This description of crowding-out and crowding-in, and why crowding-in is likely to dominate in recessions, is from Baumol and Blinder's principles text, Macroeconomics: Principles and Policy. The idea is that investment is a negative function of the interest rate and a positive function of income, i.e. I = I(r,y), where Ir<0 and Iy>0. Since an increase in government spending increases both r and y, and since each moves investment in the opposite direction, the net effect on investment (and hence future growth) depends upon which variable, r or y, has the largest impact on I. In a recession, a change in government spending does not have much impact on r, but it does have an effect on y so that the increase in government spending is likely to bring about an increase, not a decrease, in investment. Thus, unlike deficit spending near full employment, government spending in recessions can lead to higher growth rates in the future thereby alleviating concerns about the costs to future generations:
Debt, Interest Rates, and Crowding Out
So far, we have looked for possible problems that the national debt might cause on the demand side of the economy. But the real case for cutting the deficit or raising the surplus in the United States comes on the supply side. In brief, large budget deficits discourage investment and therefore retard the growth of the nation's capital stock. Conversely, budget surpluses speed up capital formation and growth.
The mechanism is easy to understand by presuming (as is generally the case) that the Fed does not engage in any substantial monetization. We have just seen that budget deficits tend to raise interest rates. By the same logic, budget surpluses tend to reduce interest rates. But we know from earlier chapters that the rate of interest (r) is a major determinant of investment spending (I). In particular, higher r leads to less I, and lower r leads to more I. The volume of investment made today will, in turn, determine how much capital we have tomorrow-and thus influence the size of our potential GDP. This, according to most economists, is the true sense in which a larger national debt may burden future generations and a smaller national debt may help them:
A larger national debt may lead a nation to bequeath less physical capital to future generations. If they inherit less plant and equipment, these generations will be burdened by a smaller productive capacity-a lower potential GOP. In other words, large deficits may retard economic growth. By the same logic, budget surpluses can stimulate capital formation and economic growth.
Phrasing this point another way explains why this result is often called the crowding-out effect. Consider what happens in financial markets when the government engages in deficit spending. When it spends more than it takes in, the government must borrow the rest. It does so by selling bonds, which compete with corporate bonds and other financial instruments for the available supply of funds. As some savers decide to buy government bonds, the funds remaining to invest in private bonds must shrink. Thus, some private borrowers get "crowded out" of the financial markets as the government claims an increasing share of the economy's total saving.
Some critics of deficit spending have taken this lesson to its illogical extreme by arguing that each $1 of government spending crowds out exactly $1 of private spending, leaving "expansionary" fiscal policy with no net effect on total demand. In their view, when G rises, I falls by an equal amount, leaving the total of C + I + G + (X - IM) unchanged.
Under normal circumstances, we would not expect this to occur. Why? First, moderate budget deficits push up interest rates only slightly. Second, private spending appears only moderately sensitive to interest rates. Even at the higher interest rates that government deficits cause, most corporations will continue to borrow to finance their capital investments.
Furthermore, in times of economic slack, a counterforce arises that we might call the crowding-in effect. Deficit spending presumably quickens the pace of economic activity. That, at least, is its purpose. As the economy expands, businesses find it profitable to add to their capacity so as to meet the greater consumer demands. Because of this induced investment, as we called it in earlier chapters, any increase in G tends to increase investment, rather than decrease it as the crowding-out hypothesis predicts.
The strength of the crowding-in effect depends on how much additional real GDP is stimulated by government spending (that is, on the size of the multiplier) and on how sensitive investment spending is to the improved profit opportunities that accompany rapid growth. It is even conceivable that the crowding-in effect can dominate the crowding-out effect in the short run, so that I rises, on balance, when G rises.
But how can this be true in view of the crowding-out argument? Certainly, if the government borrows more and the total volume of private saving is fixed, then private industry must borrow less. That's just arithmetic. The fallacy in the strict crowding-out argument lies in supposing that the economy's flow of saving is really fixed. If government deficits succeed in raising output, we will have more income and therefore more saving. In that way, both government and industry can borrow more.
Which effect dominates-crowding out or crowding in? Crowding out stems from the increases in interest rates caused by deficits, whereas crowding in derives from the faster real economic growth that deficits sometimes produce. In the short run, the crowding-in effect-which results from the outward shift of the aggregate demand curve-is often the more powerful, especially when the economy is at less than full employment.
In the long run, however, the supply side dominates because, as we have learned, the economy's self-correcting mechanism pushes the growth rate of actual GDP toward the growth rate of potential GDP. With the economy approximately at full employment, the crowding-out effect takes over: Higher interest rates lead to less investment, so the capital stock and potential GDP grow more slowly. Turned on its head, this is the basic long-run argument for running budget surpluses: They speed up investment and growth.
The Bottom Line
Let us summarize what we have learned so far about the crowding-out controversy.
• The basic argument of the crowding-out hypothesis is sound: Unless the economy produces enough additional saving, more government borrowing will force out some private borrowers, who are discouraged by the higher interest rates. This process will reduce investment spending and cancel out some of the expansionary effects of higher government spending.
• But crowding out is rarely strong enough to cancel out the entire expansionary thrust of government spending. Some net stimulus to the economy remains.
• If deficit spending induces substantial GOP growth, then the crowding-in effect will lead to more saving-perhaps so much more that private industry can borrow more than it did previously, despite the increase in government borrowing.
• The crowding-out effect is likely to dominate in the long run or when the economy is operating near full employment. The crowding-in effect is likely to dominate in the short run, especially when the economy has a great deal of slack.
• Surpluses have just the opposite effects. When slack exists, they are likely to slow growth by reducing aggregate demand. But in the long run, budget surpluses are likely to foster capital formation and speed up growth.
Update: For more on crowding-in, see ZIRP Deficits cause Crowding In of Investment, by reducing Deflation.