Road Transport in Ethiopia: Trends, Stock of Achievements, and Impact on Overall and Sectoral Economic

Growth

By Ibrahim Worku

Hassan

Ethiopian Development Research Institute (EDRI)

January 21, 2011

Outline• Introduction• Objective• Motivation• Literature• Descriptive analysis• Econometric analysis• Findings and conclusion• Recommendation• Caveats

Introduction• Successful implementation poverty alleviation programs requires an efficient

infrastructure facility : road network takes the priority

• Essential for expanding education and health service provision, trade facilitation- better public as well as private services, including banking and insurance services, to the destitute and marginalized rural dweller

• Creates linkages to other modes of transportation like railways, shipping, and airways

• Creation of market access opportunities for agricultural products

• Fan and Rao (2003), road transport facilities play a role in both the production and consumption decision of every household in their day-to-day activities

• sector is vital for developing countries where advanced means of transportation is expensive

• In sum, it affects overall and sectorial economic growth

Motivation• EDRI recognized the importance of studying the sector• In Ethiopia road transport is the dominant mode and accounts for

90 to 95 percent of motorized inter-urban freight and passenger movements.

• However, because of its limited road network, provision of infrastructure has remained one of the formidable challenges

• Recent trends in pattern of government expenditure on road projects have changed dramatically over the last few years.

• Thus, it is important to study the trends in the levels and composition of government expenditures in the road transport sector

• It is relevant to analyze the contribution on overall growth and sectorial growth

• It contributes to the stock of knowledge regarding the road transport sector.

Objective of the Study • To analyze the stock of achievements and the performance of road

transport in Ethiopia• Review the available reports and policy strategy documents • Identify and characterize the links that exist between road network

development and overall and sectoral growth • To capture the impacts different types of road on the overall and

sectoral economic growth• Come up with recommendation for improved performance of the road

sector development on the overall and sector specific economic growthTo achieve the stated objective: • literature survey, on road sector policy, link b/n road network and

economic growth; method of analysis• Descriptive Analysis and • Econometric Analysis are Undertaken

• Dercon, Gilligan, Hoddinott and Woldehanna (2006): public investment on road leads to lower growth of poverty and improvement in consumption

• Fan and Chan-Kang (2005) indicated that different types of roads can have different economic return

• Fan and Chan-Kang (2005) indicated that there is strong link between road development economic growth and poverty reduction

• Also reasoned out to incorporate interaction term in the model• Khandker et al. (2006), using panel data, estimate the income-consumption

benefits of road investment• Fan et al.( 1999) on linkages between government expenditure on roads lifts

poor section of the society to bypass the poverty line: increase agricultural wages, lowering food prices, increase non-farm employment and higher rural wages also enhance rural income

• ERA studies: on policies, planning, expenditure and financing • Ethiopia has had a respectable growth performance in the post-1991 period• 2004-2009, the economy averaged 10 percent growth rate of GDP• 50 % of GDP agric, 85 percent of the population livelihood, 90% of export

earning • Overall, Agriculture, Industry and Service GDP vis-à-vis road network

Literature Survey

Descriptive analysis: Policies: RSDP During EPRDF, PASDEP and RSDP are introduced• PASDEP consistent with RSDP• RDSP I, II, III• RSDP (1997-2002) focused on the restoration of the road network

to acceptable condition. Specifically the Program focused on a) Rehabilitationb) Upgrading c) Construction d) Regular maintenance the program also considered major policy and institutional reforms.

RSDP• The first five-year of the program (RSDP I) launched, 1997-2002, • physical and financial accomplishment rate of 97% and 74.2%,

respectively. • Accomplishment under RSDP II is rather encouraging. The total

disbursement rate of investment on federal and regional roads is about 125% and 73%,

• whilst the corresponding physical accomplishment is 134% and 145% of the planned.

• the total disbursement of projects planned for the execution amount Birr 25.4 billion (US$ 2.9 billion).

• This would enhance the integration of domestic markets and the potential growth of exports in terms of volume and international competitiveness

• The road network has been increasing

•According to world bank (2010), it is only 10 percent of the rural population who are living two kilometers closer to all weather roads.•70 percent of rural people are far away from all weather roads by more than two kilometers•EPDRF the government has invested much for construction of asphalt road•After 2001 there is significant growth in asphalt road length•In recent years; 2003, 2005, 2008 and 2009, however, there is unhealthy growth in gravel road.

Road Network

Road Density• The proper level of road network is assessed by road density• It is measured by taking road length per 1000 persons or the length of road per

1000 km2

• In the three RSDP periods, there was a plan to increase the road density form 0.46 to 1.5 km per 1000 persons and form 24.14 to 116km per 1000km2

• After the implementation of RSDP I, the road density has reached 0.5 per 1000 persons and by 28.69 per 1000km2.

• That is, at the end of the first phase, the target was achieved • At the expiration of RSDP II, 0.55 per 1000 persons and 38.6 per 1000km2

• End of RSD PIII 0.57 per 1000 persons and 42.6 Km per 1000 Km2 where 1.5 km per 1000 persons and 116 km per 1000km2 were the targeted ones, respectively.

• Despite the increment, the road density has not improved that much as planned• At the current the road density is still much below the average road density of

Africa, which is 60km per 1000 km2

Road Density• International Road Federation Statistical Report (2006) indicated

road density per square kilometer of lower middle-income countries is 0.30 km/Sq. Km.

• Ethiopia’s current road density is only 0.0468 Km/Sq. km of the total land area.

• If we consider 80% of the land area to be populated, the country by now has 0.058 Km/Sq. km of road density.

• Ethiopian Road Network compares poorly even with other low income countries including Sub-Saharan Africa.

• This indicates that a lot has to be done in expanding the road network, which requires about four fold of the existing road network.

Road Accessibility• Theoretically, accessibility has three elements:I. The location of the individual; II. The location of the supply, service or facility to which the individual

needs access; III. The link to bring the two together.• Three approaches to estimate network demand, namely,1. The random model2. The graph theory3. The squire grid approach• 200,000 km of optimum to the lower middle-income country : 200,000

km may be considered as target road network • Will give reasonably good accessibility.• To secure the rural population to have access to all weather road is

estimated to be 0.3 km/Sq. km, middle income road density, in which case a road transport network has to reach 330,000 km

Accessibility • Problem of accessibility is resolved only very sluggishly. With a period of

seven years only, additional eight percent of the population is provided with access to road.

• The recent trend indicates that with the recent construction of new roads, the average distance from a road has been reduced from 21km in 1997 to 11.7 km in 2009.

• Trends over access to all-season road per percentage of rural population

• The future is not unwelcoming. • Requires a serious commitment of the sector offices and other

stakeholders to achieve enhanced access to the rural population.

Road Sector Financing• The share of internal cost of financing by now has come to be more than ten

times what it has been a decade before. • External financing is also growing but is rampant while the domestic share is

consistently rising

Share of internal and external financing in the road sector • Government is making a relentless effort towards the development of the road

network• Key constraint is the availability of sufficient funding• Inadequate funding and resource mobilization to be major problems in the road

sector. • Insufficient public resources lead to under-funding of road transport

infrastructure needs :maintenance needs as well as expansion

Few Points on Community Roads• According to ERA (2008) study, community roads are being developed in different

woredas. • The roads are being constructed as part of Productive Safety Net Project (PSNP) under

“food for work” scheme. • Currently 57,000 km of community roads are constructed in the period 2004-2008. The

roads are constructed

• Development and its sustained maintenance is a key point in ensuring the accessibility/mobility of the rural poor

• Despite its relevance, studies reveal that community roads are not well accounted and managed by either the federal or regional road administration

RegionExisting by

2006 (Km)

Constructed in

2007 (Km)

Total Community

Road Stock (km)

Tigray 5,803 578 6,381

Afar - - -

Amhara 11,980 4,000 15,980

Oromia 13,839 3,489 17,328

Somali 1,273 - 1,273

RegionExisting by

2006 (Km)

Constructed in

2007 (Km)

Total Community

Road Stock (km)

SNNP 10,754 2,027 12,781

Benishangul-Gumz 2,230 1,001 3,231

Gambella 464 - 464

Total 46,640 11,124 57,438

Econometric Modeling and EstimationConceptual Framework of the Model• Literature : infrastructure economics . • Canning and Bennathan (1999), Fan (2002), Fan and Chan-Kang

(2005), Canning and Pedroni (2004)• Dercon, Gilligan, Hoddinott and Woldehanna (2006) • To start with the simplest production function that has the

following specification GDP = f (L, K)• Such specification has to be augmented with road so as to identify

its impact on economic growth GDP = f (L, K, R)• Chan-Kang (2005) criticize previous studies for not recognizing the

fact that return from different type of road on economic growth to be different

GDP = f (L, K, RT) GDP = f (L, K, Rp, Rg) Where Rp and Rg represent paved and gravel road respectively.

• Such specification makes it interesting to see the impact of gravel road on agricultural GDP

Specification (contd.)• The model could now be stated as; • On agriculture

AGDP = f (L, K, RT)

AGDP = f (L, K, Rp, Rg)

• On industry

IGDP = f (L, K, RT)

IGDP = f (L, K, Rp, Rg)

• On service

SGDP = f (L, K, RT)

SGDP = f (L, K, Rp, Rg)

• In order to account for other factors, such as policy shift, in the above functional specification, policy dummies are introduced

• The specification, in general, will be stated as follows:

GDP = f (K, H, Rp, Rg, Dpi)

• Dpi denotes dummy for policy interventions, which are introduced to account for any policy intervention over the analysis period.

Specification of the Model

• An estimable production function, augmented Cobb-Douglas production function with the road component

• Where: Ht is human capital at time t

• Kt is physical capital at time t

gtpttttt RRKHGDP .

tgtptttt khGDP lnlnlnlnln 21

tpgtptttt DkhGDP 54321 lnlnlnlnln

Data • Data span: 1970-2009• MoFED: GDP, Road network: ERA ADI:, Labour CSA: Capital stock• Physical capital is derived from kohle’s (1988), fan and chan-kang (2005)

and similar other studies used the following capital accumulation function

Where kt is capital stock in period t , It is investment and is rate of depreciation

• The initial capital stock is computed

Where r is the real interest rate• The initial stock of capital is computed for the year 1970• For instance, Kohle (1988), taking 8 per cent depreciation rate of the stock

of capital, computed the initial capital stock to be 7.4 times gross capital formation of the beginning year, which is 1978.

• For this study 8 percent depreciation rate is taken. The computed initial capital stock is found to be 5.88 times the gross capital formation of the year 1970.

• Human capital: following Soto (2001), fan and Chan-Kang (2005), Calderón (2009) and others proxied by secondary year school enrolment on growth regression function to see of infrastructure and growth in Africa.

1)1( ttt kIK

rI

k tt

Econometric Analysis :Estimation Time series data• Test for Unit Roots• Haris (1995) stated that; • …models containing non-stationary variable will often lead to a problem of

spurious regression, whereby the results obtained suggest that there is statistically significant relationship between the variables in the regression model when in fact all that is obtained is evidence of contemporaneous correlation rather than meaningful causal relation.

• Phillips-Perron (1988) test. • Pperron test in stata command• Most macroeconomic variables, in practice, are suspected of having time

trends• Graphic Visual Inspection Test on the Existence of Trend is undertaken • Unit root tests require a trend term in both the ADF and pperron tests

• 1%, 5%, and 10% Critical Values at levels are -3.662, -2.964 and -2.614 • 1%, 5%, and 10% Critical Values at levels with trend are -4.260, -3.548 and -3.209 • 1%, 5%, and 10% Critical Values at difference are -3.668, -2.966 and -2.616• Lag length of three is chosen to be the optimal lag length• * Based on the graphics inspection, de-trending is made for variables showing a trend behavior

Statistic Z(t)

at level

Statistic Z(t) at level

with trend*

Statistic Z(t) at first

difference

lnrgdp_pw -0.926 -0.731 -5.701

lnargdp_pw -2.407 -2.167 -6.193

lnsrgdp_pw 1.209 -0.428 -5.729

lnirgdp_pw -1.189 -1.696 -4.887

lncapnet_pw 1.747 -1.335 -4.26

lnhcap_pw -1.813 -2.283 -4.313

lnelect_pw 0.389 -3.035 -7.596

lnroadtot_pw -0.977 -2.693 -6.457

lnroadtotsq_pw -0.425 -2.56 -5.461

lnaspurb_pw -1.63 -1.194 -5.707

lngrvroad_pw -1.599 -1.553 -5.903

lnrdcapnet_pw 1.429 -1.954 -4.016

Phillips-Perron:Unit root test

Engle-Granger: test of cointegration• As the above table indicates, all of the variables are found to be not-stationary in levels

whereas few of the variables are marginally stationary in the case of ADF test, only at 10% significant value

• Regarding the existence of trend component with in a time series data set, • Harris (1995) underlined that trend in a data set can lead to spurious correlation the time

trend can either be removed by regressing the variable on time trend or nullified by including a deterministic time trend

Test for co-integration• Having tested our time-series for stationarity, the next step is checking whether the

linear combination of the variables is also stationary or not• Engle and granger (1987) argue that for such relationships to exist, the error terms of the

model should be stationary• Applied the engle-granger procedure to test for co-integration. When variable x is said to

granger-cause variable y if, given the past values of y, past values of x are useful for predicting y in two stages

• The Engle-Granger two step estimates indicate that the variables are co-integrating at 10% significant level. The Durbin-Watson test statistic also shows the existence of serial correlation in both models. Therefore, it is possible for us to estimate the model using ECM.

Co-integration • The Engle-Granger two step estimates indicate

that the variables are co-integrating at 10% significant level.

• The Durbin-Watson test statistic also shows the existence of serial correlation in both models.

• Possible for us to estimate the model using ECM. • But the problem with OLS is that it fails to account

for the problem of endogeneity• In addition, it is also difficult to deal with a model

of variable with a suspect of heteroskedasticity of the error terms.

• Interaction term : multicollinearity

Method of Estimation :OLS, IV, 1st DIff-GMM, 2-Step-GMM

• When the above problems are persistent in a time series data set, the best way to deal with is to opt for generalized method of estimation (GMM) technique

• The recently devised two-step GMM estimator would be an ideal tool to deal with multiple time series data with endogenous variable and as well is suspected of having heteroskedasticity

• Baum et al. (2007) unlike instrumental variable or two-stage least square estimators, GMM estimator does not require additional assumption on the error terms.

• Iv requires homoskedastic assumption and independent error terms.

• Without stating those assumptions one can estimate a model efficiently and consistently using two-step GMM estimation technique

Estimation (contd.)• Lanne and Saikkonen, (2009) highlight the problem in using the first-differenced

GMM panel data estimator to estimate cross-country growth regressions, time series data set.

• When the time series are persistent, the first-differenced GMM estimator can be poorly behaved, since lagged levels of the series provide only weak instruments for subsequent first-differences.

• Revisiting the work of, Caselli, Esquivel and Lefort (1996), the authors show that this problem may be serious in practice

• the second and third lags and lagged differences are used as potential instruments: 2-step GMM

Post estimation test of exogeneity of instruments, In this regard, Lanne and Saikkonen (2009) argue when there is

causality, as causality is already is tested using Engle-Granger , lagged differences and lag levels could be taken as best instruments with GMM

C statistic (also known as a "GMM distance" or "difference-in-Sargan" statistic weak identification and Kleibergen-Paap rk Wald F statistics: ten or greater overidentifying restrictions• . Hansen J statistic:p-value result that is greater than 0.1

Summary of Regression Results Model 1

RGDP

Model 2

RGDP

Model 3

ARGDP

Model 4

ARGDP

Model 5

SRGDP

Model 6

SRGDP

Model 7

IRGDP

Model 8

IRGDP

Model 9

RGDP

lncapnet_pw .356*** .184* -0.0304 -.117* .423*** .43*** .428*** .436*** .815***

lnhcap_pw .0539*** .0455** 0.038 .0572* -.0585* -.0644*** .0661*** .0632*** 0.0195

lnelect_pw .345*** 0.165 0.352 0.294 .286* -0.00567 .251* -0.0209 0.163

lnroadtot_pw .575* 1.31*** 1.18*** .568** .414*

lnroadtots~w -.401** -0.186 -.759*** -.537*** -0.126 -.453*** -0.122

prersdp -.0921*** 0.0222 0.0209 .129* -.378*** -.22*** -.211*** -0.0764 -0.0933

rsdp1 -0.0253 .131* .162*** .256*** -.341*** -.18** -.152*** -0.0284 -0.0216

rsdp2 0.0316 0.117 .162* 0.15 -.253*** -0.121 -0.0788 0.0264 0.0576

rsdp3 .194*** .222* .444*** .27*** -.172* -0.112 0.0808 0.126 .292*

year -.0182*

lnaspurb_pw .582*** .585*** .62*** .426**

lngrvroad_pw -0.0699 -.374*** 0.214 -0.102

lnrdcapnet~w -.604***

_cons 45.3*** 10.9*** 17.3*** 9.1*** 11.7*** 10.2*** 8.05*** 7** 14.5***

r2_a 0.808 0.786 0.623 0.609 0.856 0.873 0.928 0.933 0.779

rmse 0.0416 0.0438 0.0598 0.0609 0.0583 0.0535 0.0481 0.0454 0.0455

N 35 35 35 35 35 35 35 35 35

K 20.406 87.309 106.206 42.852 106.206 131.911 106.206 99.961 99.961

K P 0.5148 0.5166 0.4921 0.3173 0.5201 0.4201 0.658 0.5863 0.5596

J 0.5368 0.4467 0.328 0.1302 0.2671 0.2073 0.4782 0.4475 0.4475

C 0.8396 0.8894 0.4969 0.9401 0.7833 0.7036 0.5633 0.4835 0.4835

F 2685 978 157 115 2520 696 2378 1828 595

legend: * p<0.05; ** p<0.01; *** p<0.001

Estimation Result • Model1: Road network per worker is positively related with

economic growth• Model2:Asphalt road has positive influence on overall GDP

growth, statistically significant. Whereas gravel road, it is statistically insignificant,

• To capture the degree of substitutability and complementarity between road network and capital stock, interaction term introduced,

• Put as model9, indicates that the two inputs are highly substitutable, indicating that government is forced to make a choice over these two

• i.e. Road and capital stock accumulation are competing over scare government-finance

• Model3 up to model8 are estimated with the aim of investigating the impact of road network on sectoral output, i.E. Agricultural GDP, industrial GDP and service GDP,

• In order to see the impact of road quality on these GDP, each sector GDP is estimated on asphalt and gravel road.

Estimation Result (contd.)• Model3 and model4 on total and disaggregated road on agricultural GDP. • Total road has positive and statistically significant impact• Impact of the disaggregated road. Asphalt road is positive;• Gravel road is negative: The possible explanations: not highly integrated with the

agriculture sector, in particular, in terms of accessibility to the rural dweller, i.e, not as such networked with asphalt and gravel road

• : disaggregated roads fails to reach the threshold level required spurring agricultural GDP growth.

• Community roads play significant role than these two types of road networks.• Community road is not well accounted to be in incorporated within the regression

equation• Model5 and model6 on industrial GDP. Both have positive impact on this sector

GDP growth. Asphalt road has significant and positive impact • Intuitive explanation, asphalt road is networked where industries are operating,

i.e. City areas. • In contrast, gravel network in urban areas are not prevalent. Its impact on this

sector GDP is negligible• Model7 and model8, are on service sector GDP is positive only for the total and

asphalt road network. In contrast, gravel road is negative. • Road quality is essential for the well performing service sector.

Findings• When other interaction terms introduced, results in

multicollinearity, two-step efficient GMM estimator, the stata routine command automatically drops those variables.

• Impact of policy interventions : for all of the models RSDP3 has significant growth spurring impact than the other two.

• We have impact lag: pronounced impact in later years than the moment the policy is introduced.

• Prersdp dummy in all the models has a negative influence for the country fails to have any growth spurring impact. The period was a period of political stabilization than growth.

• Finally, regarding the other inputs used in the models, in all cases the coefficients are found to be with the expectation

Conclusion• The study investigated achievements in road transport

in Ethiopia and its impact on both the overall and sectoral economic growth.

• To this end, reviewed theoretical and empirical researches related to the road development over a wide array of perspectives.

• Based on the findings of the descriptive analysis, regarding accessibility we have seen that the country has the lowest access rate to all weather roads.

• Accessibility is a better indicator to investigate the impact of poverty on socioeconomic development.

• This indicator is linked with rural population. Therefore, in order to reduce rural poverty or provision of public utilities, expanding road network would be the best way to reach the rural population.

Findings from the Descriptive Analysis• The government is making a relentless effort towards expanding the road network of the country.

• Issue of accessibility: The country’s overall accessibility is far below from what is needed to graduate to

the club of lower middle income countries.

• By now, only a quarter is achieved regarding rural accessibility to have a competent road network in

sub Saharan Africa

• In this decade the government has made sufficient attention in financing road projects. By now the

government’s expenditure has reached tenfold of what it has been a decade before.

• Reveals that the share of donors is not as such significantly improving in financing road projects in the

face of rising road expenditure

• Community roads are not given sufficient attention both in terms of expansion and proper accountancy

by either regional or federal road authorities.

• ERA should have every bit of information regarding community road networks.

• Future road expansion need to be an integral part of this road networks for this roads might be an easy

way to ascertain access to the destitute rural poor.

• Community roads better reflect the community demand that the network has to be constructed or

upgraded.

Findings• Road network capacity fails to reach the required threshold level to spur the growth

process. • The significantly negative coefficient of the gravel road network highlights the need for big

investment in expanding it to a wider section of rural dwellers.• Doing so, could lead to better productivity, ease of access to public services, health and

education centers and better market access to their produce. • Concerted effort to upgrade the country’s poor rural road connectivity. • Key question is, are the efforts made so far sufficient to have an impact on the livelihood of

the rural poor? • Less strong on the agricultural gdp growth and addressing the issue of accessibility to the

rural poor. • The effort is relatively better and has pronounced impact on industrial and service sector

gdp. • But for it to have a far reaching impact on agricultural growth and poverty reduction, a lot

more has to be done. • In sum, the country is lagging behind in the overall road network, rural accessibility,

community road management and upgrading and winning donors support to be competent in road infrastructure.

• Even though the recent trend is appreciable, the sector requires kind of ‘big-push’ investment to have competent road network with lower middle income countries.

Recommendation• Regarding community road, proper accounting and better manage ; for

improved performance.

• Government has shown commitment in financing road projects, donors need to

follow the footsteps of the government in financing road projects

• Gravel roads have so far failed to significantly affect growth

• For it to spur growth, agricultural gdp in particular , sufficient attention has to

be given by the government either in expanding the gravel road network or

upgrading and community roads.

• Road network expansion is found to be substitute to capital stock. This shows

that the two are competing over the government’s meager resource.

• Therefore, the government needs to design an innovative way of financing

road projects or at least create a mechanism until it become complementary

Caveats • Road network vis-à-vis number of cars • Road sector and its impact on socio economic

development : data, per planned study with the new road network expansion

• Impact analysis: data+++ planed• ERA; and institutional set up• Digression; Transport Economics

Special thanks goes to:Reviewer: Heady Derec

EDRIEDRI staff

ERA, MoFED, CSA Friends

Thank you all!!!