Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
For Official Use STD/NA(2002)8 Organisation de Coopération et de Développement Economiques Organisation for Economic Co-operation and Development ___________________________________________________________________________________________ _____________ English - Or. English STATISTICS DIRECTORATE STD/NA(2002)8 For Official Use National Accounts EMPLOYEE STOCK OPTIONS AND HOLDING GAINS IN NATIONAL ACCOUNTS: AN EMPIRICAL PAPER FROM THE FINNISH HOUSEHOLD SECTOR POINT OF VIEW Paper prepard by Ilja Kristian Kavonius and Eeva Hamunen Statistics Finland OECD MEETING OF NATIONAL ACCOUNTS EXPERTS Château de la Muette, Paris 8-11 October 2002 Beginning at 9:30 a.m. on the first day [email protected] English - Or. English Document complet disponible sur OLIS dans son format d'origine Complete document available on OLIS in its original format STD/NA(2002)8 EMPLOYEE STOCK OPTIONS AND HOLDING GAINS IN NATIONAL ACCOUNTS: AN EMPIRICAL PAPER FROM THE FINNISH HOUSEHOLD SECTOR POINT OF VIEW I. Introduction 1. There has been discussion in recent years about the personal saving ratio in the US. For most of the first half of the 1980s personal saving exceeds 10 percent of disposable income, yet in 2000, it is only 2,8 percent of disposable income. There are various pathways along which the personal saving ratio has fallen in the US. Stock market capital gains are pushing down the measured ratio of personal saving. Another pathway is the conventional wealth effect according to which capital gains lead households to consume more. Since stock market capital gains do not appear in SNA93 income, the net effect from stock market wealth is reduction of SNA93 saving (the difference between income after tax and consumption).1 2. The role of durable consumer goods has also been examined in the US. These goods are treated as final consumption expenditure when purchased by households, yet the consumer may regard them as assets that are to be consumed gradually over a long service life. In this paper we are not going to concentrate to consider the setting the boundaries between consumer durable goods and investments.2 3. The US discussion has awoken economist to notice and discuss that there is similar phenomenon is in evidence also in other countries.3 A similar phenomenon can also be detected in Finland. From 1975 to 2000, the average household sector´s saving ratio was 4.2 percent, but in the year 2000 it was down to 0.3 percent. The importance of capital gains has grown enormously. In 1993, capital gains amounted to FIM 3.6 billion, but in the year 2000 they were FIM 24 billion. 4. In this paper, we shall study how capital gains and income from employee stock options influence household sector´s saving ratio and disposable income. We shall introduce three different concepts of disposable income. The first concept is the Finnish national accounts disposable income, which does not include employee stock options and capital gains, but includes their taxes. The second concept counts employee stock options and capital gains as income, and their taxes as expenditure. In this sense, the second concept of disposable income resembles the concept of disposable income in income distribution statistics. The third concept of disposable income does not include employee stock options, capital gains, or their taxes. The second and third concepts are, thus, diametrical opposites. We shall also study the effects of wealth on consumption and saving ratio. 1 Lusardi, Skinner and Venti 2001, pp. 1-4. Reindorf and Yan 2002, pp. 2. 2 Moulton 2001, pp 2-5. 3 See: Reindorf and Yan 2002. 2 STD/NA(2002)8 II. Sources and Methods 5. Where accounting conventions can affect the household sector´s saving ratio is when individuals sell appreciated stock and pay capital gains taxes. The gains realised have no effect on income, but the taxes paid have the effect of reducing income. Even under the extreme assumption that individuals do not increase their consumption when they realise capital gains, SNA93 saving would still decline. Capital gains tax affects the household sector´s saving ratio if the paid taxes are expounded to current taxes, not to capital transfers. In the US, as in Finland, the taxes paid are assumed to be current taxes, according to SNA93. 6. Pensions are treated differently in Finnish and US national accounts. In the US, the saving for pensions is based on private pension plans. Therefore, it influences the household sector´s saving ratio.4 In Finland, the saving for pensions is mostly statutory and organised by the public sector. Therefore, it does not influence the personal, but the government, saving ratio. Because of this the theoretical influence of saving for pensions is not discussed in this paper. 7. As already mentioned, the first disposable income concept is the Finnish national accounts disposable income, which does not include employee stock options and capital gains, but does include their taxes. The second concept counts income from employee stock options and capital gains as income, and their taxes as expenditure. The data on capital gains are based on annually published taxation levy statistics. In this paper, holding gains are measured using taxation data. Taxation data describes actual income flow. This is also symmetrical concept with the concept of Finnish Income Distribution Statistics. Basically, this means that all holding gains from the financial markets are included. In taxation, the concept of holding gains comprises the so-called “initial outlay assumption”. According to this assumption, a minimum of 20 percent of an assignment price has to be deducted as initial cost. If a vendor has had a property for more than 10 years, the minimum is 50 percent of an assignment price. This regulation was imposed in 1999. Previously, a minimum of 30 percent of an assignment price had to be deduced, and if the property had been acquired before 1989, then the initial outlay assumption proportion was 50 percent. The vendor can choose whether he/she wants to reduce the real initial price, or this initial outlay assumption. Because this initial outlay assumption exists, it is possible that a vendor’s holding gains are lower in taxation than they are de facto.5 However, it is difficult to estimate what is the correct amount of holding gains. Therefore taxation data is used as a proxy for holding gains in national accounts meaning. 8. In taxation, the holding gains from a dwelling in which a vendor and/or his/her family live are excluded from the concept of holding gains. The regulation is in force if the vendor and/or his/her family have lived in the dwelling for more than two years. The aim of this regulation is only to tax speculative acquisition of dwellings. The concept, what we are looking for is realised, speculative holding gains.6 9. The phenomenon of employee stock options is quite a recent phenomenon. SNA93 does not give guidance for their treatment. In many countries, income from stock options has been assumed to represent compensation of an employee. In Finland, employee stock options have been expounded to capital gains in national accounts. Employee stock options are not a cost element in company bookkeeping in Finland. Because option benefits do not constitute a cost for the employer (except for the social contributions payable to the Social Insurance Institution, SII), qualifying them as 4 See more the US pension system: Reindorf and Yan 2002, pp. 6-9. 5 Henkilöverotuksen käsikirja (The Handbook of Personal Taxation) 2001, pp. 6:14d-6:15. 6 Henkilöverotuksen käsikirja (The Handbook of Personal Taxation) 2001, pp. 6:26-6:27. 3 STD/NA(2002)8 wages and salaries would distort the operating surpluses of corporations and the national economy. 7 However, income from employee stock options affects extensively basic economic indicators, such as household disposable income and household sector´s saving ratio. Holding gains Assumed taxes Options Assumed taxes Social contributions 1993 3,583 896 1994 4,327 1,082 1995 4,096 1,024 1996 4,214 1,180 1997 7,709 2,159 1998 1999 10,726 17,077 3,003 4,782 996 5,200 578 3,016 15 78 2000 24,027 6,968 5,929 3,439 89 Table 1. Holding gains and options and their assumed taxes and social contributions used in the calculations (FIM million). Source: The calculations of Ilja Kristian Kavonius. Taxation levy statistics 1980-2000, Finnish Tax Administration 2001. 10. Before 1998, employee stock options were a fairly marginal phenomenon in Finland, with little effect on economic indicators. Since 2000, data concerning employee stock options have been published in annual taxation levy statistics. Before 2000, there was no direct source of information on them, and their amounts were difficult to estimate. The amounts of employee stock options for 1999 and 1998 are based on various estimations. According to an inquiry conducted by Business Statistics, employee stock options totalled FIM 996 million in 1998 and FIM 5.2 billion in 1999. Comparing these figures to other estimations based on taxation data shows that they can be considered reliable. According to taxation levy statistics, employee stock options amounted to FIM 5.929 billion in 2000. Table 1 shows the amounts of options and holding gains for the years 19932000 used in this paper. 11. The third concept of disposable income excludes employee stock options, capital gains, and their taxes. The taxation of capital gains has changed several times between 1987 and today. From 1987 to 1993, capital gains from financial assets were also exempt from taxation if their holding period exceeded five years. Large capital gains were partially taxable even after that, but the tax could be avoided by realising the gains over several years. Following the method Tarmo Valkonen used in his study, we shall also assume in this paper that the tax rate was zero before the reform.8 12. After the reform, the tax rate on holding gains was 25 percent from 1993 to 1995. In 1996, the rate rose to 28 percent and in 2000 it was 29 percent. From 1998 to 2000, employee stock options were taxed by progressive income taxation. It is safe to assume that the highest income classes received most of the employee stock options. During the same period, the highest marginal tax rate was 60 percent. In this paper, it is assumed that the average tax paid on employee stock options was 58 percent. Option benefits do not constitute a cost for the employer except for the social contributions payable to the SII, which average 1.5 percent. Table 1 shows the amounts of taxes and social contributions on holding gains and employee stock options as assumed in this paper. 13. A main reform of the Finnish tax system was carried out during 1987–1993. Among the main elements of the reform were a separation of taxation on earned income from that on capital income and a major revision of corporate income tax. The separation of taxation on different kind of income is not yet finalised. Housing loan interest expenditure is deductible in income taxation, but if there is not enough taxable capital income, expenses can also be deducted in earned income taxation (with 7 Compare: Hamunen and Weckström 2000. Moylan 2000. Postner 2000. 8 Valkonen 1999, pp. 44-45. 4 STD/NA(2002)8 an upper limit). Another exemption is that income together with wealth is taken into account when certain means tested transfers and payments for public services are determined.9 14. The tax reform of 1987–1993 causes a major problem with the comparability of time series. As we have seen above, the concept of holding gains changed fundamentally during this time period. The new concept of holding gains can be regarded as quite exhaustive. Holding gains, which in 1992 amounted to less than FIM 220,000 were tax-free. After the tax reform this minimum limit was abolished. Therefore, the concept of holding gains has changed fundamentally and it is, thus, also impossible to make reliable estimations going back to the years before 1993 of the extent to which the old concept of holding gains covers the new one. The Ministry of Finance has estimated that in 1991 holding gains amounted to FIM 4 billion, while the annually published taxation levy statistics put them at FIM 1 billion. It can be concluded that the Finnish concepts of holding gains are not comparable before and after the year 1993. This is also the reason why Tables 1, 2 and 3 only cover the years 1993 to 2000. On the other hand, the economic effects of holding gains and employee stock options were much more extensive after 1993 than before it. Nevertheless, the 1980–1993 time series gives a rough picture of the development trend of holding gains.10 III. The effect of capital gains and employee stock options 15. As can be seen from Figure 1, before 1997 holding gains and employee stock options did not have a substantial effect on real disposable income. As already stated, the concept of holding gains changed in 1993, rendering it incomparable. After 1997, the difference in the used concepts is so obvious that it is reasonable to presume that income from the financial markets then affected real disposable income for the first time. Assuming that the new concept of holding gains is about four times as exhaustive as the old one, as the estimate of Ministry of Finance indicates, then holding gains in 1990 were FIM 6 to 7 billion. In 2000, holding gains and income from employee stock options amounted to roughly FIM 30 billion. The indicator for 1990 is, of course, rough but allows us to form an approximate picture of the situation. 9 Valkonen 1999, pp. 46-48. 10 Pääomatulojen verouudistuksen taloudellisia vaikutuksia sevittäneen työryhmän muistioita 1992:21, pp. 8-13. 5 STD/NA(2002)8 130,00 120,00 110,00 100,00 90,00 80,00 00 20 8 99 19 7 1 99 6 1 99 5 1 99 4 1 99 3 99 99 1 92 1 90 89 91 19 19 19 87 86 88 19 19 19 84 83 85 19 19 19 81 82 19 19 19 19 80 70,00 Disposable income 1 fp (same as in national accounts) Disposable income 2 fp (the concept of national accounts + options + holding gains) Disposable income 3 fp (the concept of national accounts - the taxes of holding gains and options) Figure 1. Volume indices of the three different concepts of disposable income. Source: The calculations of Ilja Kristian Kavonius. National accounts 1980-2000, Statistics Finland 2002. Taxation levy statistics 1980-2000, Finnish Tax Administration 2001. 16. The disposable income that includes holding gains and income from employee stock options was 8.1 percent higher than the disposable income of national accounts. Approximately 20 percent of this difference come from employee stock options and the rest from holding gains. The disposable income that does not include holding gains, employee stock options or their taxes was 2.8 percent higher than in the year 2000. One third of this difference is accounted for by employee stock options and the rest by holding gains. 1994 1995 1996 1997 1998 Disposable income 2 fp (the concept of national accounts + options + holding gains) 1999 2000 Effect of options 1.19 0.18 0.00 0.00 0.00 0.00 0.30 Effect of holding gains 0.26 -0.08 0.04 1.15 0.91 1.81 1.83 Disposable income 3 fp (the concept of national accounts – the taxes on holding gains and options) Effect of options 0.00 0.00 0.00 0.00 0.18 0.72 0.11 Effect of holding gains 0.07 -0.02 0.05 0.32 0.26 0.51 0.57 Table 2. Contributions of holding gains and options and their assumed taxes and social contributions to the volume growth of disposable income. Source: The calculations of Ilja Kristian Kavonius. Taxation levy statistics 1980-2000, Finnish Tax Administration 2001. National accounts 1994-2000, Statistics Finland 2002. 17. As we can see from Table 2, in 1999 and 2000 the contribution of holding gains to the volume growth of disposable income would have been almost two percent, and even that of their taxes more than 0.5 per. In 1995, the contribution of holding gains would have been slightly negative, because holding gains fell from FIM 4.3 billion to FIM 4.1 billion. The contribution of employee stock options to the volume growth of disposable income would have been the most extensive in 1999 because options went up from FIM 1 billion to FIM 5.2 billion. Then the contribution of employee stock options would have been 1.2 percent and that of their taxes 0.7 percent. It is important to note 6 STD/NA(2002)8 that in that year the combined effect (i.e. effect of employee stock options and holding gains) on the growth would have been three percent, and even that of their taxes 1.2 percent. 18. Holding gains and employee stock options have a much more notable influence on saving and on the saving ratio than on disposable income. This is, naturally, rather logical since options and holding gains are received by the highest income groups. On the one hand, the propensity to consume is much smaller in high-income groups than in low income ones. On the other hand, saving is a residual of disposable income and individual consumption expenditure. Therefore, it is quite obvious that even a small increase in income has an influence on the household sector´s saving ratio. 12,0 10,0 8,0 6,0 4,0 2,0 19 80 19 81 19 82 19 83 19 84 19 85 19 86 19 87 19 88 19 89 19 90 19 91 19 92 1 99 3 1 99 4 1 99 5 1 99 6 1 99 7 1 99 8 19 99 20 00 0,0 Personal saving rate 1 cp (same as in national accounts) Personal saving rate 2 cp (the concept of national accounts + options + holding gains) Personal saving rate 3 cp (the concept of national accounts - the taxes od holding gains and options) Figure 2. Household sector´s saving ratios of the three different concepts of disposable income. Source: The calculations of Ilja Kristian Kavonius. National accounts 1980-2000, Statistics Finland 2002. Taxation levy statistics 1980-2000, Finnish Tax Administration 2001. 19. As Figure 2 shows, before 1997 the trends in the saving ratios were rather similar. During the recession, income from the financial markets was naturally quite insignificant, and the saving ratio before 1993 is not fully comparable with that of today. Holding gains and other income from the financial markets have increased considerably in the last few years. The Finnish economy experienced a deep recession in the early 1990s, and it was impossible to realise significant holding gains from stock markets then. Until the end of the 1980s, the financial, capital and stock markets were somewhat regulated in Finland and investing in stocks was, therefore, not as common as it is today. The deregulation of the financial, capital and stock markets can be seen in Figure 2. In 1988, the household sector´s saving ratio was 0.2 percent, while the household sector´s saving ratio inclusive of holding gains was approximately one percent. Had the taxation concept of holding gains been the same as it is today, the difference would have been much more significant than that shown in the figure. Nevertheless, it would not have been as significant as it is today. 7 STD/NA(2002)8 1993 1994 1995 1996 1997 1998 Household sector´s saving ratio 2 cp (the concept of national accounts + options + holding gains) Effect of options 0 0 0 0 0 0.20 Effect of holding gains 1.10 1.50 1.30 1.30 2.20 3.00 Household sector´s saving ratio 3 cp (the concept of national accounts – the taxes of holding gains and options) Effect of options 0 0 0 0 0 0.10 Effect of holding gains 0.30 0.30 0.30 0.40 0.60 0.90 1999 2000 1.20 1.50 4.40 6.00 0.80 0.90 1.30 1.80 Table 3. Contributions of holding gains and options and their assumed taxes and social contributions to household sector´s saving ratios (%). Source: The calculations of Ilja Kristian Kavonius. Taxation levy statistics 1980-2000, Finnish Tax Administration 2001. National accounts 1994-2000, Statistics Finland 2002. 20. In 2000, the household sector´s saving ratio was 0.3 percent according to Finnish national accounts. The saving ratio of disposable income inclusive of holding gains was 7.7 percent. The effect of employee stock options was 1.5 percent and that of holding gains 6.0 percent. The saving ratio of disposable income exclusive of holding gains and employee stock options or their taxes was 3.0 percent. The influence of holding gains was 1.8 percent and that of employee stock options 0.9 percent. The different effects of holding gains and employee stock options are shown in Table 3. IV. The wealth effect 21. In the US, aggregate estimates indicate that an additional $1 of stock market wealth raises the level of aggregate spending by around three cents. More recent estimates suggest that the whole effect of total wealth on consumption is closer to two cents for each $1 of total wealth. Estimates, however, tend to be different across different time periods. In the US, sharp increase in stock market wealth has been suggested to be one of the main culprits for the decline in saving, and a good deal of media attention has been focused on the stock market wealth effect on consumption. The trends of the Dow Jones Industrial Average Index and the private saving ratio move closely together and in opposite directions in the US.11 22. A similar phenomenon is detectable in Finland. At the end of the 1980s, the household sector´s saving ratio declined, while the values of stocks and dwellings rose. In the late 1990s, after the deep recession, the household sector´s saving ratio was also falling while the values of stocks and dwellings were rising. During the recession, the household sector´s saving ratio was exceptionally high, but then plunged very steeply. As we have seen above, one of the main reasons for this was the way holding gains, income from employee stock options and their taxes were treated in national accounts. In all probability, however, the wealth effect also influenced household sector´s saving. 11 Lusardi, Skinner and Venti 2001, pp. 7-8. 8 STD/NA(2002)8 12 10 8 6 4 2 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 Figure 3. Regression coefficient (vertically, %) between consumption and gross wealth in twenty income classes (horizontally). Source: The calculations of Markku Säylä. Survey on the Financial Situation of Finnish Households 1998, Statistics Finland 2002. 23. In this paper, we have presented very rough estimations of the wealth effect. The estimations are based on the Survey on the Financial Situation of Finnish Households 1998, which covers the wealth, consumption and income of households. The survey data put the regression coefficient between consumption and gross wealth at 5.8 percent. This calculation includes households whose gross wealth is under FIM 2.5 million. Since the study is a survey, imposition of these kinds of limitations is necessary. The calculation does not take into account the effect of income, either. It is safe to assume that the gross wealth of households with high incomes would exceed the set limit.12 24. In the following calculations households have been divided into twenty groups according to their income. The division produced twenty households groups for which average consumption and average wealth were calculated. We then proceeded to calculate the regression coefficient between average gross wealth and consumption for each income group. The results, which are very logical, can be seen in Figure 3. The wealth effect is greater in low-income groups than in high income ones. If, for example, a family belonging to a low income group owns a dwelling, it is quite oblivious that the money they do not have to spend on rent will most likely go to other consumption. It is interesting to see that this effect is the most dramatic in the third decile and the least dramatic in the fifth decile. The average wealth effect for the whole economy is 4.2 percent. 25. The method is naturally very rough, but it is interesting to discover that the wealth effect in Finland is very similar to that in the US. We have not also calculated confidence intervals, since on the one hand we wanted to give in this point only a rough idea on wealth effect in Finland. On the other hand the calculations of confidence intervals on this material is quite demanding process. Thus we have left it to forthcoming papers. 26. The wealth of Finnish households is mostly in dwellings, and it is logical that owning a personal dwelling makes it possible to consume more. In 1998, dwellings accounted for two thirds of Finnish households’ wealth, while stocks only made up six percent of it. It is essential to bear in mind that 12 Survey on the Financial Situation of Finnish Households 1998 and the calculations of Markku Säylä. 9 STD/NA(2002)8 four years earlier this proportion for stocks was only three percent. The role of the financial markets in this effect is likely to be very small. This finding is as anticipated. Takala et al., for example, have obtained similar results. This theme will be examined more closely in forthcoming papers.13 V. Conclusions 27. In this paper we have discussed the household sector´s saving ratio and the way it is used to describe actual saving. We have introduced three different concepts of disposable income. The first concept is the Finnish national accounts disposable income, which does not include income from employee stock options and capital gains, but includes their taxes. The second concept counts stock options income and capital gains as income, but treats their taxes as expenditure. The third concept of disposable income does not include employee stock options, capital gains, or their taxes. 28. We have observed that income from the financial markets has been influencing the saving ratio increasingly in recent years. The most important effect comes from the way employee stock options and holding gains are treated in national accounts. These incomes have grown increasingly important in recent years. 29. What do the different saving ratios describe? The saving ratio of the Finnish national accounts describes the transactions and incomes of the real economy, as it should. The saving ratio exclusive of options or their taxes describes perhaps most closely the behaviour of a normal Finnish family. Most Finnish families do not make their living on stock markets. The third concept, which includes options and holdings gains as well as their taxes, describes perhaps the behaviour of the whole economy and the behaviour of all Finnish households. Here we end up facing the basic question: how do the different indicators describe the essential nature of economy – what do we want to describe? 30. We have also studied the effects of wealth on consumption and saving ratio. The presented estimations are very rough and we shall examine this theme further in forthcoming papers. Nevertheless, it is quite obvious that the financial markets have a slight effect on consumption, and that this effect is continuously increasing. The major effect on consumption can be attributed to the ownership of dwellings. 13 Compare for instance Clapham, Hyytinen & Takala 2001. 10 STD/NA(2002)8 References Clapham , Erik; Hyytinen, Ari and Takala, Kari: Household Wealth, Credit and Consumption: Evidence from Finland and Sweden. Working paper. Hamunen, Eeva and Weckström, Kaisa: Share Options: Finnish Experiences and Prospects. Agenda item 10, OECD Meeting of National Accounts Experts. STD/NA(2000)31, OECD 2000. Henkilöverotuksen käsikirja (The Handbook of Personal Taxation), Finnish Tax Administration 2001. Lusardi, Annamaria; Skinner, Jonathan and Venti, Steven: Saving Puzzles and Saving Policies in the United States. NBER Working Paper No.w8237, National Bureau of Economic Research 2001. Moulton, Brent R.: Issues in Measuring Household Net Saving and Wealth. Agenda item 2, OECD Meeting of National Accounts Experts. STD/NA(2001)29, OECD 2001. Moylan, Carol: Treatment of Employee Stock Options in the U.S. National Economic Accounts. Agenda item 10, OECD Meeting of National Accounts Experts. STD/NA(2000)34, OECD 2000. National accounts 1980-2000, Statistics Finland 2002. National Income and Product Accounts Tables, Bureau of Economic Analysis, U.S. Department of Commerce, http://www.bea.doc.gov/bea/dn/nipaweb/index.asp Postner, Harry H.: Employee Stock Options and National Accounts: Using Economic Theory to Clarify Measurement Issues. Agenda item 10, OECD Meeting of National Accounts Experts. STD/NA(2000)35, OECD 2000. Pääomatulojen verouudistuksen taloudellisia vaikutuksia selvittäneen Valtiovarainministeriön työryhmämuistioita 1992:21, Helsinki 1992. työryhmän muistio. Reindorf, Marshall B. And Yan, Xiaoyi: Alternative Measurese of Household Saving in the US, the UK and Canada. Paper prepared for 27th General Confrence of The International Association for Research in Income and Wealth. Stockholm, Sweden. August 18-24, 2002. http://www.iariw.org/papers/rein.pdf Säylä, Markku: Survey on the Financial Situation of Finnish Households 1998 (Kotitalouksien varallisuus 1998), SVT; Income and Consumption 2000:26, Statistics Finland 2000. Taxation levy statistics 1980-2000, Finnish Tax Administration 2001. Valkonen, Tarmo: The Finnish Corporate and Capital Income Tax Reform – A General Equilibrium Approach. The Research Institute of the Finnish Economy Series A 29. Taloustieto Oy; Helsinki 1999. 11