Thread by @fatihguvenen

1. (Thread): Here is a tweet summary of our paper on wealth taxation: “Use It or Lose it: Efficiency Gains from Wealth Taxation”: bit.ly/2k46stT
Along the way, I discuss some qualifications, and clarify a few things.

2. Our bottom line result is that wealth taxation is preferable to capital income taxation for several reasons. It improves aggregate productivity, reduces some key types of inequality, and raises welfare for a large majority of households.

3. Key feature of wealth taxes that delivers these results is “use-it-or-lose-it” effect (hence the title), which is a powerful mechanism that reallocates wealth to improve aggregate efficiency/productivity. It emerges from a combination of two features/assumptions:

4.First feature: some individuals are better entrepreneurs than others. So, they earn a higher profit/return per dollar of wealth they invest. Recent empirical studies document evidence of such differences that are persistent.

5.Second feature: entrepreneurs face borrowing constraints, which hits the most productive entrepreneurs the hardest (since they want to borrow to invest). As a result, they produce well below their optimal level.

6.With these two features, wealth taxation has a “use-it-or-lose-it” effect that rewards the productive use of wealth and discourages idleness and low productivity.

7.Details: There hasn’t been much academic work on wealth taxation because without the first feature—persistent heterogeneity across entrepreneurs—wealth taxes work very similarly to capital income taxes (no use-or-lose effect).

8.Note: we don’t study the effects of wealth taxes *in addition* to existing capital income taxes in the US (as many current policy proposals argue for). Instead, we compare having either one or the other. This is a key distinction that should be kept in mind.

Tax Reform: *Replacing* capital income taxes (as in the US) with a wealth tax that keeps labor tax rate and government revenues unchanged raises output and average welfare substantially. This is one of the simplest policy reforms, as it does not touch any other taxes.

10.Gains can be quite large—about 7% of consumption for every year during the average individual’s lifetime. The gains are broad based across income and wealth groups.

11.Gains don’t come from shifting tax burden from workers to the wealthy but rather from a redistribution *among* the wealthy: from high-productivity toward low-productivity ones. Reallocation improves efficiency and raises output, wages, and consumption.

12.Who are the low-productivity wealthy? These can include wealthy entrepreneurs whose best years are behind them, some children who inherit large fortunes but have no interest or ability to grow the wealth, one-hit wonders, etc.

13.Next we ask: what is the optimal wealth tax rate if the government can choose tax rates on wealth and labor income simultaneously? The optimal wealth tax ranges from 2% to 3% per year. This high rate allows the government to reduce labor taxes substantially.

14.Optimal Wealth Tax Rate: What if the government can also choose an exemption level and tax only wealth above that (so far all wealth got taxed)? With threshold, around 60% get taxed, average welfare gain is modest, but % who gain goes up quite a bit.

15.About 75% of welfare gains from optimal wealth taxation are due to efficiency gains and 25% due to a more equitable distribution of resources across households. Tax policies often have an equity-efficiency trade-off but surprisingly not here (comp. w/ capital income tax).

16.So far, these results were about a comparison across economies with different tax systems. In reality, if the new tax system is implemented, the economy will go through an adjustment period (transition), so it may take a while for gains to be realized.

17.So, the next question is: how do the individuals who are alive at the time of the reform fare relative to their no-reform future? We find that they still gain substantially, ranging from 80% to 100% of the gains found above.

18.This result was surprising to us because transitions often imply much lower welfare gains than comparisons across economies. But, often, the reason is they involve large sacrifices, like a higher aggregate savings rate to increase the total capital stock in the economy.

20. Now for qualifications. First, this is quantitative work, so the final numbers we report here depend on many choices we make. We conduct quite a bit of robustness analysis, but as always there is more to explore going forward.

In the paper, we don’t get into implementation questions. But one issue I will mention: the wealth tax should be levied on the “book value” of firms, not “market value”. The latter accounts for entrepreneur’s productivity, so taxing it will undo benefits discussed here.

22.The End. Thanks for reading. More details in the paper:

bit.ly/2k46stT

#wealthtax #taxingwealth #taxingcapital