Black economy: different approaches to the estimation of its size.
On the fringes are people who earn small amounts of income from which income tax is not deducted and who conveniently 'forget' to declare this income to the tax authority at the end of the year. Four main issues relating to the black economy are: (i) Size and measurement ii) Participation (iii)Economic consequences (iv) Policy
Some economists have devoted a great deal of attention to trying to measure the size of the black economy. Much of the early work on this front was undertaken in the USA by Gutmann, Feige and Tanzi. However interest in measurement has spread to other countries. To most people to question 'how big' is the black economy? seems absolutely fundamental to further analysis of the topic. After all, if it is large, then surely it poses a threat to the legitimate economy, whereas if it is small it can be safely ignored.
There are various direct and indirect methods used to estimate the size of the black economy. The most frequently used are those based on accounting and monetary indicators, and in particular the amount of currency in circulation. Other method is fiscal approach. To provide a definite path to the government in the implementation of different policies, the evaluation of the black economy with accurate information about various macroeconomic variables is imperative.
So far the work done in the accounting sphere comprises of independent estimates of income and expenditure. The discrepancy between the income and expenditure gives the clue to the size of hidden economy. There are two approaches conducted in this context. The first approach deals with the comparison of income and expenditure at the macro level. The second deals with the evaluation of the discrepancy between income and expenditure at the micro level of households. Discussion of each approach is as follows:-
a) The Macro Approach: This approach adopts two methods: first, being the one in which the comparison of national account estimates of income with income estimates built of from tax returns. The difference between the two represents the estimates of income not reported to the tax authority. Such comparisons have been undertaken by Park (1979) for United States which measure the 'unexplained difference', or 'residual error' between the estimates of personal income and that of adjusted gross income on the basis of a sample of tax returns.
O'Higgins (1980) finds for United Kingdom in 1970's an opposite development of discrepancy between the aggregate amount of income in national accounts and aggregate income estimates based on adjusted tax returns. The officially measured national income increased three folds between 1970 and 1978, the unexplained difference increased nine folds. However, the absolute size of hidden economy ranged between 2.5 per cent to 3 per cent of GNP, in 1978 (Macfee 1980).
The second method focuses on the estimation of discrepancy between the GDP from expenditure side and GDP on the income side. The estimate of the expenditure being higher than those of income for the existence of the submerged economy. Studies carried out in various countries indicate that in Anglo Saxon and Scandinavian countries the discrepancies are small of up to 10 per cent of GNP, while in Latin American countries larger discrepancies are observed around 20 per cent of GNP and more. For United States the measure of hidden economy has been declining since postwar period as is suggested by Park (1979) from 9.4 percent in 1948 to 4 per cent in 1977.
b) The Micro Approach: The residual factor evaluated from the differences between expenditure and income is also done from household side. The method adopted for the estimation is through survey techniques. The survey results are extrapolated to give an estimate of the hidden economy. Research is underway in this direction in the United States. However, at micro level attempts have been made to calculate the difference between expenditure and income for specific types of households occupational groups, and income classes and to compare them with the unexplained residual at the macro level. O'Higgins (1980) arrived at an estimate for the United Kingdom in 1978, the private households headed by self employed persons have not reported Pound Sterling 2.10 billion in their answers to the consumer surveys. This correspond to 2 per cent of the national income in Britain. The two approaches mentioned above have been criticized by Macfee (1980) as follows:-
a) There are errors in both estimates of aggregate income.
b) There are errors due to the difference of statistical coverage.
c) The national income estimates are not always completely independent of the tax data based income estimates i.e. income not included by tax authorities may also not appear in the national income data. Besides this it is also observed that, there are many income creating activities which escape the measurement like income generated from bartering common in Latin American countries, Sweden and most of Asian countries. It corresponds to 2 per cent of the National income of Britain.
Fiscal Approach: Kaldor (1956) Wanchoo Committee (1971) and Chopra (1982) had adopted the fiscal methodology to estimate the size of unreported income due to income tax evasion. Their method requires:-
1) Estimation of gross personal income accruing to the households.
2) Estimation of distribution of income for households is obtained. It is converted to the distribution of income by earner.
3) Major exclusions and deductions are made from the earner wise distribution of income to get the estimate of the income that should have been taxed.
4) This estimated total is compared to the with the income that is actually assessed to tax to get the extent to evasion occurring in the economy.
Theoretical treatment to tax evasion has been given by Allingham and Sandmo (1972) in which a model for income tax evasion is discussed in the context of proportional tax system. The tax payer is assumed to be a moral and risk averse, utility maximizing individual. Srinivasan (1973) has used a more general case in which he allows for a progressive tax system and assumes the tax payer to be risk neutral. Tax structure in his paper is assigned the role of revenue collection and avoids the role of altering the distribution of wealth. Both srivasan and Allingham and Sandmo have found that the proportion of income reported falls as the probability of detection decreases. This means that the higher the income of the people the more difficult it is to control.
Mork (1975) on the other hand has carried out empirical analysis for evasion of income taxes. His analysis is led to the conclusion that as true income of the masses increases the proportion of unreported income to the tax authorities increases. This finding is also supported by theoretical models put forward by Srinivasan, and Allingham and Sandmo.
The monetary approach to the measurement of the black economy dates back to the work done by Peter Gutmann (1977). His analysis is subject to a number of assumptions. The first, activities in the underground economy are carried out through the use of currency holding. Second, the underground economy is the result of heavy imposed restriction and high taxes by the government. Third, the ratio of currency to demand deposits are only influenced by changes in taxes and restrictions introduced since 1973-41. Fourth and final that there was no underground economy from the period of 1937-41. By assuming that the level of demand deposits in 1976 were normal, he calculated the extra currency due to the underground economy then multiplied it by the velocity of money to get the estimates of the black economy. The velocity of money was found by taking the ratio of (legal) GNP to legal money (i.e. excluding extra currency).
Feige's analysis (1979) is based on the Irving Fishers Quantity Theory of Money written as, MV = PT; where M includes currency and demand deposits, P is the average price level of all goods exchanged and not newly created goods and T is the index of physical volume of transactions. By estimating MV, PT is found out and then it is divided by nominal GNP.
He assumed that there was no underground economy in 1939 and calculated the ratio of PT/GNP. After that he estimated the PT for the years of 1976 and 1978 and divided the result by ratio of 1939, by doing so he was able to derive the nominal GNP for the years of 1976 and 1978. The difference between the indirectly obtained GNP's and the one's in the national account were compared to get the size of the underground economy.
Vito Tanzi (180, 1983) has refined the estimates of the black economy using the current ratio, which in turn is based on an earlier insight of the work done by Cagen (1958). Cagen was interested in explaining the long run behaviour of the currency ratio defined as currency to money supply over the period from 1875 to 1955. He identified a number of factors that were expected to influence this ratio. They are as follows:-
1) The opportunity cost of holding currency.
2) Expected real income per capita.
3) The volume of retail trade (cash being more acceptable in retail trade).
4) The volume of travel per capita (cash being more acceptable between strangers).
5) The degree of urbanization.
6) The rate on tax on transactions
The last variable provides the link between the currency ratio and the tax evasion. According to Cagen (1958, pp. 312), "Some people evade taxes by making as many transactions as possible with the currency and not reporting to the tax collector". He therefore postulates a direct positive relation between the income tax rates and the currency ratio. This hypothesis was confirmed using multiple regression analysis for the period of 1919-55. The dependent variable was the ratio of currency to M2 and the independent variables were:
1) The interest rate paid on bank deposits
2) Expected real income per capita (based of previous income levels with exponentially declining weights).
3) The annual percentage of percentage of personal income collected for income taxes.
Tanzi (1980) adopted the idea linking the currency ratio to tax rates and used to drive alternative estimates of the size of black economy for USA. According to Tanzi (1980, 1983) the reason for people to engage in the underground economy is the existence of taxes. His approach focuses on the specification of a demand for currency equation to be able to see the change in the tax level on the demand.
Tanzi used ratio of currency to M2 being a function of share of wages and salaries to national income and interest rate on time deposits and a tax variable taken in three forms, that is, ratio of personal income taxes to personal income net of transfers, top bracket statutory tax rate and weighted average tax rate on interest income. He then evaluated the predicted level of currency holdings for given level of M2 with taxes and without taxes. The difference between the two was multiplied by the product of income velocity of money gave that gave the extent of the underground economy.