By John Ricco, Efraim Berkovich and Richard Prisinzano
The New York Times reports that the Trump administration is exploring indexing capital gains to inflation. On March 23, 2018, PWBM scored a similar proposal on a purely static basis at $102 billion, which does not include the potential impact on economic growth. In this blog entry, we use PWBM’s OLG model to explore the dynamic effects of capital gains indexation, which includes the impact of the proposed policy change on economic growth. We project that this policy change will produce no meaningful economic feedback effect over the next decade.
We model two different versions of capital gains indexation policy. The first is a “retroactive” version where all capital gains, including gains from past investments, are eligible for inflation indexation. The second is a “prospective” version where asset basis is only adjusted for inflation which occurs after policy implementation, which we assume starts on January 1st, 2019. Of course, both policies increase the after-tax return to additional savings decisions by the same amount. However, the retroactive policy also gives an effective lump-sum tax rebate to previous investment, which produces more debt relative to the prospective version.
As Table 1 shows, on a static basis, we estimate that the retroactive policy costs $95 billion over 2019-2028, whereas the prospective policy costs $49 billion over that same timeframe. The $95 billion static estimate is slightly smaller than our previous estimate of $102 billion for a few reasons: we have updated our methodology for imputing real asset basis; we now include a behavioral response in capital gain realizations1; and we are now examining a budget window starting in 2019 as opposed to 2018.
| Cumulative revenue change | Change in debt | |||||||
|---|---|---|---|---|---|---|---|---|
| Retroactive | Prospective | Retroactive | Prospective | |||||
| Year | Static | Dynamic | Static | Dynamic | Static | Dynamic | Static | Dynamic |
| 2019-2028 | -$95 | -$96 | -$49 | -$49 | $108 | $110 | $54 | $55 |
| 2019-2038 | -$227 | -$232 | -$158 | -$158 | $299 | $306 | $197 | $199 |
Consistent with the empirical evidence, the dynamic projections above assume that the U.S. economy is 40 percent open and 60 percent closed. Specifically, 40 percent of new government debt is purchased by foreigners. The dynamic projections above assume a high rate of return to private capital.
Table 1 also presents revenue estimates on a dynamic basis, which includes economic feedback effects. Notice the revenue loss does not differ meaningfully between the static and dynamic measures over the next 10 years. Table 2 gives the reason: we estimate that indexation spurs roughly zero net additional economic growth during the 10-year budget window.
| Change in GDP | Change in labor income | Change in capital services | ||||
|---|---|---|---|---|---|---|
| Year | Retroactive | Prospective | Retroactive | Prospective | Retroactive | Prospective |
| 2028 | -0.01% | 0.00% | -0.01% | 0.00% | -0.02% | 0.01% |
| 2038 | -0.04% | -0.01% | -0.04% | -0.01% | -0.10% | -0.01% |
Consistent with the empirical evidence, the dynamic projections above assume that the U.S. economy is 40 percent open and 60 percent closed. Specifically, 40 percent of new government debt is purchased by foreigners. The dynamic projections above assume a high rate of return to private capital.
Why are the estimated effects on economic activity so small? As Gravelle (2018) notes, indexing capital gains to inflation is more limited in its potential for growth effects compared to other types of tax cuts on capital:
...unlike some other tax cuts (such as expensing or corporate rate cuts) that occur at the firm level and have the potential to draw capital from abroad as well as potentially increase saving, capital gains are on the savers side, which means their effects operate solely through saving with some of that saving leaking into investments in other countries.
In fact, looking beyond the usual 10-year window, Table 1 shows that the retroactive policy loses an additional $5 billion on a dynamic basis over the next 20 years. The reason is that additional debt accumulation reduces private capital formation at a greater rate than the positive impact of a slightly higher after-tax returns to savings coming from indexation.
-
We use a realization elasticity with respect to a rate cut of -0.66 per Bakija and Gentry (2014). Gravelle (2018) argues that indexation would have a smaller impact than a pure rate cut. To the extent this assertion is correct, our estimate is an upper bound on the realization response and therefore, a lower bound on the revenue loss. ↩