State implementation of estate tax is empirically effective—the benefits outweigh the cost.

**Read this on state CP solvency against aff, suggests that the CP solves a sufficient portion of the aff

Based on these assumptions, we estimate that the cost-benefit ratio of having a billionaire estate tax in 2017 is 0.47 for the average state (with at least one billionaire), indicating that the the additional revenues from the estate tax exceeds the cost of foregone income tax revenues by a significant margin. Remarkably, despite the high elasticity of billionaires to state estate taxes, the gain for the average state from taxing the estates of the remaining billionaire population exceeds the loss of income tax flows from those billionaires that flee.

The cost-benefit ratio varies greatly across states due to the variation in state income tax rates. Table 7 shows estimates of costs and benefits for each state. Panel A includes states that had an estate tax as of 2017. The first row indicates that Connecticut had 7 Forbes billionaires in 2017, with an aggregate wealth of $47.2 billion. Based on the state personal income tax rate, we estimate the revenue benefit of having an estate tax to be 26% larger than the revenue cost, with a cost-benefit ratio equal to 0.74. In the last column we show that net expected present value (EPV) of having an estate tax is $376 million. We stress that entries in column 5 are not annual figures, but the expected present value computed over the expected remaining lifetimes of the 2017 Forbes 400.

By contrast, for Hawaii the benefit of having an estate tax is 43% larger than the cost, with a cost-benefit ratio equal to 1.43. The expected present value of having an estate tax is -$73 million. The difference between Connecticut and Hawaii is largely due to the difference in their personal income tax (PIT) rate. Hawaii’s PIT is higher than Connecticut’s. The higher PIT rate in Hawaii means a higher opportunity cost of foregoing billionaires’ income tax streams.

In general, we find that states with high personal income tax rates tend to have higher cost-benefit ratios, while states with low personal income tax rates tend to have low cost-benefit ratios. Indeed, for the state of Washington, which has no PIT, the cost-benefit ratio is 0. We estimate that for this state expected present value of having an estate tax is $2.005 billion, the largest in this group. Maryland, Massachusetts and Pennsylvania are other examples of states with benefits much larger than costs, with cost-benefit ratios equal to 0.56, 0.50 and 0.35, respectively. New York has a cost-benefit ratio of 0.66, and an EPV of having an estate tax equal to $5.18 billion.

Overall, of the 10 states that currently have an estate tax and have at least one billionaire, 9 states have a cost-benefit ratio below one and only Hawaii has a ratio above one. The last row in panel A indicates that for the average state in this group, the benefit of having an estate tax is 39% larger than the cost.

Panel B shows results for states that did not have an estate tax in 2017. In this group, California has by far the largest cost-benefit ratio—1.45—because it has the highest PIT rate. We estimate that if California adopted the estate tax, it would lose $8.26 billion in EPV of revenue. However, California is more the exception than the rule. For most other states in this panel, we estimate that the cost-benefit ratio is well below 1, indicating that estate tax adoption would increase their tax revenues. For example, Florida and Texas have no personal income tax and therefore the cost-benefit ratio for them is 0. We estimate that they would gain $7.68 and $7.06 billion, respectively, in EPV of revenues if they adopted estate tax. The two other states with the largest absolute estimated gains from adoption in this group are Nevada and Tennessee.

The last row indicates that on average, the cost-benefit ratio for this group is 0.42, indicating that for the average non-adopting state the tax revenue benefit of having an estate tax on billionaires would significantly exceed the cost from foregone income tax revenues. Overall, of the 28 states that currently do not have estate tax and have at least one billionaire, 27 states would increase their revenues if they adopted estate taxes. However, the relative magnitude of costs and benefits varies widely, and depends strongly on the personal income tax rate, as the cost of adoption tends to be much higher in states with high income tax rates.

In Panel A of Appendix Table A6 we probe the sensitivity of our results to alternative parameterizations. First, we allow for wealth and income to grow over time. Recall that the baseline estimates are based on wealth and income set at their 2017 levels. But historically, wealth and income of Forbes billionaires have grown significantly, as we saw in Figure 2. Alternative 1 in the table assumes that real wealth and income grow beyond 2017 at 7.0% per year, which is the average annual growth rate of Forbes 400 real wealth from 1982–2017. Next, we increase the assumed longevity of the surviving spouse. In the baseline we used IRS life expectancy tables that implicitly assume that one’s spouse is within 10 years of age. Alternative 2 uses IRS surviving spouse life expectancies, by age of older spouse, for a couple 20 years apart in age. Alternative 3 assumes states discount future revenues using a real interest rate of 1% (vs. 2% in the baseline). Alternative 4 assumes states discount using a real interest rate of 3%. Alternative 5, following Saez and Zucman (2019), assumes that the Forbes 400 have income equal to half of the gross adjusted income reported by the top 400 income taxpayers according to IRS SOI data (vs. the baseline where we assume it is 10.3% of Forbes 400 taxable wealth). All other parameter assumptions are the same as in the baseline scenario. The finding that benefits exceed costs for most states is found to be robust to these alternative assumptions.