The Growth of the Shadow Economy in the OECD: Some Preliminary Explanations(1).

SCHNEIDER, FRIEDRICH

"... growing concerns have led many economists to the difficult task of measuring the size and development of the shadow economy [in OECD countries], tracing its main causes and analyzing the interactions between official and unofficial economies."

Over the last two decades, officials, politicians and social scientists have become increasingly concerned with the phenomenon of the shadow economy In many countries of the world, including the most developed, there are several important reasons why informed observers should be concerned about the size and growth of the shadow economy

First, an increase in the size of the shadow economy is usually a result of a rise in the overall tax and social security payments burden. This increased burden creates incentives for individuals to undertake commercial activity in the informal sector to avoid additional payments to the government. As a result, the rise in the shadow economy may lead to erosion of the tax and social security bases and a decrease in tax receipts. In turn, this may lead to a further increase in the budget deficit or to further increases in direct as well as indirect taxation. This may eventually create a vicious circle of increasing tax rates and ever-growing shadow economy.

Second, the very nature of the shadow economy makes it difficult to accurately measure official economic indicators, such as unemployment, income and consumption. Thus, economic policy, which is based on interpretation of these indicators, can be inefficient or even harmful. Where there is a prospering shadow economy, politicians must base policy on unreliable official indicators. The very direction of intended policy measures may therefore be questionable.

Third, a growing shadow economy creates strong incentives for workers--domestic and foreign--to work more in the shadow economy and less, or less efficiently, in the official economy

These growing concerns have led many economists to the difficult task of measuring the size and development of the shadow economy, tracing its main causes and analyzing the interactions between the official and unofficial economies. This paper provides an overview of this research. The first section lays out the latest results of the size of the shadow economies for 18 Organization for Economic Cooperation and Development (OECD) countries. The second section examines the development of the shadow economy over time, using Austria as a case study The third section looks at the influence of a changing tax structure on the shadow economy The final section summarizes my conclusion and gives some prospects for future research.

EMPIRICAL FINDINGS ON THE SIZE OF THE SHADOW ECONOMY FOR OECD COUNTRIES

As literature on the topic indicates, the currency demand approach is most often used to estimate the size of the shadow economy.(2) This approach estimates the size of the shadow economy by calculating the difference between the size of the economy (based on actual levels of cash) compared to the expected level of cash demand (given its GDP and its velocity of money).(3) It has been applied to estimate the size of the informal economies in most OECD countries. The results are shown in Table 1. Note that Greece has the largest informal economy, which comprises 29 percent of its GDP, while Switzerland has the smallest, with 8 percent. The US is also on the low end of the spectrum, with 8.9 percent.

Table 1 The Development of the Size of the Shadow Economy in Selected OECD Countries Over Time Applying the Currency Demand Approach

 Size of the Shadow Economy
 (in percent of official GDP)
 Using the Currency Demand
 Approach in the Year

 1960 1970 1980

Australia -- -- --
Austria 0.4 1.8 3.0
Belgium -- 10.4 16.4
Canada -- -- 10.1-11.2
Denmark 3.8-4.8 5.3-7.4 6.9-10.2
France -- 3.9 69.0
Germany 2.0-2.1 2.7-3.0 10.3-11.2
Greece -- -- --
Ireland -- 4.3 8.0
Italy -- 10.7 16.7
Netherlands -- 4.8 9.1
New Zealand(*) -- 6.9 9.2
Norway 1.3-1.7 6.2-6.9 10.2-10.9
Spain -- 10.3 17.2
Sweden 1.5-1.8 6.8-7.8 11.9-12.4
Switzerland 1.2 4.1 6.5
United Kingdom -- 2.0 8.4
USA 2.1-4.1 2.6-4.6 3.9-6.1

 Size of the Shadow Economy
 (in percent of official GDP)
 Using the Currency Demand
 Approach in the Year

 1994 1995 1996

Australia 13.0 13.2 14.0
Austria 6.7 7.3 8.3
Belgium 21.4 21.6 21.9
Canada 14.6 15.0 15.1
Denmark 17.6 18.1 18.3
France 14.3 14.8 14.9
Germany 13.1 13.9 14.5
Greece 26.0 26.6 28.5
Ireland 15.3 15.6 15.9
Italy 25.8 26.2 27.0
Netherlands 13.6 14.1 14.0
New Zealand(*) 11.3 -- --
Norway 17.9 18.5 18.9
Spain 22.3 22.6 22.9
Sweden 18.3 18.9 19.2
Switzerland 6.6 6.9 7.5
United Kingdom 12.4 12.6 13.1
USA 9.4 9.0 8.8

 Size of the Shadow Economy
 (in percent of official GDP)
 Using the Currency Demand
 Approach in the Year

 1997 1998

Australia 13.9 14.1
Austria 8.9 9.1
Belgium 22.4 22.6
Canada 14.8 15.0
Denmark 18.1 18.4
France 14.7 14.9
Germany 15.0 14.7
Greece 28.7 29.0
Ireland 16.1 16.3
Italy 27.3 27.8
Netherlands 13.5 13.5
New Zealand(*) -- --
Norway 19.4 19.7
Spain 23.1 23.4
Sweden 19.8 20.0
Switzerland 8.1 8.0
United Kingdom 13.0 13.0
USA 8.8 8.9

Source: E Schneider, "The Shadow Economies of Western Europe," Journal of The Institute of Economic Affairs, 17, no. 3, pp. 42-48; F. Schneider, "Ist Schwarzarbeit ein Volkssport geworden? Ein internationaler Vergleich des AusmaBes der Schwarzarbeit von 1970-97" in S. Lamnek and J. Luedtke, eds., Der Sozialstaat avischen Markt und Hedeonismus, Obladen (West Germany: Luedtke and Budrich publishing company,, 1999) pp.293-318.

(*) Source: D. Giles, "Modelling the hidden economy in the tax cape in New Zealand," Working paper (Department of Economics, University of Victoria, Canada, 1999).

Table 1 clearly shows that the increase of the shadow economy in all of these countries is quite remarkable and evident in all investigated countries. Whereas in 1960, the size of the shadow economy was below 5 percent of GDP, the shadow economy increased dramatically by 1998. By this time, in many OECD countries, the informal sector constituted over 13 percent of GDP; elsewhere, it was closer to 10 percent. In Belgium, the size of the shadow economy was 10.4 percent of GDP in 1970; by 1998, it had more than doubled to 22.6 percent. In Italy, the size of the shadow economy was 10.7 percent in 1970 and 27.8 percent in 1998--an increase of more than 160 percent in only 20 years.

Yet even in countries where the shadow economy is considerably smaller, the increase has been remarkable. In the United States, the shadow economy constituted between only 3.6 and 4.6 percent in 1970. By 1998, it had increased to 8.9 percent, a growth of 168 percent. The shadow economy of the Netherlands increased from 4.8 percent in 1970 to 13.5 percent in 1998, an increase of more than 181 percent. These findings demonstrate convincingly that in the OECD countries--the most affluent and stable economic systems of the world--there was a strong and steady increase in the size of the shadow economy over the last three decades.

Table 2 presents one possible explanation of the different sizes of the shadow economies in the OECD countries by comparing their overall tax and social security contributions with the size of their shadow economies for 1996.

Table 2 Size of the Shadow Economy and the Burden of Taxes and Social Security Contributions in OECD Countries

 Size of the
 shadow
 economy Value added Average direct
 (in % tax rate tax rate
 of GDP) (in %)(*) (in %)(**)

 1996 1996 1996

 (1) (2) (3)
Greece 28.5 18.0 11.0
Italy 27.0 19.0 12.0
Spain 22.9 16.0 13.0
Belgium 21.9 21.0 19.0
Sweden 19.2 25.0 20.0
Norway 18.9 23.0 19.0
Denmark 18.3 25.0 36.0
Ireland 15.9 21.0 20.0
Canada 14.6 7.0 21.0
Germany 14.5 15.0 18.0
France 14.3 20.6 6.0
Netherlands 14.0 17.5 10.0
U.K. 13.1 17.5 16.0
USA 8.8 3.0 17.0
Austria 8.3 20.0 8.0
Switzerland 7.5 6.5 10.0

 Social Social Total social
 security security security
 contributions contributions contributions
 by employees by employers rate (in %)
 rate(***) rate(***) sum of
 (in %) (in %) (4)+(5)

 1996 1996 1996

 (4) (5) (6)
Greece 15.8 27.5 43.3
Italy 9.9 32.0 41.9
Spain 6.6 31.6 38.2
Belgium 10.0 26.0 36.0
Sweden 4.0 29.6 33.6
Norway 7.0 12.8 19.8
Denmark 9.0 0.0 9.0
Ireland 7.2 12.3 19.5
Canada 7.0 8.0 15.0
Germany 16.1 16.1 32.2
France 13.0 31.0 44.0
Netherlands 31.0 8.8 39.8
U.K. 10.7 10.2 21.4
USA 7.6 13.8 21.4
Austria 18.2 24.2 42.4
Switzerland 11.6 11.6 23.2

 Total social
 security Total tax and
 contributions + social security
 direct tax burden: burden: sum
 sum (4)+(5)+(3) (2)+(3)+
 (in %) (4)+(5)

 1996 1996

 (7) (8)
Greece 54.3 72.3
Italy 53.9 72.9
Spain 51.2 67.2
Belgium 55.0 76.0
Sweden 53.6 78.6
Norway 38.8 61.8
Denmark 45.0 70.0
Ireland 39.5 60.5
Canada 36.0 43.0
Germany 50.2 65.2
France 50.0 70.6
Netherlands 49.8 67.3
U.K. 37.4 54.9
USA 38.4 41.4
Austria 50.4 70.4
Switzerland 33.2 39.7

(*) Rates of the year 1996; USA: Average sales tax

(**) Average direct tax rate is calculated as the stun of all income taxes (+ payroll and manpower taxes) paid on wages and salaries (including income of self-employed) divided by gross labor costs of an average income earner.

(***) The rate is calculated on the basis of the annual gross earnings of an average income earner.

Source: W. Leebtritz, J. Thornton. and A. Bibbee, "OECD, 1997: Taxation and Economic Performance," Working Paper 176 (Paris, 1997).

With a few exceptions, countries that had the largest shadow economies in 1996--Greece, Italy, Belgium and Sweden--also had the highest tax and social security burden. Countries such as Switzerland and US, which had the lowest overall tax and social security burden, also had the smallest shadow economies. There are exceptions; the United Kingdom and Austria, with relatively high overall tax and social security burden, had a rather low level of shadow economic activity. Yet, the overall picture convincingly indicates a correlation between higher overall social security and tax burden and a larger shadow economy, ceteris paribus.

After this brief discussion of the increase in the size of the shadow economies, we can make some observations on the labor market in the shadow economy Table 3 presents the estimates of the size of the shadow economy labor force in several OECD countries from 1974 to 1998.

Table 3 Estimates of the Size of the "Shadow Economy Labor Force" in Some OECD Countries 1974-1998

 Year Participants
 in 1000
 people(*)

Austria 1990-91 300
 1997-98 500

Denmark 1980 --
 1986 --
 1991 --
 1994 --

France 1975-82 800-1,500
 1997-98 1,400-3,200

Germany 1974-82 2,000-3,000
 1997-98 5,000

Italy 1979 4,000-7000
 1997 6,600-11,400

Spain 1979-80 1,250-3,500
 1997-98 1 ,500-4,200

Sweden 1978 750
 1997 1,150

EU 1978 10,000
 1997-98 20,000

OECD 1978 16,000
 1997-98 35, 000

 Shadow
 Participants Economy
 in % of (% of DP)
 Labor Currency
 Force(**) Demand
 Approach

Austria 9.6 5.47
 16.0 8.93

Denmark 8.3 8.6
 13.0 --
 14.3 11.2
 15.4 17.6

France 3.0-6.0 6.9
 6.0-12.0 14.7

Germany 8.0-12.0 10.6
 22.0 14.7

Italy 20.0-35.0 16.7
 30.0-48.0 27.3

Spain 9.6-26.5 19.0
 11.5-32.3 23.1

Sweden 13.0-14.0 13.0
 19.8 19.8
EU 14.5
 --
OECD 15.0
 --

(*) Estimated full-time jobs, including unregistered workers, illegal immigrants, and second jobs. The estimations are based either on surveys (e.g. Denmark) or on a calculation using the value-added values of the shadow economy (subtracting all material inputs) and assuming certain average values of earnings paid per hour in the shadow economy.

(**) In percent of the population aged 20-69. In Denmark: In percent of the population aged 20-69, survey method (% heavily engaged in shadow economy activities).

Source: F. Schneider and D. Enste, "Shadow Economies Around the World: Sizes, Causes and Consequences," The Journal of Economic Literature (March 2000).

These figures only provide a rough estimate; however, they indicate a substantial increase in the size of the populations engaged in shadow economic activity For example, the results for Denmark show an increase from 8.3 percent of the total labor force in 1980 to 15.4 percent in 1994. In Germany, this proportion rose from around 8 percent to 12 percent of the total labor force in the period from 1974 to 1982 and rose again to 22 percent from 1997 to 1998. While estimates for Italy's informal labor force in 1979 were relatively large compared to its OECD neighbors (20 to 35 percent), estimates now range from 30 to 48 percent of the total labor force. Even France, which had a relatively low percentage of informal labor from 1975 to 1982 (3 to 6 percent) has doubled its informal labor force to a level of 6 to 12 percent of the total labor force in 1997 and 1998. In the European Union, at least 10 million people are engaged in the shadow economy. In OECD countries combined, about 16 million people work "informally" These figures demonstrate that the shadow economy labor market is vibrant and extensive. They also may provide an explanation to the puzzling situation, for example, in Germany, where we observe a high and persistent unemployment rate in the "official" labor market.

To better understand the root causes behind the increases in shadow economic activity, this paper examines Austria as a case study. Austria was chosen because there are comprehensive data on various kinds of direct and indirect tax burdens as well as the intensity of its regulations. Furthermore, based on the available data for Austria, we can investigate which of the major causes, such as tax pressure and regulations, contributes most to the size and growth of the shadow economy

EMPIRICAL DATA ON THE DEVELOPMENT OF THE AUSTRIAN SHADOW ECONOMY

The prevalent method used to estimate the size of the shadow economy in OECD countries is the currency demand approach. This approach assumes that hidden transactions are undertaken in the form of cash payments, so as to leave no observable traces for the tax authorities.(4) An increase in the size of the shadow economy will therefore increase the demand for cash. The economic theory of income velocity predicts the amount of cash an economy will demand, given its GDP and the speed at which money moves through the economy. Therefore, it is possible to estimate the size of the shadow economy by isolating the excess demand for cash--that is, comparing the predicted level of cash in the economy to the actual level of cash. Note, however, that the entire difference of cash levels can not be attributed to the shadow economy. Some factors can be explained by changes in conventional factors such as income, payment habits and interest rates. The remaining difference can be attributed to variables, such as a rising tax burden and increased government regulations, which encourage entrance into the shadow economy. Assuming the income velocity for cash is the same in both the official and shadow economies, the size of the shadow economy can be computed in comparison to the official GDP. The development of the shadow economy can then be calculated by comparing the difference between the calculated levels of cash when the tax burden and government regulations are at their lowest levels with the levels of cash when the tax and regulatory burden is at its highest level.

The currency demand approach is one of the most commonly used. Nevertheless, it has been criticized on a variety of grounds.(5) The first objection is that not all transactions in the shadow economy are paid in cash. A study by Isachsen and Strom discovered that in Norway in 1980 roughly 80 percent of all transactions in the hidden sector were paid in cash: 20 percent of the shadow activity are thus not taken into account by the currency demand method.(6) This shows that the size of the total shadow economy may be even larger than previously estimated.

The second objection is that most studies only consider one factor, the tax burden, as a cause of the shadow economy. Other causes, such as the impact of regulation, the complexity of the tax system, taxpayers' attitudes toward the state and "tax morality" are not taken into consideration. If, as seems likely; these other factors also have an impact on the extent of the hidden economy, it might be larger than reported in most studies.(7)

A third objection to this approach is in the actual measurement of money supply. A study by Garcia(8) and Park(9) point out that increases in the amount of cash in an economy, which consists of currency and demand deposits, are largely due to a decreased desire for demand deposits rather than an increased demand for cash caused by activities in the shadow economy. A related issue is that the US dollar is used as an international currency. Economists need to consider, and compensate for, the amount of US dollars held abroad.

A fourth objection is that the assumption that the velocity of money is the same in both formal and informal economies is a faulty one. Klovland argues that there is considerable uncertainty about the velocity of money circulation in the official economy. So the velocity of cash in the shadow economy is even more difficult to estimate. While Klovland's observations stem only from his research on Scandanavia, the point is a relevant one.(10)

In spite of these weaknesses, the currency demand approach has been widely used and serves as a useful tool for cross-country comparative analysis.(11) Moreover, there is a large amount of data available--spanning from 1956 to 1998--on Austria's monetary sector, its tax structure and its regulatory system. From this data, I have built a model to determine the size and development of its shadow economy, utilizing four principal causal variables in the model:(12)

i) The burden of total direct taxation (DIRT), both average and marginal(13) with:

[DIRT.sub.t] = [TATRY.sub.t] + [AMTRY.sub.t] where [TATRY.sub.t] = total average tax rate (including social security payments) on wage income in the year t, and [AMTRY.sub.t] = average marginal tax rate on wage income in year t.

A rising burden of direct taxation provides a strong incentive to work in the shadow economy (positive sign expected).

ii) The burden of indirect taxation (INDT) defined as the percentage ratio of the sum of all indirect taxes to gross domestic net production of indirect taxes. As with direct taxation, a rising burden of indirect taxation provides a strong incentive to work in the shadow economy (positive sign expected).

iii) The complexity of the tax system (VIST)(14) which following Wagner(15) (1976) and Clotefelter(16) is defined with the help of the Herfindahl-Hirschman concentration measure:

[MATHEMATICAL EXPRESSION NOT REPRODUCIBLE IN ASCII]

where, [REV.sub.it] equals the i-th revenue share of the total revenue amount for all in revenue items in year t; [EXEM.sub.jt] equals the j-th tax exemption of the total tax exemptions n in year t, which refers to direct and indirect taxes and, for statistical reasons, REV and EXEM are combined into one variable.

Thus, VIST is equal to 2 if only one revenue source exists in year t and if there are no tax exemptions. In this case, the tax system is highly simple and every change will be immediately recognized by the taxpayer. For example, an increase in the tax burden will immediately lead to more shadow economic activity. As the number of revenue sources and tax exemptions increases, the tax system becomes more complex. In this case, a tax increase is much less recognizable (that is, felt less) by the taxpayer and hence leads to a lower increase in shadow economic activity than under an extremely simple tax system.(17) Moreover, the more "legal" tax loopholes and tax exemptions created, the lower the incentive to work in the shadow economy, ceteris paribus (positive sign expected). iv) The intensity of regulation (REG) refers to all existing and enforced laws concerning federal, state and local regulatory activities.(18) An increase in the burden of regulation is expected to provide a strong incentive to work in the shadow economy (positive sign expected).

The estimation results of the currency demand function are shown in Table 4.

Table 4 Estimation Results of the Currency Demand Function for Austria(*)

 Dependent Variable: Real
 Currency per Capita,
 In ([CUR.sub.t]/[POP.sub.t])
 Independent Variables Estimation Period

 1956-1998 1956-1990

Lagged Dependent Variable 0.64(**) 0.673(**)
In ([CUR.sub.t-1] /[POP.sub.t-1]) (10.93) (11.87)

Real Consumption per capita 0.734(**) 0.749(**)
In ([C.sub.1]/[POP.sub.1]) (6.53) (6.09)

Number of Eurocheque Systems per- 0.175(**) -0.204(*)
capita In ([ES.subt-1] /[POP.sub.t-1] (-2.29) (-2.39)

Real Interest Rate on Bonds -0.113(*) -0.122(*)
In ([IR.sub.t]) (-2.31) (-2.55)

Direct Tax Burden (including social 0.224(**) 0.209(**)
security payments); ln ([DIRT.sub.t] (3.80) (3.21)

Indirect Tax Burden 0.129(*) 0.119(*)
In ([INDT.sub.t]) (2.28) (2.02)

Complexity/"Visibility" of the 0.124(*) 0.119(**)

Tax System In ([VIST.sub.t]) (2.49) (2.71)
Intensity of Regulation 0.203(**) 0.189(**)
In ([REG.sub.t]) (3.36) (2.92)

Constant Term 1.05 1.98
 (0.80) (-1.63)
Test Statistics:
[R.sup.2] 0.990 0.993
S.E. 0.014 0.015
Durbin's h 1.07 1.18
rho (1) 0.21 0.23
D.E 35 27
Ex-post Forecast 1990 - 1998
RMSE -- 1.39
Theil's U1 -- 0.39

(*) All equations are estimated by all ordinary least-squares procedure using annual data. [R.sup.2] is the adjusted coefficient of determination (corrected for the degrees of freedom); S.E. shows the standard error of the estimation. Durbin's h is Durbin's h-test against autocorrelation, when a lagged dependent variables used as regressor. Rho (1) is the autocorrelation coefficient of first order. D.E stands for the degrees of freedom. RMSE is the root mean squared error and Theil's U 1 stands for Theil's inequality coefficient. The term "In" indicates that these variables have been transformed to natural logarithms. Numbers in parentheses below coefficient estimates are t-values. (*), * and ** indicate significance at the 90%, 95% and 99%-level, respectively Real Currency is defined as cash money held outside banks.

In most studies of Austria's neighbors, as well as the Scandinavian countries, it is generally assumed that there would have been no shadow economy if the direct and indirect tax burdens and the amount of regulation had remained at their historical minimum. Thus, to calculate an estimate of the shadow economy, I first estimated "normal" currency demand by using the historical variables, the minimum values in 1956. Then, I calculated actual currency demand from 1956 to 1998. The difference between the actual and the simulated currency holdings is assumed to reflect the amount of currency used for shadow economic transactions. Assuming the same velocity for currency used in the shadow economy as for the official economy, I can compute the size of the shadow economy, as well as compare it to official GDP. Table 5 shows the absolute values and the growth of the Austrian shadow economy from 1965 to 1998. Since 1965, the Austrian shadow economy increased steadily and reached a peak value in 1998 with 9.12 percent of official GDP.

Table 5 Size of the Austrian Shadow Economy*

 "Official" Gross Value Added in the
Year Domestic Produc- Shadow Economy,
 tion, real, Bill. AS real, Bill AS

1965 440.95 5.14
1966 466.20 5.69
1967 479.98 5.49
1968 502.18 6.21
1969 533.44 8.31
1970 571.25 10.47
1971 600.31 9.58
1972 637.69 10.41
1973 669.29 13.34
1974 695.79 12.46
1975 692.84 12.02
1976 724.75 15.39
1977 756.12 18.01
1978 760.23 20.06
1979 795.96 22.88
1980 820.03 24.98
1981 818.58 26.96
1982 828.62 29.09
1983 845.54 31.86
1984 862.66 36.38
1985 887.39 36.98
1986 904.46 39.64
1987 924.84 41.22
1988 959.91 46.37
1989 1,004.09 50.89
1990 1,061.32 54.56
1991 1,126.32 59.89
1992 1,160.37 68.23
1993 1,181.90 74.46
1994 1,228.95 82.34
1995 1,243.89 89.56
1996 1,273.98 105.86
1997 1,303.99 116.32
1998 1,355.50 123.62

 Size of the Shadow
Year Economy in percent
 of the "official" GDP

1965 1.16
1966 1.22
1967 1.14
1968 1.24
1969 1.56
1970 1.83
1971 1.59
1972 1.63
1973 1.99
1974 1.79
1975 1.73
1976 2.12
1977 2.38
1978 2.63
1979 2.87
1980 3.05
1981 3.29
1982 3.51
1983 3.77
1984 4.22
1985 4.16
1986 4.38
1987 4.46
1988 4.83
1989 5.07
1990 5.14
1991 5.32
1992 5.88
1993 6.30
1994 6.70
1995 7.20
1996 8.31
1997 8.92
1998 9.12

(*) Assumptions made for the calculation of the shadow economy: (i) All transactions in the shadow economy are made in cash. (ii) Direct and indirect tax burdens, the complexity of the tax system and the intensity of regulations are the reasons for working in the shadow economy (iii) In 1960 the shadow economy did not exist. (iv) The velocity of currency is the same in the shadow economy as in the official economy and is calculated by dividing total income by M1. (v) For the calculation, first the currency-demand equation from table 4 is used.

THE INFLUENCE OF CHANGING TAX STRUCTURES ON THE AUSTRIAN SHADOW ECONOMY

Table 5 indicates that the Austrian shadow economy has grown steadily over the last 33 years. This is surprising since there were major changes in the direct and indirect tax rates, as well as the actual tax structure during this period. In order to determine why Austria's shadow economy has increased, I investigate what effects these tax changes had on the development of the shadow economy I will do this by considering three chronological cases.(19)

The first period considered is from 1972 to 1973. On 1 January 1973, Austria introduced a value-added tax of 16 percent. At the same time, taxation was changed from family (joint) to individual income basis, and additional tax exemptions were legislated to stimulate investment and accumulation of capital.

The second period of tax change is from 1983 to 1984 when Austrians experienced an increase in indirect tax rates. During this period, value-added tax increased from 18 to 20 percent, while the value-added tax on luxury goods increased from 30 to 32 percent. The government also introduced a tax on interest. At the same time, tax exemptions were legislated to stimulate venture capital.

The third period is from 1988 to 1989, which saw major income tax reform. The Austrian tax system underwent a considerable reduction of all marginal tax rates on income, from 62 percent down to 50 percent (top marginal tax rate). There were simultaneous reductions in the number of income-tax rates from 10 percent to 5 percent, a decrease in the average tax rates on capital gains and profits and a general simplification of the tax system.

Table 6 shows the influence of these three major tax changes on the development of the shadow economy In the first case, the changes in direct taxation (for example, changing from family/ joint to individual taxation) had the strongest effect on the shadow economy This policy change caused a 42 percent increase of the shadow economy The introduction of a value-added tax, which replaced the old turnover tax, reduced complexity, but led to a rise in the indirect tax burden.

Table 6 The Influence of Changing Tax Systems on the Size and Growth of the Austrian Shadow Economy

 Amount of the Shadow
 Size of the Economy Attributed to
 Shadow
 Economy Inten-
 Direct Indirect
Year Tax Tax
 Burden Burden
 Bill. AS Bill. AS Bill. AS
Case (i)
1972 10.41 5.58 1.45
1973 13.34 6.82 2.20

Diff. (73-72) 2.93 1.24 0.75
(rel. weight) (100.0%) (42.3%) (25.6%)

Case (ii)
1983 31.86 13.48 7.74
1984 36.38 16.12 9.57

Diff. (84-83) 4.52 2.64 1.83
(rel. weight) (100.0%) (58.4%) (40.5%)

Case (iii)
1988 46.37 19.01 11.27
1989 50.89 17.86 12.21

Diff. (89-88) 4.52 -1.15 0.94
(rel. weight) (100.0%) (i-25.4%) (20.8%)

 Amount of the Shadow Economy
 Attributed to

 Com- Inten-
 plexity sity of
Year of Tax Regula-
 System tion
 Bill. AS Bill. AS

Case (i)
1972 2.37 1.01
1973 3.24 1.08

Diff. (73-72) 0.87 0.07
(rel. weight) (29.7%) (2.4%)

Case (ii)
1983 5.19 5.45
1984 5.16 5.53

Diff. (84-83) 0.-03 0.08
(rel. weight) (-0.7%) (1.8%)

Case (iii)
1988 6.12 9.97
1989 8.90 11.91

Diff. (89-88) 2.79 1.94
(rel. weight) (61.7%) (42.9%)

The second case, covering the years 1983 and 1984, shows that the "cold progression effect" of income taxation had more of an impact on the shadow economy than the increase in indirect taxes. Whereas income taxation was responsible for 58.4 percent of the increase of the shadow economy, the increase in indirect taxes contributed only 40.5 percent.

The third case, the major tax reform of 1989, is the most interesting. Ceteris paribus, the reduction in the direct marginal and average tax burden would have decreased the shadow economy by AS1.15 billion (Austrian Schillings) ma reduction of 2.5 percent of the shadow economy of AS46.37 billion in 1988. However, the negative effect of the decreased personal income tax rates on the shadow economy is more than offset by the strongly reduced complexity of the tax system, which renders the shadow economy more sensitive to changes in the tax system. The increased regulation intensity compensates for the negative effect of the reduced income tax rates as well.

Table 7 shows the proportional importance of the four causes responsible for the size and growth of the Austrian shadow economy The direct tax burden had the largest influence of all the factors operating over this period. However, the influence is a strongly declining one: up to 1977, roughly 50 percent of all shadow economic activities were caused by the burden of direct taxation. In 1995, this impact had diminished to 29 percent. The opposite trend holds for the influence of the indirect tax burden. Whereas in the 1960s only 12 percent of shadow economy activities were caused by this factor, in 1994 and 1995, its influence had risen to 26 percent.

Table 7 Breakdown of the Shadow Economy According to the Four Causes

 The Shadow Economy in Percent Share of

 Complexity Intensity of
Year Direct Tax Indirect Tax of the Tax Regulation
 Burden % Burden % System % %

1965 51.2 12.1 25.9 9.8
1966 51.8 12.4 25.3 11.1
1967 49.3 11.2 26.1 13.4
1968 49.6 11.9 25.3 13.2
1969 50.9 12.6 24.6 11.9
1970 51.3 13.5 23.4 11.8
1971 52.3 13.3 23.0 11.4
1972 53.6 13.9 22.8 9.7
1973 51.1 16.5 24.3 8.1
1974 52.0 16.0 24.0 8.0
1975 50.9 15.9 23.4 9.8
1976 51.4 18.4 21.2 9.0
1977 51.0 20.1 20.6 8.3
1978 49.6 22.0 18.7 9.7
1979 47.6 21.9 18.8 11.7
1980 46.3 21.6 17.9 14.2
1981 45.0 23.3 17.0 14.4
1982 42.2 24.9 17.1 15.8
1983 42.3 24.3 16.3 17.1
1984 44.3 26.3 14.2 15.2
1985 44.0 25.2 15.2 15.6
1986 43.1 25.2 14.1 17.6
1987 40.9 24.0 13.7 21.4
1988 41.0 24.3 13.2 21.5
1989 35.1 25.6 17.1 22.2
1990 34.6 25.1 17.3 23.0
1991 34.1 25.4 17.0 23.5
1992 33.1 25.6 17.4 23.9
1993 31.9 25.9 17.9 24.3
1994 30.5 26.2 18.2 25.1
1995 28.7 26.6 18.7 26.0

The influence of the complexity of the tax system on the shadow economy also declined in the 1980s and 1990s. In the 1960s, Austria had a simple tax system, causing 25 percent of all shadow economic activities. This influence diminished to 18 percent in 1994 and 1995. On the other hand, we can observe a strong increase in the percentage share of the shadow economy due to the intensity of regulation. In the 1960s and 1970s this factor caused only 10 to 12 percent of all shadow economic activities, but increased to approximately 25 percent in the years 1994 and 1995. These findings confirm the theoretical hypothesis about the influence on the complexities of the shadow economy The decline of the direct tax burden in 1989 had a negative effect on the size of the shadow economy. However, it was offset by a decrease in complexity of the tax system and an increase in regulations--especially in the labor market.

CONCLUSIONS

Though there are many obstacles, progress has been made in measuring the size of, and analyzing the consequences of, the shadow economy. This paper has shown that it is possible to estimate the size of the shadow economy. I have demonstrated that the currency demand approach helps collect useful insights about the size and development of the shadow economy of the OECD countries. The most general conclusion is that the shadow economy is significant. For 15 of the 18 OECD countries investigated, the estimated size of the informal economy was more than 13 percent of GDP in 1998.

Although different methods provide a rather wide range of estimates, there is general agreement that the size of the shadow economies for the OECD countries has been growing strongly over the recent decades. This conclusion was demonstrated using the specific case of Austria, where the shadow economy increased from AS12.02 billion (or 1.73 percent of official GDP) in 1975 to AS89.56 billion (or 9.12 percent of the official GDP) in 1995.

This research enables us to analyze the effects of a changing economic policy on the development of the shadow economy. When the government changes the tax structure, for instance, decreasing the direct tax burden, one would expect that the shadow economy would decrease. However, in the period from 1989 to 1995, even when a major tax reform with significant lowering of tax rates had been undertaken, the shadow economy still increased. This was due to other important factors, such as increased regulation, encouraging people to work in the shadow economy. Hence, the effect of a lower direct tax burden was more than offset by other factors.

Shadow economies are a complex phenomenon, present to a large extent in both developed and developing economies. People engage in shadow economic activity for a variety of reasons-mostly in response to government actions, notably taxation and regulation. A government aiming to decrease shadow economic activity has to first analyze the complex and frequently contradictory relationships among the consequences of its own policy decisions.

(1) Earlier versions of this paper were presented at the 54th Congress of the International Institute of Public Finance (IIPF), Cordoba, Argentina, 24-27 August 1998, and at seminars at the Universities of Vigo (Spain) and San Diego de Compostella, at the Austrian National Bank, the German Bundesbank as well at the Canadian Public Economic Study Group, Ottawa (Canada). I would like to thank Eduard Hochreither (Austrian National Bank), Othmar Issing (German Bundesbank) and T. K. Rymes (Carleton University, Ottawa, Canada) for most helpful comments and suggestions.

(2) Compare, for example, B. S. Frey and W. W. Pommerehne, "The Hidden Economy: State and Prospects for Measurement," Review of Income and Wealth, 30 (1984) pp. 1-23; F. Schneider, "Estimating the Size of the Danish Shadow Economy Using the Currency Demand Approach: An Attempt," Scandinavian Journal of Economics, 88 (1986) pp. 643-668; "Determinanten der Steuerhinterziehung der Schwarzarbeit im internationalen Vergleich," in Ch. Smekal and E. Theurl, eds., Stand und Entwicklung der Finanzpsychologie (Baden-Baden: Nomos Verlag, 1994) pp. 247-288; "The Shadow Economies of Western Europe," Journal of The Institute of Economic Affairs, 17, no. 3 (1997) pp. 42-48; "Ist Schwarzarbeit ein Volkssport geworden? Ein internationaler Vergleich des Ausma[Beta]es der Schwarzarbeit von 1970-97," in: S. Lamnek and J. Luedtke, eds., Der Sozialstaat zwischen Markt und Hedeonismus, Obladen (West Germany: Luedtke and Budrich publishing company, 1999) pp. 293-318; F. Schneider and D. Enste, "Shadow Economies Around the World: Sizes, Causes and Consequences," The Journal of Economic Literature (March 2000).

(3) The currency demand approach is explained in the section entitled "Empirical Data on the Development of the Austrian Shadow Economy"

(4) The currency demand approach was first used by P. Cagan, "The Demand for Currency Relative to the Total Money Supply," Journal of Political Economy, 66 (1958) pp. 303-328. He calculated a correlation between the currency demand and the tax pressure as one cause of the shadow economy for the United States for the period 1919 to 1955. Twenty years later P. M. Gutmann in "The Subterranean Economy," Financial Analysis Journal, 33, no. 6 (1977) pp. 26 - 27. Thirty-four used the same approach, but did not use any statistical procedures; instead he "only" looked at the ratio between currency and demand deposits for the years 1937 to 1976. Cagan's approach was further developed by V. Tanzi, "The Underground Economy in the United States: Estimates and Implications," Banca Nazionale del Lavoro Quarterly, 135 (1980) pp. 427-453; and in "The Underground Economy in the United States: Annual Estimates, 1930-1980," IMF-Staff Papers, 30 (1983) pp. 238-305. He estimated a currency demand function for the United States for the period from 1929 to 1980 to measure the shadow economy.

(5) See, for example, G. Garcia, "The Currency Ratio and the Subterranean Economy," Financial Analysis Journal, 34, NO. 6 (1978) pp. 64-66; D. Blades, "The Hidden Economy and the National Accounts," OECD Economic Outlook, (Paris, 1982) pp. 28-45; B.S. Frey and W. W. Pommerehne, "The Hidden Economy: State and Prospects for Measurement," Review of Income and Wealth, 30 (1984) PP. 1-23; J. Klovland, "Tax Evasion and the Demand for Currency in Norway and Sweden: Is there a Hidden Relationship?, Scandinavian Journal of Economics, 86 (1984) pp. 423-439; G. Kirchgassner, "Size and Development of the West-German Shadow Economy 1955-1980," Zeitschrift fur die gesamte Staatswissenschaft, 139 (1983) pp. 197-214; G. Kirchgassner, "Verfahren zur Erfassung des in der Schattenwirtschaft erarbeiteten Sozialproduktes," Allgemeines Statistisches Archiv, 68 (1984) pp. 378-405; H. Weck, Schattenwirtschaft: Eine Moglichkeit zur Einschrankung der offentlichen Verwaltung? (Bern: Verlag Peter Lang Publishing Company. 1983); E Schneider (1986, 1994). See also J. J. Thomas, "The Underground Economy in the United States: A Comment on Tanzi," IMF-Staff Papers, 33, no. 4 (1986) pp. 782-789; "Informal Economic Activity," London School of Economics Handbooks in Economics (London: Harvester Wheatsheaf, 1992); and "Quantifying the Black Economy: `Measurement Without Theory' Yet Again?" The Economic Journal, 109, no. 456 (1999) pp.381-389. And Susan Pozo, ed., Exploring the Underground Economy: Studies of Illegal and Unreported Activity, (Michigan: W.E. Upjohn, Institute for Employment Research, 1996).

(6) A. Isachsen and S. Strom, "The Size and Growth of the Hidden Economy in Norway," Review of Income and Wealth, 31 (1985) pp. 21-38.

(7) One (weak) justification for the use of the tax variable only is that this variable has by far the strongest impact on the size of the shadow economy in all studies known to the author. The only exception is a study in which the variable "tax immorality" has a quantitatively larger and statistically higher influence in the model than the direct tax share by B.S. Frey and H. Weck-Hannemann, "The Hidden Economy as an Unobserved Variable," European Economic Review, 26, no. 1 (1984) pp. 33-53. In a study of the U.S. shadow economy where data on various tax measures as well as on regulation, tax immorality and minimum wage rates are available, the tax variable has a dominating influence and contributes roughly 50-60 percent to the size of the shadow economy, by W. W. Pommerehne and E Schneider, "The Decline of Productivity Growth and the Rise of the Shadow Economy in the US," Working Paper, University of Aarhus (1985).

(8) G. Garcia, "The Currency Ratio and the Suberranean Economy," Financial Analysis Journal, 34, no. 6 (1978) pp. 64-66.

(9) T. Park, "Reconciliation between Personal Income and Taxable Income, 1947-1977," Bureau of Economic Analysis (Washington, DC, 1979).

(10) Klovland (1984).

(11) In applying the currency demand approach, I follow the procedure developed by Klovland (1984).

(12) Compare, for example, W. W. Pommerehne, "Steuerhinterziehung und Schwarzarbeit als Grenzen der Staatstatigkeit," Schweizerische Zeitschrift far Volkswirtschaft und Statistik, 119 (1983) pp. 261-284; and "Was wissen wir eigentlich uber Steuerhinterziehung?," Rivista Internazionale di Science Economiche e Commerciali, 35 (1986) pp. 230-262. Also, M.F. Hofreither and F. Schneider, "Die Erfassung der Schattenwirtschaft durch den Bargeldansatz--plausible Ergebnisse mittels unzulassiger Methode?" Wirtshaftspolitishe Blatter, 34 (1987) pp. 99-118; R. Neck, F. Schneider and M.F. Hofreither, "The Consequences of Progressive Income Taxation for the Shadow Economy: Some Theoretical Considerations," in D. Bos and B. Felderer, eds., The Political Economy of Progressive Taxation, (Heidelberg, 1989) pp. 149-176; F. Schneider (1994, 1999) and Tanzi (1999).

(13) Most studies stress that both average and marginal tax rates induce people to work in the shadow economy.

(14) For a theoretical underpinning to include this variable compare R. Neck and E Schneider, "The Development of the Shadow Economy Under Changing Tax Systems and Structures," Finanzarchiv, 50, NO. 3 (1993) pp. 344-369, where a simple theoretical microeconomic model of household behavior is formulated, where the household can participate in the official and shadow economy. Using comparative statistics, Neck and Schneider show that the complexity of the tax system affects participation in the shadow economy negatively, i.e. a more "complex" tax system implies, ceteris paribus, a smaller labor supply in the shadow economy.

(15) R.E. Wagner, "Revenue Structure, Fiscal Illusion and Budgetary Choice," Public Choice, 25 (1976) pp. 45-61.

(16) C.T.C. Clotefelter, "Public Spending for Higher Education: An Empirical Test of Two Hypotheses," Public Finance, 31 (1976) pp. 177-195; and "Tax Evasion and Tax Rates: An Analysis of Individual Returns," Review of Economics and Statistics, 65 (1983) pp. 363-373.

(17) There is an extensive literature on this type of "fiscal illusion" compare, for example, Pommerehne (1983, 1986).

(18) Regulatory activities are here defined in the areas of foreign labor, social security, working hours and other working conditions. The author is aware that this is an extremely crude measure, which is open to severe criticism, but I do not know any better measure which is available on a time series basis.

(19) In the following, only the most important tax rate and structure changes are discussed; for detailed information, see D. Bos, "Die finanzpolitische Entwicklung in Osterreich 1971-1974," Finanzarchiv, N.F. 33 (1975) pp. 305-338; B. Genser and R. Holzmann, "Die finanzpolitische Entwicklung in Osterreich 1985-1986," Finanzarchiv, N.F. 45 (1987) pp. 500-526; and "Die finanzpolitische Entwicklung in Osterreich 1987-1988," Finanzarchiv, N.F. 48 (1990) pp. 143-168.

Friedrich G. Schneider is the Chair in Economic Policy and Public Finance in the Department of Economics at Johannes Kepler University in Linz, Austria. For the past few years, he has been the President of the Austrian Economic Association and the Vice President for Foreign Affairs of Johannes Kepler University His main research interests are Economic Policy, the Shadow Economy, Environmental Economics, Privatization and Deregulation Policies. He has also authored several articles and publications, including: "The Shadow Economies of Western Europe," "The Informal Underground Economy" and "Shadow Economies Around the World: Size, Causes and Consequences."