In analyzing the President's budgetary proposals for 2006, the Congressional Budget Office considered how those policies--in comparison with the policies incorporated in CBO's baseline--would affect the economy. CBO concluded that under the policies in the President's budget, economic output would probably be slightly lower, on average, over the 2006-2010 period but somewhat higher over the 2011-2015 period than it would be under the policies in CBO's baseline. However, CBO's analysis suggests that the differences are likely to be small, affecting output by an average of less than one-half of 1 percent over the full 10-year period.

The small size of the effects on output stems in part from the small magnitude--relative to the overall economy--of the proposals' budgetary impact. For example, in CBO's estimation, revenues from 2006 to 2010 under the President's proposals--excluding economic effects--would be lower by 0.14 percent of cumulative gross domestic product than they would be under CBO's baseline. Spending, including interest on government debt, would be higher by 0.01 percent of GDP. Moreover, various proposals in the budget have offsetting effects: some tend to imply greater output, and some tend to imply less.

The macroeconomic effects of the proposals could in turn alter their budgetary cost. Under CBO's economic assumptions, the President's budgetary proposals would increase the cumulative deficit for the 2006-2010 period, CBO estimates, by $104 billion (0.1 percent of cumulative GDP), compared with the deficit under the baseline's policy assumptions. If the budgetary effects of the economic changes resulting from those proposals were included in the estimate, the projected increase in the cumulative deficit over that period would range from as much as $165 billion to as little as $105 billion (see Figure 2-1 and Table 2-1).

Figure 2-1.

Source: Congressional Budget Office.

Notes: The estimates in the figure reflect the proposals' supply-side effects on the economy but exclude demand-side economic impacts, as explained in the text. A negative change indicates an increase in the cumulative deficit relative to CBO's baseline.

CBO's analysis used the following models (which are described in the text): (A) "textbook" high model, (B) "textbook" low model, (C) closed-economy life-cycle model with lower government spending after 2015, (D) closed-economy life-cycle model with higher taxes after 2015, (E) open-economy life-cycle model with lower government spending after 2015, (F) open-economy life-cycle model with higher taxes after 2015, (G) infinite-horizon model with lower government consumption after 2015, (H) infinite-horizon model with higher taxes after 2015, (I) Macroeconomic Advisers' model, and (J) Global Insight's model.

a. Because this model is designed primarily to capture business-cycle developments, which are hard to predict beyond a few years, CBO did not compute an estimate for the 2011-2015 period.

Table 2-1.

For the years 2011 to 2015, the President's proposals, as assessed using the baseline's economic assumptions, would increase the cumulative deficit by $1,497 billion (1.7 percent of cumulative GDP). If the budgetary effects of the economic changes resulting from those proposals were also considered, the projected increase in the cumulative deficit over that period would range from as much as $1,523 billion to as little as $1,366 billion.
 

How Fiscal Policy Affects the Economy

Budgetary policies can affect the economy in a variety of ways. Changes in the tax rates that people face on their income can affect incentives to work and save; government spending on goods and services can reduce the resources available for investment; and spending and tax policies can influence the overall level of demand in the economy.⁽¹⁾ Those factors and other possible economic influences can be broadly divided into long-run supply-side effects and short-run demand-side effects.

The economy's underlying potential to produce goods and services depends on the size and quality of both the labor force and the stock of productive capital (such as factories and information systems) as well as on the level of technological know-how. Analysts refer to long-term changes in those three determinants of potential output as supply-side changes because they alter the quantity of goods and services that the economy is capable of supplying on a sustainable basis. Supply-side changes have a lasting effect on the economy.

In the short run, however, economic output can deviate from its potential level, as the total demand for goods and services moves above and below that level, causing employment to rise and fall and the stock of capital to be used more or less intensively. Those movements are referred to as demand-side, or cyclical, variations. Unlike movements on the supply side of the economy, cyclical changes are temporary--built-in corrective forces that usually tend to move the economy back toward the sustainable potential level determined by the supply side.

Both supply-side and demand-side economic developments depend on the choices of millions of individuals about what and how much to buy, how much to save and what assets to hold, and where and how much to work. The government plays a crucial role in establishing the legal and institutional framework within which the economy operates and an overall level of government spending and taxes. Once that general structure is in place, however, changes in government spending and tax policies influence individuals' choices only to a limited degree. Decisions by businesses, personal circumstances, and preferences play a much larger role in economic performance than do modest changes in the federal government's fiscal policies.

Supply-Side Effects

The supply-side effects of the President's budgetary proposals could include influences on the quantity and quality of labor, the size and composition of the capital stock, and technological progress. Changes in any or all of those factors are capable of altering potential output.

The Quantity and Quality of Labor. The overall quantity and quality of labor in the economy is an important determinant of potential output. Put simply, a long-term increase in the overall number of hours worked raises the economy's potential to generate output. Moreover, increases in workers' educational attainment, level of training, and experience or degree of effort on the job raise the quality of each hour worked, which will also increase potential output. Some analysts argue that certain policies in the President's budget--for example, making current tax rates permanent--will ultimately affect the quality of labor. However, the channels through which budgetary policies influence that supply-side variable and the pace at which such effects occur are not well understood. For that reason, CBO's analysis concentrated on the hours of labor supplied, which the President's proposals would affect in two main ways.

First, several provisions, such as the extension of the child tax credit and the exemption from taxation of some dividend and capital gains income, would increase people's after-tax income but not significantly change the marginal tax rates on income resulting from labor. (In general, the marginal rate is the rate on the next dollar of income.) A rise in after-tax income without a change in marginal rates tends to reduce the number of hours of labor that people supply because they can maintain the same standard of living with less work.

Second, some of the President's proposed policies, such as the extension of the marginal tax rates enacted in the Economic Growth and Tax Relief Reconciliation Act of 2001, would both increase the compensation after taxes for each additional hour of work and boost after-tax income overall. Evaluating how that kind of change affects the number of hours people work is complicated by the fact that the policy has opposing effects: people earn more for each extra hour they work, which tends to encourage them to supply more labor, but they can also work fewer hours and earn the same after-tax income, which tends to discourage work. Most studies find that, on balance, reductions in marginal tax rates such as those that the President is proposing will increase the hours of labor supplied, primarily because they will draw secondary earners (for example, the spouse of a household's primary breadwinner) into the labor force.

To summarize as a single number the changes proposed by the President in the schedule of marginal tax rates, CBO estimated the impact that those changes would have on the effective marginal tax rate on labor income--the rate at which the average additional dollar of compensation for labor is taxed (see Table 2-2). The largest changes in rates occur for the years 2011 to 2015, because the President proposes to make permanent various reductions in taxes that are scheduled to expire after 2010. For those years, the average effective marginal tax rate on labor income would be between 4 percent and 5 percent lower under the President's proposals than under the current-law policies incorporated in CBO's baseline.

Table 2-2.

The President's proposals might also influence the level of the capital stock (discussed in the next section), which could change people's productivity and wages and thus affect the hours of labor supplied. That is, if the proposals reduced investment, the stock of productive capital would be smaller and wages would be lower, which would discourage work. (Increasing investment would imply the opposite results.)

Another way in which the President's proposals could affect the hours worked in the economy would be by changing people's expectations about future policies. Under the President's proposals, the cumulative federal budget deficit over the 2006-2015 period would grow. That rise could lead people to expect that at some time after that period, fiscal policy would have to change to finance the increase in the federal government's interest payments on the money that the government had borrowed to cover the bigger deficit. Either taxes would have to be raised or spending cut.⁽²⁾

If people expected to pay more in taxes or to receive fewer services or smaller transfer payments (such as Social Security benefits), they might try to work and save more now so they would have more resources to compensate for the larger burden in the future. In addition, if people expected to face higher tax rates on their income from labor in the future, they might try to work more before the rates went up and then work less when the rates were higher.

But it is difficult to gauge the degree to which such foresight influences people's decisions. Also unclear is the time horizon that people consider in making plans and the future changes in policy that they actually expect. To illustrate the importance of those factors, CBO used various assumptions in its analysis about the extent of people's foresight and the expectations they might have about future policies and assessed the differences in results.

In sum, the President's budgetary proposals would have relatively small effects, on average, on the number of hours worked over the 2006-2010 period, CBO estimates, and those effects could be positive or negative. Over the 2011-2015 period, however--when the reductions in marginal tax rates were largest relative to CBO's baseline policy assumptions--the proposals would probably increase the hours of labor supplied.

The Size and Composition of the Capital Stock. The President's budgetary policies would influence the size of the capital stock primarily by affecting consumption and therefore investment. The President's proposals would directly alter government consumption of goods and services relative to the level in CBO's baseline. In 2006 and 2007, the proposals imply an increase in consumption relative to the baseline level, which would tend to lessen investment in private capital by reducing the resources available for that purpose. However, from 2008 through 2015, the proposals imply a reduction in government consumption, which would tend to boost private investment.⁽³⁾

The President's budgetary policies would also produce offsetting influences on private consumption. The reductions in taxes that the budget proposes would increase after-tax income, which would tend to increase consumption. Other things being equal, a higher level of consumption could crowd out investment in capital goods. But some of the President's tax proposals--for example, extending EGTRRA's marginal income tax rates, extending the tax rates on dividends and capital gains enacted in the Jobs and Growth Tax Relief Reconciliation Act of 2003, and expanding tax-free savings accounts--would tend to reduce consumption during the years that they were in effect. The reason is that the proposals would provide an incentive to save by lowering the effective marginal tax rates on capital income and thus increasing the after-tax rate of return on savings. (Appendix C analyzes in more detail the potential economic effects of the President's proposals for dividend and capital gains taxation, savings accounts, and the estate tax.)

Again, to summarize in one number the effects of the President's proposals on the rate of return on savings, CBO calculated the average effective marginal tax rate on capital income. That calculation was performed under two sets of policy assumptions: one comprising the current-law policies in CBO's baseline and the other, the President's policies (see Table 2-3). In both instances, the estimated effective tax rates were below all but the lowest statutory marginal rate because some capital income (for example, the interest that flows into tax-free savings accounts or pension funds) is not taxed. According to CBO's estimates, during the 2011-2015 period, the effective marginal tax rate on capital income would be about 8½ percent lower under the President's proposals than under the policies in CBO's baseline.

Table 2-3.

The reductions in taxes on capital income that the President's budget proposes would raise the rate of return on savings and affect consumption in two opposing ways (just as lowering the marginal tax rate on labor income would have opposing effects on the supply of labor). The higher return on savings that the reductions imply would tend to increase saving and reduce current consumption. But the higher return would also increase savers' wealth by boosting their after-tax income, both now and in the future--which would tend to push up their current consumption. On balance, the implications for consumption that higher returns would have could be either positive or negative. The general conclusion that researchers have drawn from empirical data is that the return on savings has a relatively small effect on how much people spend. Nevertheless, to cover other possibilities, CBO included in its analysis a range of assumptions about how the rate of return on savings might affect consumption. Some of CBO's estimates incorporate the assumption that the rate has little or no effect on how much people spend; for other estimates, CBO assumed that increasing the rate of return on savings would reduce consumption--and increase saving--significantly.

Finally, as described earlier, the larger cumulative 10-year deficit under the President's budgetary proposals (larger relative to CBO's baseline estimate) might lead some people to anticipate changes in policy in the future. If people expected higher taxes, smaller transfer payments, or fewer government services in the years to come, they might reduce their spending and build up their savings to compensate for those anticipated policies. CBO used a range of assumptions about those expectations in its analysis. Most of its estimates indicated that, on average, the domestically owned capital stock would be smaller under the policies in the President's budget than under those in CBO's baseline--but the differences were modest. Positive effects on the capital stock were projected only under the most dramatic assumption about foresight--that people care just as much for future generations as they do for themselves. In effect, if people had a sufficiently long time horizon, they might recognize and counter the deleterious effects of fiscal policy on the formation of capital and thus on future standards of living.

In addition to changes in the level of the capital stock, changes in the mix of different types of capital within that stock can affect potential output. Among the policies in the President's budget, the proposal to extend the tax rates on corporate dividends and capital gains would probably have the biggest effect on the stock's composition because it would lessen the taxation of corporate income and thus encourage a shift of capital from the noncorporate to the corporate sector. Some corporate income is taxed once at the level of the firm, through the corporate income tax, and again at the personal level, through the individual income tax on dividends and capital gains. That tax treatment distorts the way that capital is allocated in the economy because it discourages investment in the corporate sector relative to investment in the housing and noncorporate business sectors. As a result, less capital is held in the corporate sector than is optimal for the economy's efficient operation.

The taxation of dividends and capital gains also encourages firms to finance investment by borrowing rather than by issuing stock. Because firms can deduct their interest payments on debt (for example, bonds they have issued) from their taxable income, they can lessen the tax they pay. (The interest payments will then be taxed only once, at the individual level.) That tax policy may influence firms' decisions about financing and lead to economic inefficiencies. Extending the tax rates on dividends and capital gains would lessen those inefficiencies.

Technological Progress. New and improved technical processes and products are the source of most long-term growth in productivity, and the President's budgetary policies could affect the economy by influencing the rate of such progress. Researchers, however, lack a basis for estimating how fiscal policy influences technological innovation. Because so little is understood about the genesis of that innovation, CBO for the most part has not incorporated in its analysis effects on technological progress arising from the President's proposals.⁽⁴⁾

Demand-Side (Cyclical) Effects

The federal government's fiscal policies also affect the economy by adding to or subtracting from the total demand for goods and services. Increases in demand can cause firms to temporarily gear up production and hire more workers; decreases in demand can have the opposite effects. From a demand-side perspective, budgetary policies that reduce consumption (and other purchases) might slow the pace of the economy's current cyclical expansion.

Demand-side effects, however, can only temporarily raise or lower output beyond what it would otherwise have been because built-in economic forces tend to move output toward its sustainable potential level. Moreover, policies that increase demand by raising government or private consumption are likely to lower output in the long run because they tend eventually to decrease investment and the size of the capital stock.
 

A Description of CBO's Models and Their Results

In estimating the economic effects of the President's budgetary proposals relative to its baseline assumptions, CBO used several different economic models that--although similar in some respects--capture different features of the economy and reflect different ways of thinking about it. The models fall into two broad categories. Three estimate only supply-side effects. The other two are commercial macroeconometric models that emphasize the business-cycle aspects of the economy and are designed primarily to analyze demand-side effects--although they incorporate some supply-side influences as well. (Figure 2-2 presents, year by year, the impact of the President's proposals on effective tax rates on labor and capital and on the deficit--some of the key inputs for CBO's various models. Table 2-4 presents details of CBO's baseline projections of GDP and other economic variables.)

Figure 2-2.

Source: Congressional Budget Office.

a. The bars represent the effects of the President's proposals on the budget balance under CBO's baseline economic assumptions. A negative change indicates an increase in the annual deficit relative to CBO's baseline.

b. Changes in the effective federal marginal tax rate on income from capital (the share of the last dollar of such income taken by federal individual income and corporate income taxes).

c. Changes in the effective federal marginal tax rate on income from labor (the share of the last dollar of such income taken by federal individual income and payroll taxes).

Ten-Year Analysis of Supply-Side Effects

CBO used three growth models--a "textbook" growth model, a life-cycle growth model, and an infinite-horizon growth model--to analyze the supply-side effects of the President's proposals from 2006 through 2015 (the same period that CBO's baseline projections cover). The models differ in part in their assumptions about how far ahead people look in making their plans (see Appendix D). The textbook growth model is not forward-looking--it assumes that people do not explicitly take into account expected future policies in making their current plans. The life-cycle model incorporates the assumption that people make lifelong plans for working and saving but do not take into consideration events that might occur after their death. By contrast, the infinite-horizon model incorporates the assumption that people care as much about the welfare of their descendants as they do about their own welfare. That assumption means that people behave as if they will live forever.

CBO used the textbook growth model to estimate effects under two assumptions about the degree to which people will alter their hours of labor in response to changes in marginal tax rates: a "low" assumption, in which there is little response, and a "high" assumption, in which the response is at the upper end of the consensus of empirical estimates.⁽⁵⁾ Under either response assumption, the President's budgetary proposals would have little effect on GDP in the 2006-2010 period. Over the 2011-2015 period, the proposals would reduce GDP by 0.2 percent under the low assumption, in CBO's estimation, and raise GDP by 0.1 percent under the high assumption. Those estimates illustrate the budgetary proposals' opposing effects, as estimated by the textbook growth model: higher deficits crowd out investment, reducing the capital stock, but lower marginal tax rates increase the labor supplied (especially in the last five years of the projection period). Which effect dominates depends on the assumption made about how sensitive the supply of labor is to changes in marginal tax rates.

The textbook growth model's results differ from those of the other two growth models largely because the textbook model does not incorporate forward-looking behavior: people, as it represents them, do not shift hours of labor from the earlier to the later period and do not work and save more in anticipation of policy changes after 2015. In addition, unlike the life-cycle and infinite-horizon models, the textbook growth model does not incorporate any direct effects on private consumption from lower marginal (as opposed to average) tax rates and higher pretax interest rates.

The estimates that CBO produced using the life-cycle and infinite-horizon growth models are based on the assumption that people behave as if they are certain that the assumed budgetary policies--those of the President or of CBO's baseline--will be maintained over the 10-year modeling period. That assumption is an important characteristic of CBO's estimates. In reality, people would probably think it possible that the policies could change at some point during that time.

Another characteristic of the estimates produced by the life-cycle and infinite-horizon models is that they depend in part on how the President's budgetary proposals would affect people's expectations about fiscal policy beyond 2015. Assumptions about those expectations are complicated by the fact that the policies reflected in the baseline are likely to be unsustainable, owing to the extent of both current deficits and projected increases in spending for health and retirement programs.⁽⁶⁾ However, CBO assumed that people already expect that the fiscal imbalances under current-law policies will be resolved, though in some unspecified way. In projections using its forward-looking models, CBO made explicit assumptions only about the way in which increases in deficits under the President's budgetary proposals, relative to the baseline, would eventually be financed--not about the way that deficits projected in the baseline would be addressed.

The life-cycle and infinite-horizon models each generated two sets of estimates based on different assumptions about that financing (the models require such an assumption about financing because they are forward-looking). Under one assumption, people believe that the proposals will be financed by gradually reducing government spending on goods and services and on transfer payments (as shares of GDP) over the 2016-2025 period. (The reductions in the two categories would be proportional to their shares of government spending in CBO's baseline.) Under the other assumption, people believe that the proposals will be financed by gradually increasing marginal tax rates over the same period.

The life-cycle model projects that under the President's proposals, GDP will fall by between 0.2 percent and 0.4 percent over the 2006-2010 period relative to CBO's baseline and will rise by between 0.3 percent and 0.8 percent over the 2011-2015 period. Again, the difference between the projected effects over the two periods stems partly from the assumption that people will shift some hours of work from the earlier to the later period, when tax rates will be lower (relative to those assumed in CBO's baseline).

However, in the projections that incorporate the assumption of an open economy, the higher deficits under the President's proposals would attract net inflows of capital from the rest of the world. That means that people in other countries would have increasing claims on the output produced in the United States--that is, on GDP. An alternative measure of output, gross national product (GNP), measures the output produced by labor and capital supplied by U.S. residents, regardless of where those factors of production are located. That measure by definition nets out those increased foreign claims--that is, the capital income that nonresidents earn from their investments in the United States. CBO estimates that under the President's proposals, GNP will fall by about 0.4 percent over the 2006-2010 period relative to CBO's baseline and will rise by between 0.1 percent and 0.2 percent over the 2011-2015 period.

Depending on which of the assumptions is used and on whether the economy is assumed to be open or closed to flows of foreign capital, the infinite-horizon model projects that the President's proposals will subtract (relative to CBO's baseline) between zero and 0.1 percent from GDP during the first five years of the period but then add between 0.5 percent and 0.7 percent during the second five years (see Table 2-5). The difference in projected effects arises partly because people are estimated to shift some of their hours of work from the first five-year span to the second.

Table 2-5.

The effects on the economy from the President's proposed changes in fiscal policy would in turn affect the federal budget (see Tables 2-1 and 2-6). Using the economic assumptions incorporated in its baseline, CBO projects that the President's proposals will expand the cumulative deficit over the 2006-2010 period by $104 billion. Under the various assumptions used in the growth models that CBO employed in its analysis, the economic effects of the President's proposals over that period could further increase the deficit by $6 billion to $61 billion.

Table 2-6.

For the 2011-2015 period, the President's budgetary policies are projected to boost the cumulative deficit by $1,497 billion under the baseline's economic assumptions. The economic effects of the President's proposals over that period, according to the models, could add as much as $26 billion to that increase or subtract as much as $131 billion from it.

Five-Year Analysis Including Demand-Side Effects

CBO turned to macroeconometric forecasting models created by two private forecasting firms--Macroeconomic Advisers and Global Insight--to analyze both the demand-side and supply-side effects of the President's budgetary proposals on the economy over the next five years. Although the models include embedded growth models, their design concentrates on demand-side (cyclical) economic effects. Such effects are increasingly harder to estimate as the projections extend further. Therefore, CBO used those models to produce estimates only for the first five years of the 2006-2015 projection period.

Like the textbook growth model, Macroeconomic Advisers' and Global Insight's models are not forward-looking--people, as the models represent them, do not behave as though they have specific expectations about future policies or economic developments. Instead, people respond to economic changes in the same way as they have in the past, regardless of the source of those changes. For example, they react to the tax proposals in the President's budget (which would boost disposable income) by increasing their purchases to about the same degree as they have, on average, in the past when disposable income rose. However, in reality, people might increase their spending by a smaller amount in response to a change in taxes than they would in response to some other change that expanded their income (such as an increase in productivity), because they might feel that tax legislation was more likely to be reversed in the future.

The lack of forward-looking behavior in the macroeconometric models implies that specific policy changes that are scheduled to occur in the future do not affect the current behavior of people represented in the models unless special adjustments are made to mimic such behavior. For example, the President's proposal to extend many of the provisions in EGTRRA would reduce taxes from 2011 to 2015 compared with the levels in CBO's baseline. Those lower taxes would increase the amount of after-tax income that people expected in the future, which might cause them to boost their spending today (as the forward-looking models imply). In the macroeconometric models, however, those changes in taxes affect consumption only when they occur.

The projections by Macroeconomic Advisers' and Global Insight's models also required an adjustment to the supply of labor. Like the textbook growth model, those models do not incorporate the effects of taxes on the number of hours worked. Therefore, CBO adjusted the models' equations to incorporate its own estimates of those effects.⁽⁷⁾

Two sets of estimates were produced to illustrate the magnitude of demand-side effects in the models. For one set, CBO ran the models using the standard assumption that monetary policy would allow both demand- and supply-side effects. For the second set, CBO attempted to isolate the supply-side effects by altering interest rates in such a way as to hold the unemployment rate at the level projected in CBO's baseline. That approach produced an estimate of the implications of the proposals for potential (noncyclical) GDP.⁽⁸⁾

Macroeconomic Advisers' model and Global Insight's model differed in their estimates of the effects of the policies in the President's budget relative to the policies in CBO's baseline (see Table 2-5). However, both models predicted that together, the demand- and supply-side effects of those changes would have little effect on GDP, on average, between 2006 and 2010. Also similar were the models' estimates of the supply-side impacts of the President's proposals. Both macroeconometric models projected that the supply-side effects of the President's budgetary policies would decrease output by between 0.1 percent and 0.2 percent.

The projected economic impacts of the proposals would in turn affect the budget. According to the projections from Macroeconomic Advisers' model, the proposals' impact on the economy could add $5 billion to the $104 billion increase in the deficit projected for the 2006-2010 period under the baseline's economic assumptions (see Table 2-6). By the estimates of Global Insight's model, the supply-side and cyclical effects of the President's proposals could add $1 billion to the deficit over the same period.