Data and preliminary findings from the quantitative phase (as the core method)

• 397 survey responses comprising 196 self-

employed and 201 employed taxpayers were

analysed using IBM SPSS Amos v.21.

• 12 structural models has been proposed to

scrutinise the impact of five types of

perceptions of corruption (refer to Figure A).

• So far, the data suggest while intention to

underreport (ITC) is a good predictor of

underreporting behaviour (TCB), attitudes

(ATB) and situation towards underreporting

(PBC) appear influential to intention to

underreport (see Table C).

• Perceptions of corruption generally affect

both attitudes and views on underreporting

tax, but with mixed results. As shown in

Figure D, while perceived grand corruption

(GCO) tends to affect attitude towards

underreporting (ATB), perceived petty

corruption (PCO) seems more influential to

views on underreporting (SNO).

Do perceptions of corruption influence personal income taxpayer

reporting behaviour? An empirical study of Indonesia

Arifin Rosid │School of Taxation and Business Law

Why is it important?

No country is immune from corruption and any tax system inevitably

suffers tax evasion. While there is no single tax compliance theory that

has been widely accepted, it is becoming increasingly difficult to ignore

the role of social and psychological factors in dealing with tax

(non)compliance issues.

High level of perceived corruption appears to

negatively influence taxpayers’ attitudes and

views on paying tax.

Primary research questions:

• To what extent do the perceptions of corruption affect taxpayers'

intentional tax underreporting behaviour?

• In what ways do the perceptions of corruption affect taxpayers'

intentional tax underreporting behaviour?

Acknowledgements: The author owes great debt of gratitude to his supervisors, Professor Chris Evans and Professor Binh Tran-Nam, for their constant support and invaluable assistance. He would also like to thank Ministry of

Finance of the Republic of Indonesia and the World Bank for providing SPIRIT Scholarship.

Introduction

Many studies have identified a wide range of behavioural

factors, but very little analytic attention has been paid to the

impact of perceptions of corruption. Specifically, what is not yet

clear is the impact of perceptions of corruption upon intentional

non-compliance behaviour.

Why Indonesia?

Indonesia is considered appropriate for this study for two

reasons. First, Indonesia is a major developing countries with

continual tax compliance issues. Second, Indonesia has been

classified as one of the most corrupt of major countries in Asia-

Pacific.

Research design

A sequential mixed-methods of “qualQUANT” has been adopted.

Phase 1 : In-depth interviews with 3 tax agents, 3 taxpayers and 3 tax officers

Phase 2 : Mixed-modes survey questionnaire to collect 397 respondents

Data and preliminary findings from qualitative phase

• A total of 432 minutes of interview data was converted into a verbatim format and

analysed with CDC EZ-Text 4.0 by using deductive thematic analysis.

• 28 categories and 272 assigned codes were identified from the data.

• Diagrammatic interpretation of the findings and hypothetical relationships among

variables were then developed (see Figures A and B below).

A. Diagrammatic interpretation of qualitative findings

Motivational factor

High level of perceived corruption

Negative attitudes towards

compliance

Negativesubjective

norms

Intention to underreport

tax

High level of perceived

behaviouralcontrol for non-

compliance

Level of tax under-

reporting

A

B

C

D

E F

Situational factor

B. Hypothetical relationships of qualitative findings

Locations of the survey

The survey was conducted in 12 tax offices across four regions in Indonesia

and took almost four months to complete. It was conducted with assistance

from two research intermediaries from Tax Center Faculty of Administrative

Science University of Brawijaya, Indonesia

Total effects Intervening

variables

Endogenous

variables

ATB SNO ITC TCB

Exogenous

variables

GCO .317** .036 -.131 -.073

PCO -.041 .331** -.027 -.015

PBC -.165** .011

Intervening

variables

ATB -.400** -.223

SNO -.130* -.073

ITC .559**

D. One of theoretical-based structural models in the study

C. SEM Output: Path coefficients

** significant at p = .05* *significant at p = .10

GCO = Perceptions of grand corruption

PCO = Perceptions of petty corruption

ATB = Attitudes toward tax underreporting

SNO = Views on tax underreporting

PBC = Situations toward tax underreporting

ITC = Intention to underreport

TCB = level of underreported tax

(+)

(+)

(+)

(+)

(+)

(+) = influential

1

2

3

4

5

6

7

Malang (100 taxpayers)

Yogyakarta (100 taxpayers)Surabaya (97 taxpayers)

Bali (100 taxpayers)

INDONESIA

6 participants

3 participants

90 87 8577 74

63 6156 55

49

40 39 37 36 36 34 33 32

Ne

w Z

ea

land

Sin

gap

ore

Au

str

alia

Ho

ng

Ko

ng

Ja

pan

Bh

uta

n

Ta

iwa

n

Ko

rea

(S

ou

th)

Bru

ne

i

Ma

laysia

Sri L

an

ka

Ch

ina

Th

aila

nd

Ind

ia

Mo

ng

olia

Ph

ilip

pin

es

Tim

or-

Le

ste

Ind

on

esia

Corruption Perceptions Index 2012 among major selected Asia-Pacific countries

Source: Transparency International, 2013

On a 0-100 scale, 0 means a country is

perceived as very corrupt and 100 means it is

perceived as very clean.

Indonesia

y = 0.1264x + 10.334

0

10

20

30

40

50

60

0 20 40 60 80 100

TAX

REV

ENU

E TO

GD

P R

ATI

O (

20

08

-20

12

)

CORRUPTION PERCEPTION INDEX 2012

Correlation between share of tax revenue in GDP and perceived level of corruption

Source: Author’s calculation on 119 countries data of the CPI 2012 data (Transparency International, 2013a) and the World Bank data of tax revenue as percentage of GDP (World Bank, 2013d).