Greg Mankiw must know better than he indicates in his analysis of the debt in today's NYT. He complains that efforts to use large-scale stimulus to boost the economy may put excessive burdens on our children. Let's work with this one for a moment. Suppose the stimulus option is spending an additional 4 percent of GDP (@$600 billion a year) for the next two years. Some of this money would go to consumption uses (e.g. health care, food stamps etc.) and some would go for investment purposes (e.g. infrastructure and energy conserving building retrofits). As a result of going the stimulus option, the economy will grow faster (or contract less) and return to high levels of employment more quickly than if we follow Mankiw's advice and decide not to spend the money in order to avoid passing on debt to our children. Okay, where are we after two years in the stimulus versus no stimulus case? Using some very crude numbers, let's say that our spending had an average multiplier effect of 1.25. In other words, the secondary impact on the economy that results from people spending the money that they get paid to build roads or retrofit buildings adds an addition 25 percent to economic output beyond the original effect (this is a very modest assumption). This means that our economic output will be 5 percent larger over the next two years than would otherwise be the case. Assuming that employment is roughly proportional to output, as a result of the stimulus as additional 7.5 million people will be employed over the next two years as compared to the non-stimulus scenario. Okay, under the absurd assumption that the economy has completely recovered in 2011 in both cases (it will recover much more quickly in the stimulus case than in the no stimulus case), let's see what we have done to our kids. In the stimulus case, we spent an amount equal to 8 percent of GDP (we'll ignore any growth in the GDP that affects the denominator in this story). However, the additional growth to the economy led to additional tax receipts. Since government revenues are approximately equal to 20 percent of output, if output expanded by 5 percent, then revenues grew by an amount equal to 1 percent of GDP. This means that if we spent an amount equal to 4 percent of GDP boosting the economy, then the debt grew by an amount equal to 3 percent of GDP each year for a total increase in the debt burden over the two years of 6 percent of GDP (an amount that is in the ballpark of the budgetary cost of the Iraq War). While those of us alive today will all benefit from the increased output and employment due to the stimulus, what have we done to our kids? First, we should be clear that we gave them a more productive economy. Suppose that just one-quarter of the stimulus (1 percent of GDP) goes to useful investment projects such as roads that are actually needed or retrofits that reduce energy consumption in future years. If we assume that the payback on this investment is 10 percent a year (a standard assumption), then this public investment will have permanently increased output by an amount equal to 0.2 percent of GDP (10 percent of 2 percent) or approximately $30 billion a year. This isn't the only effect. The additional output is likely to increase private investment -- companies are far more likely to expand and get new equipment when they see demand than when the economy is in a severe slump. If we assume that the additional private investment is equal to 10 percent of the GDP growth created by the stimulus, then investment will have increased by an amount equal to 1 percent of GDP (10 percent of 10 percent) over the two year period. Again, assuming the return on this investment averages 10 percent, this will have permanently boosted our annual output by an additional 0.1 percent, or another $15 billion a year. This brings the total growth dividend from the stimulus to $45 billion or 0.3 percent of GDP. A full account of the growth benefits would also add in the gains from increased education. It is likely that more people will be able to attend college or receive other post-secondary education either as a direct result of education support in a stimulus package or as indirect result of families with jobs being better able to support the education of their children. Additional education would also be expected to increase economic output, but we'll ignore any growth dividend for the moment and see what we've done to out kids by spending so much money on this stimulus package. Okay, we have increased our debt by an amount equal to 6 percent of GDP. Let's assume that the real interest rate on government debt is equal to 3 percent, an interest rate far higher than we have seen in recent years. This means that the additional debt will have increased the tax burden on our kids by an amount equal to 0.18 percent of GDP (3 percent of 6 percent), or approximately $27 billion a year at current output levels. However, the growth dividend from the stimulus will cut this burden in half. Since GDP will be 0.3 percent higher as a result of the stimulus, the government will take in an additional amount of tax revenue equal to 0.06 percent of GDP each year (20 percent of 0.3 percent), which leaves a net interest burden from the stimulus of 0.12 percent of GDP or approximately $18 billion a year. So, if we are evil to our kids and use deficit spending to boost the economy, they will be forced to pay an additional tax burden in perpetuity equal to 0.12 percent of GDP. But wait, when we are all dead (Mankiw's scenario) who will be collecting this interest? That's right, our kids will be collecting the interest. (We're ignoring foreigners, since foreign ownership of government debt and other U.S. assets depends on the trade deficit, which in turn is primarily the result of the over-valued dollar, which has little direct relationship to the budget deficit.) So, we will be taxing our children, some of whom own the debt and some of whom don't, to pay interest to those of our children who do own the debt. This can make our kids worse off insofar as taxes lead to economic distortions, but the transfer itself is entirely among our kids -- it doesn't go to us, we're all dead. So, in the stimulus story, we have handed our kids a more productive economy than in the no stimulus story. We also may have created an economy in which tax distortions are somewhat larger (although the plunge in stock values means that after-tax returns are likely to be far higher in the future even with higher tax rates, than they were in the pre-crash years), but any plausible measure of the distortions resulting from the additional tax burden will be dwarfed by the addition to GDP resulting from the stimulus. (If the tax distortions are equal to 25 percent of the tax collections, then they will be equal 0.03 percent of GDP [25 percent of 0.12 percent]. The stimulus permanently raised output by 0.3 percent.) In short, if we think about our kids, we should do the opposite of what Mankiw argued. We should support a big stimulus package that is focused on investments for the future. This is essential for sustaining the economy now and it will help our kids for decades to come.
--Dean Baker
