Eakinomics: “Stimulus” and the TCJA

I’m not a big fan of the “stimulus” line of reasoning — evaluating government policy strictly on the basis of changes in dollars of taxes collected and spent. In thinking about the Tax Cuts and Jobs Act (TCJA), for example, I think the most important elements are the changes in incentives to innovate and invest in the United States. These changes to tax rates, capital cost recovery, and the tax base have a bigger impact on productivity and growth than simply cutting the dollars of taxes collected. Nevertheless, the discussion around the TCJA, amplified by the Bipartisan Budget Act of 2018 (BBA18), has focused on the stimulus it will deliver to the economy. Indeed, there have been assertions that it would exceed that delivered by the Obama-era stimulus (the American Recovery and Reinvestment Act or ARRA).

The gold standard for measuring Keynesian stimulus is the change in the “full-employment” surplus (or deficit) due to the policy. Using this measure ensures that one captures only the additional deficit produced by the policy and not by suppressed economic activity. This is roughly what is measured by a Congressional Budget Office (CBO) cost estimate. Let’s take a look at the numbers.

Converted to a calendar year basis, ARRA increased the deficit by $239 billion in 2009 and an additional $95 billion in 2010. These represented 1.7 and 0.6 percent of Gross Domestic Product (GDP), respectively. Combining the tax and outlay effects of the TCJA and BBA18 yields $257 billion in 2018 and an additional $162 billion in 2019. These are 1.3 percent and 0.8 percent of GDP (assuming baseline growth of 2.2 percent in real GDP and 2 percent inflation). So, not quite “bigger than Obama” but still quite substantial. (One caveat is that the Obama stimulus was incredibly poorly designed, so its impact was diminished.)

A second debate has arisen as to whether the stimulus will be offset by other factors, yielding low growth impacts for TCJA. As mentioned at the outset, this assumes that stimulus is all the bill is about. That’s off the mark, but let’s evaluate it at face value. The economy is at, or at least very near, full employment so there is little opportunity for rapid growth simply from putting the unemployed back to work. That means one should discard any notion of “multipliers” in which the 1.3 percent stimulus in 2018 creates even more economic growth. Instead, the net effect will likely be somewhat smaller.

Some argue that it might even be zero as the stimulus “crowds out” the private sector. As The New York Times put it, “In mainstream models of how the economy works, it’s the idea that if the government runs budget deficits when the economy is at full employment, its borrowing won’t spur new economic activity as desired. Instead, the borrowing will simply raise interest rates and squeeze out private-sector investment, resulting in no net improvement in the economy.” That’s also likely off the mark. A more plausible scenario is that in response to the stimulus, interest rates rise and capital flows into the United States seeking the higher returns. This serves to finance investment, but in the process the dollar strengthens, imports rise, and exports diminish. Put differently, any crowding out will likely be focused on net exports. This is important. To the extent that stimulus gets offset, it preserves the investment generated by better incentives, preserves the improved productivity it generates, and thus preserves the improved long-run growth.

The bottom line? The impact of the TCJA and BBA18 is strong insurance against a recession developing, may contribute to hitting (finally) the Fed’s inflation targets, and improves the innovation and investment incentives over the long term.