Sweden Does Soak Its Rich

Cathie Jo Martin and Alexander Hertel-Fernandez wrote a piece at Vox titled "How Sweden fights inequality -- without soaking the rich." The piece argues that the Nordics countries show that "redistribution" is not the solution to low-end inequality. This isn't true.

1. Sweden Soaks the Rich

The authors claim that lower tax progressivity in Nordic countries show that they do not soak the rich. 

Comparing the "progressivity" of tax systems across countries can lead to serious confusion of the sort that I suspect to be at play here. To see why, consider two hypothetical countries, A and B. Suppose that in both countries, there are just two people, a person who has a market income of $10,000 and another person who has a market income of $100,000.

In country A, the effective tax rate of the poorer person is 1% and the effective tax rate of the richer person is 10%. This means that the poorer person gives up $100 in taxes while the richer person gives up $10,000 in taxes. Because the effective tax rate of the rich person is 10x the effective tax rate of the poor person, this would score as a very progressive tax system.

In country B, however, the effective tax rate of the poorer person is 15% while the effective tax rate of the richer person is 30%. At these rates, the poor person would give up $1,500 in taxes while the richer person would give up $30,000 in taxes. The rich person's tax rate is only 2x that of the poor person's and so this is seen as much less progressive than country A.

Would it be correct to conclude from this that country A soaks the rich and country B doesn't? No that's manifestly absurd. Country B extracts triple the amount of taxes from the rich person as country A does. Imagine that all of the tax revenue from the two countries was spent on universal benefits. In country A, that would mean each person gets $5,050 in benefits, giving the poor person a disposable income of $14,950 and the rich person a disposable income of $95,050. In country B, that would mean each person gets $15,750 in benefits, resulting in the poor person having a disposable income of $24,250 and the rich person having a disposable income of $85,750.

Clearly, country B, with the more "regressive" system is soaking the rich by a far greater amount. It, on net, shifted $14,250 from the rich person to the poor person, while country A only, on net, shifted $4,950 from the rich person to poor person. The tax system that is massively more "progressive" and "rich-soaking" (country A) transferred just one-third the amout of money from rich to poor!

Clever people will spot the problem here. At lower tax levels, tiny tax differences (in dollar terms) yield vastly greater progressivity (in percentage of income terms). The higher the tax levels go, the greater the tax difference (in dollar terms) has to be be in order to maintain the same level of progressivity. It would be nearly impossible for countries with the highest tax levels to match the tax progressivity (in percentage of income terms) of those with the lowest tax levels, but that does not mean that they don't soak the rich.

For less hypothetical evidence, consider Sweden's marginal income tax rates, which top out around 70%:

Countries like Sweden have less progressive tax systems, not because they tax the rich less than us (they in fact tax them more), but because they tax the non-rich more than us. Looking only at progessivity regressions (without any regard for tax levels) hides this fact, which is presumably why the authors of the Vox piece totally bungled it.

2. Nordics Have Low Inequality Because of Transfers

After the above mistake, the piece goes on to claim "[t]he reason Northern European countries with more regressive taxes achieve such high levels of labor market equality, despite less progressive tax systems, is that they spend money on increasing the skills and earning power of low-end wage earners." Regular readers of mine know that reading this sentence sent me into a fit of rage. This is because it is demonstrably false.

Here are the market poverty rates (defined as those with incomes below 50% of the median income) of the four Nordic countries and the US:

The Nordics as a whole have somewhat lower market poverty rates than the US, but it's hardly something to write home about. The average market poverty rate for the Nordic countries is 27.2%, which is only a touch lower than the US market poverty rate of 28.4%. Finland actually has a higher market poverty rate than the US does.

The success of the Nordics at the low-end of the income distribution is not the result of building "human capital" in some way that causes the bottom of the income distribution to pull down all of that sweet market income. It's transfer incomes!

Here is the same graph above, but I've introduced a second bar that shows you the disposable income (i.e. after taxes and transfers) poverty rates of these countries:

It's not hard to figure out what's going on here and it's not market income equality driven by human capital policies.

Conclusion

One of the few good things about this piece is that the authors implicitly accept the idea that the way to fight low-end inequality is to do what the Nordics do. They then say the Nordics don't soak the rich and that they keep inequality at the bottom low through human-capital-building that results in higher market incomes. But this just isn't true. The Nordics soak the rich and achieve low inequality at the bottom through transfers, or "redistribution." If the goal is to do as the Nordics do, then the lesson is to do exactly the opposite of what Cathie Jo Martin and Alexander Hertel-Fernandez say to do.

Comments