As Congress presses forward with an economic stimulus package, questions are still being asked about whether it would raise interest rates, thereby negating any stimulus. There are legitimate concerns that can be raised about whether the bill recently reported by the House Ways and Means Committee will do much to increase growth in the short run, but its impact on interest rates is not one of them.
The idea is that lower revenues resulting from tax cuts will lower the budget surplus or push it into deficit. Whether or not an actual deficit emerges is irrelevant. Since surpluses are, theoretically, additions to national saving, any decline in the surplus should have exactly the same effect on interest rates as a rise in the deficit. In other words, if the government went from a surplus of $100 billion to a surplus of $50 billion, the impact should be the same as if it went from a deficit of $50 billion to one of $100 billion. Either way, there has been the same negative $50 billion change in government's net fiscal impact on financial markets.
All other things being equal, one would expect that increasing federal borrowing, or reducing the surplus by an equal amount, will raise interest rates above what otherwise would be the case. There are important theoretical questions about why this might not be the case. But leaving those aside, the real question is the order of magnitude. How much will interest rates rise if, because of the stimulus package, the government takes in $75 billion less revenue next year?
Let's first look at the record. Over the last 20 years, we have had experience with large deficits, large surpluses and major changes in both over short periods of time. The evidence, to say the least, offers little support for a direct link between deficits and interest rates.
-- In 1982, the federal government ran a budget deficit of 4.1 percent of the gross domestic product. That year, the Treasury's 30-year bond rate averaged 12.8 percent. In 1993, the government ran a deficit of exactly the same size, as a share of GDP, but the interest rate was just 6.6 percent.
-- Between 1986 and 1987, the deficit fell from 4.3 percent of GDP to 3.1 percent, yet the 30-year bond interest rate rose significantly from 7.8 percent to 8.6 percent. Conversely, between 1991 and 1992, the deficit increased from 3.6 percent of GDP to 4.7 percent, but the interest rate fell from 8.1 percent to 7.7 percent.
-- Between 1995 and 2000, the budget went from a deficit of 2.6 percent of GDP to a surplus of 2.2 percent -- a net change of 4.8 percent of GDP. This is a huge amount, equal to over $500 billion. Yet the 30-year bond only fell from 6.9 percent to 5.9 percent over this period. If the same order of magnitude holds, then a $75 billion loss in revenue -- equal to about 0.7 percent of GDP -- would only raise interest rates by at most 15 basis points (0.15 percent).
Of course, it is very difficult to isolate the effect of a change in the federal deficit on interest rates. There are always other things going on. The Federal Reserve, in particular, can overwhelm fiscal policy with changes in monetary policy. Moreover, the impact of Fed policy is not easily predictable. Sometimes, Fed-easing lowers long-term interest rates by reducing the cost of money to banks, but other times, it can raise rates by creating inflationary expectations.
The overall economic situation is also critical to setting interest rates. If the economy is slowing, as now, business investment falls and personal saving rises as individuals cut back on spending and put away extra funds for a rainy day. Moreover, the possible $75 billion reduction in the federal government's fiscal position has to be seen together with the fact that state and local governments run surpluses of about $60 billion, foreign investment adds about $100 billion to saving, and total domestic saving is about $1.8 trillion annually.
For these and other reasons, serious academic research has failed to find more than a trivial effect on interest rates from federal deficits. The latest, by economists Douglas Elmendorf of the Federal Reserve and Gregory Mankiw of Harvard, suggests that the stimulus package being considered by Congress would raise long-term interest rates by at most five basis points. This is less than many day-to-day interest rate movements simply due to changing market conditions.
Those who oppose tax cuts for ideological reasons will still oppose the stimulus package. But if they say it is because it will raise interest rates significantly, they are not telling the truth.