Transcript Incentives for Undeclared Work

Incentives for Undeclared Work
Prepared for Project on Undeclared Work in
Hungary
Jan Svejnar
December 2007
Theoretical considerations

Informal (undeclared wage) sector

Fallback (necessity)



Need policies for acquiring capabilities (or change regulation)
Simply enforcing taxes and regulations => fall in jobs and GDP,
and increase in unemployment
Choice (optimization)


Incentives to avoid paying taxes high when taxes and benefits
not tightly linked => distortions
Distortions = labor supply to the taxpaying sector becomes
more limited, while in the informal sector employment goes up
and productivity down => may lead to slower growth of GDP
Theoretical considerations (2)



If workers have incomplete information about or
time-inconsistent view of social security (old age)
benefits => less likely to work in formal sector =>
greater enforcement could increase welfare
Tax morale (inverse of tolerance of tax evasion)
matters => expenditure accountability, governance,
and public info. campaign = important issues
Watch out for unintended impacts (e.g., tiered tax
system by size of business => disincentive to grow)
Empirical Approach

Using October 2007 TARKI survey of individuals we
estimate the determinants of




% of wages the respondent thinks a typical company in his
business declares for tax and social contributions
whether respondent or his employer pays for unemployment
insurance
perceived effectiveness of reducing labor costs to employers
as a measure to reduce undeclared work
Since individuals/firms vary in degree to which they
view taxes as prepayment of benefits, explanatory
variables are


characteristics of firms/jobs
characteristics of workers
Empirical Estimates – Declared Wages

Table 1 – OLS results on % of wages declared


Respondents in more formal parts of economy think firms
declare less than do respondents in less formal parts
More formal = individual that







has more other income
qualifies for severance payments
has trade union in his/her workplace
has written work contract
lives in central region of Hungary
Exception = having a permanent contract has a positive
effect on the perception of how much firms declare
Respondents who have children (responsible, risk averse)
think firms declare less than do respondents with no children
Table 1: Determinants of Percentage of Wage Payments
Declared for Tax or Social Contributions
(2)
(3)
Other Income
Severance
Trade Union Presence
Written Contract
Permanent Status
With Children
Manager/Employer
Central Region
Firm Size (Less than 5 employees)
-1.075*
-1.021*
(0.548)
(0.541)
-11.166***
-11.979***
(3.563)
(3.533)
-6.803**
-7.124**
(3.369)
(3.067)
-37.742***
-37.759***
(9.198)
(9.097)
11.069**
10.265**
(5.202)
(5.080)
-6.371*
-6.693**
(3.284)
(3.362)
7.724
8.336
6.111
(5.938)
-4.093
-5.230
(3.824)
(3.769)
-2.181
(5.154)
Firm size (5-20 employees)
0.788
(3.767)
Constant
148.090***
150.442***
(12.709)
(12.332)
Observations
236
238
F
7.59
9.80
0.219
0.229
Adjusted R^2
Standard errors in parentheses.
* p<0.1, **p<0.05, *** p<0.01
Estimates – Paying for Unemployment
Insurance




Table 2 – only few systematic determinants of
whether the respondent or his employer pay for
unemployment insurance
= probability of paying unemployment insurance is
not strongly related to observable characteristics of
the individual and firm/job
Workers in construction are in firms that are less
likely to pay for unemployment insurance
Workers with written contracts are in firms that are
less likely to pay for unemployment insurance (?)
Table 2: Determinants of Employer/Employee Contribution to
Unemployment Insurance
(2)
(3)
Written Contract
Manager/Employer
Employer in Construction Sector
Trade Union Presence
Severance
Other Income
With Children
Central Region
-2.133***
-2.108**
(0.857)
(0.852)
-0.847
-0.862
(0.537)
(0.536)
-0.767
-0.732
(0.480)
(0.478)
-0.444
-0.448
(0.341)
(0.340)
-0.344
-0.363
(0.341)
(0.340)
-0.078
-0.078
(.058)
(0.057)
0.295
0.299
(0.370)
(0.328)
-0.400
(0.350)
Age (years)
-0.004
(0.017)
Constant
Observations
Pseudo R^2
Standard errors in parentheses.
5.050***
4.781***
(1.18)
(1.045)
278
278
0.083
0.078
* p<0.1, **p<0.05, *** p<0.01
Estimates – Reduction of Labor Costs to
Employers



Table 3 – only few systematic determinants of
whether the respondent views a reduction in costs
for employers (minimum wage, severance, etc.) as an
effective strategy for reducing undeclared work
= probability of respondent viewing a reduction in
costs for employers is not strongly related to
observable characteristics of the individual and
firm/job
The probablilty of the strategy being viewed as
effective is higher for those working in construction
and in Central Hungary
Table 3: Determinants of the Effectiveness of Reducing Labor
Law Compliance Costs on Undeclared Work
(1)
(2)
Other Income
Severance
Trade Union Presence
Employer in Construction Sector
Manager/Employer
Central Region
Written Contract
0.082
0.865
(0.050)
(0.050)
0.536*
0.552
(0.306)
(0.298)
-0.393
-0.345
(0.277)
(0.274)
1.241*
1.020
(0.647)
(0.569)
1.049
1.090
(0.661)
(0.658)
0.705**
0.589
(0.348)
(0.338)
1.208
(1.103)
Permanent Status
-0.515
(0.499)
Constant
Observations
Pseudo R^2
Standard errors in parentheses.
-0.473
0.201
(1.318)
(0.537)
301
303
0.060
0.048
* p<0.1, **p<0.05, *** p<0.01
Discussion

Survey – uneven evidence of undeclared wage sector



Few worker or firm/job characteristics affect



Only about 2% of respondents report non-payment of
contributions for pensions or health insurance
About 18% report non-payment of contributions for
unemployment insurance
probability of unemployment insurance payment
view that reducing labor costs to employers constitutes an
effective strategy for combating undeclared work
Individuals in the formal sector think undeclared
wages are more of a problem than individuals in the
less formal sector – Why (stereotype, truth telling)?