential mobility whichwould also facilitate a tenure choiceanalysis for Wales.

9.2.2 The dynamics of the residentialproperty market in Wales

The growth of home ownership was notfacilitated by a massive increase in newconstruction but by the transfer ofdwellings between tenures through privatelandlords selling properties, and counciltenants taking advantage of Right-to-Buy.The dynamic of this tenure switchingprocess holds vital clues to the operation ofthe housing market in Wales. Work hasbeen done on England and is currentlybeing done for Scotland to develop aformal analysis of this process. Usingregression analysis, estimates are producedof (amongst other things) the sensitivity of

the probability that a dwelling will transferto owner occupancy with regard toaverage rent. Such techniques require alongitudinal dataset of dwellings (such asthat contained in the Scottish HouseCondition Surveys), which is currently notavailable for Wales (although this was donein Wales between the 1981 and 1986surveys). However, the National Assemblymay wish to bear in mind the advantagesof developing a longitudinal element to thedata set when the forthcoming WelshHousehold and Dwelling Survey is carriedout.

9.3 Specific Issues for FutureResearch

A number of specific issues for futureresearch were identified throughout thisreport. They are summarised here.

Scale and distribution of the private rentedsector

• Reasons for migration from the socialrented sector to the private rentedsector.

• Scale and distribution of differenttypes of person, household anddwelling in the PRS using the 2001Census of Population.

• Use the 2001 Census of Population asa benchmark against which tocompare HMO databases andregistration schemes held by localauthorities.

• Use the 2001 Census of Population asa benchmark against which to assessthe degree of market penetrationachieved by Rent Officers’ marketevidence databases.

Dwelling characteristics

• HMO management, including repairsand maintenance.

• The impact of Housing Benefitrestrictions on repair andmaintenance.

Tenant characteristics

• Qualitative interviews with tenantsregarding their housing aspirations ingeneral and their experiences of thePRS and Housing Benefit.

The role of Housing Benefit in supporting thePRS

• The impact of Housing Benefitrestrictions on private landlords andtenants.

• The impact of Housing Benefitadministration on private landlordsand tenants.

• Evaluation of pre-tenancydeterminations.

Landlords, housing grants and tenancy types

• Assessment of why people becomelandlords and how they decide onwhat type and size of property topurchase for the purposes of letting.

• Review of housing grants, especiallyregarding their benefit and the equityissues surrounding whether landlordor tenant benefits the most, and theflexibility local authorities have inadministering housing grants.

• Review of Bond Boards, includinglandlord, letting agent and tenantperspectives.

82


• Assessment of means to increasetrust, communication and co-operation between landlords and localauthorities, particularly HousingBenefit and Environmental Healthdepartments.

Changes to the private rented sector

• Change in unfitness, repair costs, anddwelling and tenant characteristicsfrom the 1998 Welsh HouseCondition Survey and theforthcoming Welsh Household andDwelling Survey.

• 1991-2001 change in scale anddistribution of the private rentedsector from the Census of Population.

• Calculate a robust long-termstandardised rent index for Wales andother regions of the UK using theFamily Expenditure Survey.

9.4 The forthcoming WelshHousehold and Dwelling Survey

Three issues to consider in the design ofthe forthcoming Welsh Household andDwelling Survey are suggested here. Theyare: boost the private rented sector sample;introduce a longitudinal element from the1998 surveys; and additional questions.

9.4.1 Boost the private rented sectorsample.

The private rented sector plays variousroles in the overall housing system in Walesand the UK, the importance of which faroutweigh its numerical size. In addition,problems in terms of unfitness and disrepairare more prevalent in the PRS than in otherhousing tenures. Furthermore, the PRS isnot a homogenous group of properties ortenants, with a number of sub-sectors, for

example, ‘DSS lets’, the student marketand luxury rooms and houses.

If detailed information is required onvarious aspects of the private rented sectorat local authority level, then it may beadvantageous to boost the number of PRShouseholds and dwellings in theforthcoming Welsh Household andDwelling Survey. Possible means ofachieving this include:

1) Take a longitudinal sample of all the1998 PRS dwellings and repeat themin addition to those PRS dwellingsidentified by random sampling ofaddresses. This would, however,introduce a bias in favour ofproperties which have been in thePRS for at least five years.

2) Random sampling of addresses heldby the Rent Officer Service (ROS)around Wales - biases uncertain.Permission may need to be grantedfor this by the Data ProtectionCommissioner. This would also be ofinterest to the Rent Officers as theycould compare randomly selected PRSproperties with the properties on theirmarket evidence databases. The ROScould perhaps contribute to themarginal cost of surveying propertiesfrom their records.

3) Some or all local authorities may beprepared to pay for a PRS boost intheir area. Some would be able toprovide lists of HMOs, for example,from Environmental HealthDepartments. Again, permission mayneed to be granted for this by theData Protection Commissioner.

The biases introduced by these means ofboosting the PRS sample could be assessed

83


by comparison against the 2001 Census.Appropriate weights could then becalculated for grossing to reflect thehousing stock and population as whole.Because the forthcoming Welsh Householdand Dwelling Survey is likely to be close toa Census year, reasonably accurate weightscould be calculated to adjust for stratifiedsampling such as that outlined above.

9.4.2 Introduce a longitudinal elementfrom the 1998 survey

As outlined in the previous section, it maybe possible to take a longitudinal sample ofthe 1998 PRS dwellings and repeat them inaddition to those PRS identified by randomsampling. Not only would this boost thePRS sample in the Welsh Household andDwelling Survey, but it would also producea longitudinal dataset to analyse change inindividual properties. This longitudinalelement could usefully be introduced for alltenures which would allow, amongst otherthings, the type of analysis outlinedpreviously of the dynamics of the Welshresidential property market.

There may be ethical and Data Protectionissues to consider in contacting householdsfrom the 1998 survey as they may havebeen given assurances that they would notbe contacted again. If this precludessampling from the previous survey, then aprovision to allow this to be done in thefuture from the Welsh Household andDwelling Survey sample could beintroduced.

9.4.3 Additional questions

The 1998 Welsh House Condition Surveywas fairly short compared to many otherface-to-face household surveys carried outin the UK. No doubt there are good

reasons for this, not least cost. However,we suggest two areas for possibleexpansion.

1) It would be useful to ask the amountof Housing Benefit received. Thiswould allow, for example, a detailedanalysis of the importance of HousingBenefit in supporting the privaterented sector, including different sub-markets within the PRS such asHMOs. In addition, the shortfallsdifferent groups of people facebetween payable rent and HousingBenefit received could be assessed.

2) It would be of benefit to includequestions regarding recent andanticipated residential mobility,including reasons for moving.Information on recent residentialmobility could be used to calibrate adynamic tenure choice model. Thiswould help shed light on, forexample, the reasons for the possibleincreasing movement from the socialrented sector into the private rentedsector identified by a number ofinterviewees during the course of thequalitative research for this project. Itwould also shed light on the reasonsthat people choose to live, or choosenot to live, in privately rentedaccommodation in general.

9.5 Possible Monitoring of the PRS

The Rent Officer Service collects largequantities of detailed data about theprivate rented sector across Wales, both incollecting market evidence from non-HBlets and in making rent determinations forproperties let, or to be let, to a HBrecipient. This is a potentially valuableresource which, if stored centrally for all ofWales, could increase knowledge about


84

85


changes through time on an on-going basisregarding the scale, distribution, rent andcharacteristics of the private rented sectoracross Wales.

Given that the structure of the Rent OfficerService in Wales is currently under reviewby the National Assembly for Wales, thismay represent an opportunity tostandardise the way in which data arecollected and recorded by the Rent OfficerService. The Rent Agency in England, weunderstand, are moving towards thecreation of an all-England rent database,from which no doubt lessons could belearned.

There have been some calls for an annualor bi-annual survey of housing andhouseholds in Wales along similar lines tothe Survey of English Housing (forexample, see Smith, Stirling and Williams,2000). This is a potentially useful source ofinformation regarding housing andhouseholds in Wales which could be usedfor on-going monitoring, although the sizeof a survey of this frequency would besuch that it is unlikely that it would bepossible to obtain robust detailedinformation about the private rented sector,certainly below the all-Wales level.


86

87

10.1 Introduction

This final chapter summarises the keyfindings of this review as well as areasidentified for future research andmonitoring. This is not intended to be anexhaustive account, but rather to highlightthe most important issues.

10.2 Key Findings

The key research findings of this review areas follows:


• The private rented sector in Wales hasbeen in long term decline until the1980s and early 1990s when itexpanded slightly but it has stabilisedin size since the mid 1990s.

• The share of households in theprivate rented sector is greater innorth and mid Wales than elsewherein Wales.

• The share of households in theprivate rented sector is lower in southWales and industrial towns thanelsewhere in Wales. However, Cardiffand Swansea have around the Welshaverage proportion of householdsliving in the PRS.


• Dwellings in the private rented sectorare smaller compared to otherhousing tenures.

• There is a greater prevalence of flatsin the private rented sector thanhousing as a whole in Wales.

• Dwellings in the private rented sectorare on average considerably olderthan in other tenures.

• Dwellings in the private rented sectorhave a higher prevalence of unfitnessand higher repair costs than housingin other tenures.


• private tenants are substantiallyyounger and lower income than thepopulation as a whole,

• single people, both pre- and post-marriage, are more likely to be privatetenants (apart from pensioners),

• private tenants are more likely to beunemployed and members of ethnicminority groups compared to thepopulation as a whole,

• landlords and letting agents generallyrecognise the following marketsegments:

- professionals and other mobileworkers,

- young single people and couples,including families in rural areas,

- students, and

- ‘DSS lets’ (i.e. recipients of HousingBenefit), differentiating betweenlong-term and more transient sub-groups.

• landlords representatives and lettingagents reported problems at the


Chapter TenConclusions

bottom end of the ‘DSS’ market ofrent arrears and tenants vacatingproperties without serving notice toquit.

The role of Housing Benefit in supporting theprivate rented sector

• Restrictions on Housing Benefit arereported by some landlords’ to limittheir ability to carry out repairs at thebottom end of the private rentalmarket, although this is sometimesreported to also be the case withmarket rents in some sub-sectors.

• Restrictions on Housing Benefit areresulting in increasing numbers oftenants facing shortfalls between theirrent and that which Housing Benefitwill cover.

• Knowledge of the Housing Benefitsystem is variable, with some tenants,landlords and letting agents beingunsure of how it operates.


• Buy-to-let mortgages have facilitatedan expansion of individual investmentlandlords throughout the early 1990sin particular.

• There has been an increasedprofessionalisation of the PRS in thelast decade or so, with reportedincreasing use of management andletting agents.

• There is over supply or marketsaturation in the private rentalmarkets in some locations.

• Few landlords or letting agents seeany benefit to them of using BondBoards.

• Although the majority of existingtenants are reported to be happy withassured shorthold tenancies, toexpand the demand for privaterenting in the future from groups atpresent under represented in thetenure, greater security of tenurewould be required.


• There is some evidence that theprivate rented sector in Wales isconverging with the rest of thehousing stock in Wales in terms ofdwelling and tenant characteristics.

• Unfitness and repair costs in the PRSare improving considerably, but thismay be less true at the bottom end ofthe market. Furthermore, theimprovement in the PRS is slowerthan that in other housing tenuresand the condition of PRS propertiesremains significantly below that foundin housing as a whole in Wales.

• Rents are rising in Wales slower thanin the rest of Britain, particularly in2000. Furthermore, rents in Walesare lower than in the rest of Britain.

• Yields are higher in Wales than therest of Britain due to lower houseprices. Yields in both Wales andBritain are falling.

• Although nominal rents are rising inWales, after taking account ofinflation and the fact that larger andbetter quality properties are enteringthe PRS, underlying rent in Wales fellin the 1970s, recovered slightly in the1980s and has been stablethroughout the 1990s.

• Although the numbers of tenantsinvolved are relatively small, many


88

89

interviewees noted an increasingtrend in people choosing to movefrom social rented accommodation tothe private rented sector.

10.3 Key areas for future research andmonitoring

Some of the areas for future research listedbelow relate to issues, particularly HousingBenefit, which are to a large extent outwith the direct control of the NationalAssembly for Wales. However, theNational Assembly may wish tocommunicate with Westminster centralgovernment departments over the directionof UK research in these areas.

Two general areas for further research havebeen identified:

• Tenure choice in Wales.

• The dynamics of the residentialproperty market in Wales.

Specific areas for future research identifiedby this review include:


• Reasons for migration from the socialrented sector to the private rentedsector.

• Use the 2001 Census of Population asa benchmark against which tocompare HMO databases andregistration schemes held by localauthorities.

• Use the 2001 Census of Population asa benchmark against which to assessthe degree of market penetrationachieved by Rent Officers’ marketevidence databases.


• The impact of Housing Benefitrestrictions on repair andmaintenance.


• Qualitative interviews with tenantsregarding their housing aspirations ingeneral and their experiences of thePRS and Housing Benefit.

The role of Housing Benefit in supporting thePRS

• The impact of Housing Benefitrestrictions on private landlords andtenants.


• Review of Bond Boards, includinglandlord, letting agent and tenantperspectives.

• Assessment of means to increasetrust, communication and co-operation between landlords and localauthorities, particularly HousingBenefit and Environmental Healthdepartments.


• Change in unfitness, repair costs, anddwelling and tenant characteristicsfrom the 1998 Welsh HouseCondition Survey and theforthcoming Welsh Household andDwelling Survey.

• 1991-2001 change in scale anddistribution of the private rentedsector from the Census of Population.


In addition, we suggest the following forthe forthcoming Welsh Household andDwelling Survey:

• Boost of the private rented sectorsample.

• Introduce a longitudinal element fromthe 1998 survey.

• Ask additional questions regarding a)the level of Housing Benefit received,and b) past and anticipated residentialmobility.

Finally, we suggest that the proposedrestructuring of the Rent Officer Service inWales be used as an opportunity tocentralise and standardise the datacollected both for valuations and marketevidence.


90

91

Bevan, M, Kemp, P A, & Rhodes, D (1995) Private Landlords and Housing Benefit, Centre forHousing Policy Research Report, York: University of York.

Bourton, Wendy and Leather, Phillip (2000) Housing stock condition and renewal. In Smith,Robert; Stirling, Tamsin; and Williams, Peter (Eds) Housing in Wales. The Policy Agenda inan Era of Devolution. (Coventry: Chartered Institute of Housing).

Bramley, G., Pryce, G. & Satsangi, M. (1999) ‘The Supply Responsiveness of the PrivateRented Sector: An International Comparison’, Department of the Environment, Transport andthe Regions: London.

CH Associates (2000) Who cares about older private tenants? Comparing the housingexperiences of older private tenants and their landlords with those of owner occupiers in twoWelsh rural areas - Carmarthenshire and Pembrokeshire. A study for the Housing AssociationCharitable Trust, September 2000.

Coles, A. (1989) The English and Welsh House Condition Surveys 1986. Housing Finance,no.1.

Crook, T & Kemp, P A (1999) Financial Institutions and Private Rented Housing, York: JosephRowntree Foundation & York Publishing Services.

Crook, T, Hughes, J & Kemp, P A (1995) The Supply of Privately Rented Homes, York: JosephRowntree Foundation.

Crook, T, Henneberry, J, Hughes, J, & Kemp, P A (2000) Repair and Maintenance by PrivateLandlords, London: Department of the Environment, Transport and the Regions.

Daniel, Lisa (1999) Gaining and maintaining private rented accommodation for young singlehomeless people. M.Sc. Dissertation, Department of City and Regional Planning, Universityof Cardiff.

Fisk, Malcolm, J. (2000) Historical perspectives on housing development. In Smith, Robert;Stirling, Tamsin; and Williams, Peter (Eds) Housing in Wales. The Policy Agenda in an Era ofDevolution. (Coventry: Chartered Institute of Housing).

Holmans, A. (1996) Housing demand and need in Wales 1991-2011. JRF Findings summaryno.182.

Institute of Welsh Affairs (1997) Building our future: the housing challenge for Wales.Cardiff, Institute of Welsh Affairs.

Kemp, P A & McLaverty, P (1995) Private Tenants and Restrictions in Rent for HousingBenefit, Centre for Housing Policy Research Report, York: University of York.

Kemp, P A & Rugg, J (1998) The Single Room Rent: Its Impact on Young People, Centre forHousing Policy Research Report, York: University of York.

Maclennan, D. & Pryce, G. (1996) ‘Global Economic Change, Labour Market Adjustment,and the Challenges for Housing Policies’, Urban Studies, 33, 10, 1849-1865.

Murphy, D (1998) Of No Benefit: The Effects of the Single Room Rent in Wales, ShelterCymru.


References

NACAB (1998) Unsafe Deposit: CAB Clients’ Experience of Rental Deposits, London: NationalAssociation of Citizens Advice Bureaux.

National Assembly for Wales (2000a) Better Homes for People in Wales. The NationalAssembly for Wales’ proposals for a National Housing Strategy. Consultation Paper. Cardiff,National Assembly for Wales, December 2000.

National Assembly for Wales (2000b) Proposals for a National Housing Strategy for Wales.Final reports to the National Assembly for Wales by the Housing Strategy Task Groups.Cardiff, National Assembly for Wales, April 2000.

National Assembly for Wales (2001) Better Homes for People in Wales. A National HousingStrategy for Wales. Mimeograph.

Rugg, J (1997) Closing Doors?, Centre for Housing Policy Research Report, York: University ofYork.

Rugg, J (2000) The Nature and Impact of Student Demand on Housing Markets, York: YorkPublishing Services.

Shelter Cymru (1994) No way in. The availability of the private rented sector to families inAlyn and Deeside. Swansea, Shelter Cymru.

Smith, Robert; Stirling, Tamsin; and Williams, Peter (2000) Future directions: an agenda forWelsh housing policy. In Smith, Robert; Stirling, Tamsin; and Williams, Peter (Eds) Housingin Wales. The Policy Agenda in an Era of Devolution. (Coventry: Chartered Institute ofHousing).

Task Group 2 (2000) Final Report to the National Assembly for Wales. Creating Efficient andAccessible Private Housing Markets with Good Quality Accommodation, Cardiff: NationalAssembly for Wales. www.wales.gov.uk

Tewdwr-Jones, Mark and Gallent, Nick (2000) Housing in rural Wales. In Smith, Robert;Stirling, Tamsin; and Williams, Peter (Eds) Housing in Wales. The Policy Agenda in an Era ofDevolution. (Coventry: Chartered Institute of Housing).

Thomas, Rhian (1995) Access to the private rented sector for low income people. M.Sc.Dissertation, Department of City and Regional Planning, University of Cardiff.

University of York (2000) University of York Index of Private Rents and Yields. FourthQuarter 2000. (York: Centre for Housing Policy, University of York).

Welsh Consumer Council (1987) Consumer Issues in Wales. A report of suggestions forwork received by the Welsh Consumer Council for 1987/8. Consumer Issues in Wales,section 1: housing. (Cardiff: Welsh Consumer Council).

Williams, Peter (2000a) Home-ownership and the private housing market. In Smith,Robert; Stirling, Tamsin; and Williams, Peter (Eds) Housing in Wales. The Policy Agenda inan Era of Devolution. (Coventry: Chartered Institute of Housing).

Williams, Peter (2000b) Housing Finance National Markets Review - Wales. NationalHousing Finance National Markets Review 2000, p.31-40. (London: Council of MortgageLenders Publications).


92

93

This appendix provides a brief description of all the datasets reviewed in one place. Thedata review sections at the start of each empirical chapter in the main body of the reportonly cover material relevant to each particular chapter.

The following datasets were reviewed:

1) Census of Population, Local Base Statistics & Small Area Statistics, 1991 and 20012) DSS/DWP Housing Benefit data, available annually from 19923) Family Expenditure Survey, available annually from 19614) General Household Survey, available annually from 19715) JRF/Uni of York Index of Private Rents & Yields, available quarterly from 1996Q16) Student Accommodation data, 1998/997) Welsh Health Survey, 19988) Welsh House Condition Survey, 19989) Welsh Index of Multiple Deprivation, 2000

1) Census of Population

Year(s): Decennial, 2001 not yet available.Sample: Full population coverage; some more complex tables are 10% samples.

Census data are mostly based on households or persons, but selected tables based ondwellings. Given near 100% coverage of the population, this can provide information onthe precise spatial distribution of households in the private rented sector in Wales down to afine grained geography using the Local Base Statistics (LBS) and the Small Area Statistics(SAS). In the 1991 Census, data are available at the levels of Wales, County, District, Wardand Enumeration District (ED). In the 2001 Census they will be available at Wales, localauthority, Ward and ED levels.

The 1991 Census provides information by housing tenure on the number of rooms indwellings, dwelling type (detached house, terraced house, flat, etc.) and amenities (e.g.central heating and shared WC facilities). Information on household size, householdstructure, overcrowding and car ownership are also provided by housing tenure. Housingtenure in the Census differentiates between furnished and unfurnished private rentedproperties and accommodation that is rented from the tenant’s employer. These data referto households rather than dwellings.

The 2001 Census, which will not be available for some time, contains similar information.The 2001 Census includes a new question on the general state of individuals’ health,hopefully which will be made available in the 2001 SAS tabulated against housing tenure.Unfortunately, proposals to include a question on income were not implemented.


Appendix 1Data Review

2) DSS/DWP Housing Benefit Data

Year(s): From 1992Sample: 100% of HB caseload.

The Department for Work and Pensions’ Housing Benefit Management Information Systemis based on returns by local authorities on 100% of their HB caseload. It contains annualdata on the number of regulated and deregulated private tenants in receipt of HB in Walesand in each local authority from 1992 to 2000. Its also contains, for Wales and for eachlocal authority, the average eligible rent and average HB entitlement of regulated andderegulated private tenants from 1996 to 2000. It is possible to make estimates of theannual cost of HB for Wales and for each local authority, based on the data for the numberof recipients and average HB entitlement. We propose to include these estimates, togetherwith an overview of the HB caseload and average payments, in the final report. Because oflocal government reorganisation in Wales in 1996, data presented will be for all of Walesfrom 1992 and by local authority from 1996.

3) Family Expenditure Survey

Year(s): From 1961Sample: Approximately 8,000 households annually across the UK

The Family Expenditure Survey (FES) is a UK wide on-going survey which collectsinformation on approximately 8,000 households each year. The FES is available from 1961.In any year, the FES generates around 40 private rented sector observations in Wales.Therefore, analysis could only be carried out at all-Wales level, and adjacent years wouldneed to be combined to generate robust sample sizes.

Since the focus of the FES is on expenditure, it includes levels of rent and rent rebates(presently Housing Benefit). The FES also provides information on the size of properties,whether they have central heating or not, whether rented properties are furnished or not,and where properties are located. The geographical indicators contained in this dataset areStandard Regions.

4) General Household Survey

Year(s): From 1971Sample: Approximately 8,000 households annually across the UK

The General Household Survey (GHS) is available annually from 1971 and providesinformation on housing, household and individual personal characteristics. It has informationon approximately 8,000 households across the UK, so faces the same sampling constraintswith regard to the private rented sector in Wales as the FES. The geographical indicatorscontained in this dataset are Standard Regions.


94

95

The GHS also contains information on the type of landlord people rent from. This could beused to provide an insight into the long term changing type of landlord operating in theprivate rented sector, for example employers offering accommodation tied to a job, otherinstitutional landlords and private individuals. Long term change in the size and type ofproperties in the private rented sector could also be assessed.

5) JRF/ University of York Index of Private Rents and Yields

Year(s): From 1996, quarterlySample: N/A (Valuations Index); 1,000+ transactions per quarter (Transactions Index)

The University of York produces two indices (formerly sponsored by the Joseph RowntreeFoundation). The first is the Valuations Index which is based on Rent Officers’ valuations ofthe typical rent for six different types of common private rented properties, each with aprescribed set of attributes. This is done at the local authority district level and isstandardised to take account of the type of properties that are more likely to change tenantstherefore be included in a Rent Officer’s valuation. An adjustment is also made for theproperty type mix in each district.

The second index produced by the University of York, the Transactions Index, is based, asthe name suggests, on actual transactions in the private rented sector. Data on these areprovided by a sample of members of the ARLA and other principle letting agents. Note thatthis is biased towards the middle and upper end of the private rental market. TheTransactions Index is available at district level, and occasionally to smaller postally basedunits when a sufficient number of transactions are present in order to provide a reliableestimate of market rent. This index is standardised through time to allow for the fact thatdifferent segments of the market will have a higher proportion of all transactions at differentpoints in time e.g. students taking up private tenancies in September/October, and longerscale variations with the economic cycle. As with the Valuations Index, an adjustment is alsomade for the property type mix in each district.

The York Indices provide robust time series data since the first quarter of 1996 on the rentcharged for different types of property and the yield returned to landlords. The yield isannual rent as a proportion of the open market vacant repossession value of the property.In the Valuations Index, this is based on valuations of the same properties as the rentalvaluations have been made on. However, in the Transactions Index, property values aretaken from Halifax Plc sale prices, therefore rental yields in the Transactions Index areestimates.

Note that the University of York Indices may be discontinued in the near future.


6) Student Accommodation Data

Year(s): Available for academic years/ sessions (October to September).Sample: All student enrolments.

These data are recorded by the Statistical Directorate of the National Assembly for Wales.They are collected for all students (full and part time; HE and FE) for each academic year. Itis not clear from the data we have been given whether it is possible to differentiate betweenfull and part time students.

Each student’s home postcode is recorded, but not the postcode of where they live duringterm time. The identity of the institution at which the student is enrolled is recorded, as isthe local authority code in which the institution is located, for both HE and FE students.

The type of accommodation in which the student lives during term time is recorded as oneof the following: ‘institution maintained property’; ‘parental/guardian home’; ‘own home’;‘other’; ‘not known’; and ‘not in attendance at institution’. Unfortunately, this does notidentify students in the private rented sector, as ‘own home’ includes students living inowner occupation, the PRS and the SRS. However, these data do allow for the proportionof students who live with their parents/guardians to be calculated, which would give anindication of the affordability of accommodation to students generally. This would need tobe restricted to students with a parental home within a certain distance of the institution inquestion in order to control for variation in the recruitment catchments of institutions. Thisproportion could be broken down by, for example: institution, local authority area, andlocation of parental home.

7) Welsh Health Survey

Year(s): 1998Sample: 29,874 individuals

The 1998 Welsh Health Survey (WHS) provides information on a random selection of30,000 individuals. The only non-health related information it contains that is not containedwithin the Welsh House Condition Survey is ethnic group. No household information iscollected (e.g. household size) and information on housing characteristics is limited to tenureand the age of dwelling. The WHS and the Welsh House Condition Survey could, in theory,be combined to provide a larger dataset but the weights used in each survey to factor-upthe results to reflect the age, gender and geographical distribution of population in Waleswould need to be modified. In any case, this would only enhance the data common to bothdatasets, i.e. age of dwelling and individuals’ characteristics (age, gender, marital status, etc.)but nothing on household characteristics.


96

97

8) Welsh House Condition Survey (WHCS)

Year(s): 1998Sample: 40,670 household interviews; 12,037 physical inspections.

Every five years or so a survey of house conditions is carried out in Wales. Surveys havebeen carried out in 1968, 1973, 1976, 1981, 1986, 1993 and 1998, and another is plannedto take place within the next year or two. The 1998 WHCS consists of household interviewsfollowed by physical inspections of a selection of dwellings from the household survey. Thehousehold interview element was carried out in 1997 and is sometimes referred to as the1997 Welsh Household Interview Survey. The physical survey was carried out in 1998.

The 1998 Welsh House Condition Survey physical inspection is well documented by theNational Assembly for Wales and local authorities across Wales. Therefore, it was notreviewed in detail in this study. Rather, the research team concentrated on the householdinterview component. The National Assembly for Wales made the 1998 Welsh HouseCondition Survey available to the research team in electronic format and provided publishedreports of the 1986 and 1993 surveys. Note that the earlier surveys were based onsubstantially smaller samples therefore are less robust at local authority level.

The 1998 Welsh House Condition Survey includes information on:

• geographical location at various scales (for example, regions within Wales, localauthorities, parliamentary constituencies, postcode areas e.g. CF and districts e.g. CF1),

• the characteristics of properties (e.g. date of construction, number of rooms, type ofdwelling, level of insulation, presence of central heating),

• the respondents’ and the interviewers’ assessments of the condition of the properties• the characteristics of households and individuals (for example household size,

household income, and individuals’ age and socio-economic status),• the amount of rent paid,• whether Housing Benefit and other state benefits are received or not,• whether a housing grant has been received to improve the property or not,• the type of landlord (this is broadly: tied, institution or individual).

The Welsh House Condition Survey was designed to provide robust information down tolocal authority level for most variables. Therefore, a reliable estimate of the housing tenurestructure of different local authority areas can be made, but further disaggregations withinthe private rented sector at local authority level will result in small numbers of observationsin some cells.


9) 2000 Welsh Index of Multiple Deprivation

Year(s): 2000 (some data sourced in earlier years)Sample: N/A - various secondary sources used

This is at electoral division level (of which there are 856 in Wales). The Index contains thefollowing domains:

• income (based on the proportion of the population in receipt of various social securitybenefits),

• employment,• health and disability,• education, skills and training,• housing (based on the proportion of housing in disrepair, without central heating and

lacking roof/loft insulation),• geographical access to services,• crime and physical environment,• overall index.

The income domain within the Welsh Index of Multiple Deprivation was used as a criterionin the framework for the selection of case study local authority areas.


98

Documents

Review of the Private Rented Sector in Wales · 2019-06-17 · 8.1 Absolute and relative size of the private rented sector in Wales: 1986, 1992 and 1997 8.2 Tenure aspirations of