Data behind a wealth tax

Edit: There are quite a few negatives that I didn’t address, as Evan Soltas points out. I think (1) and (2) are sufficiently mitigated by an exemption and amortization of tax burden over time, but the coordination of this policy definitely requires some more thinking on my part.

I watched a shocking video earlier today. We all know the story behind wealth inequality in the United States, but it’s good to be reminded just how divided our nation is. I’ve long argued that an income tax system should be replaced with a wealth tax for several reasons:

• Donald Trump supports it (kind of)
• It’s easier to design a simple tax code
• Wealth inequality is kind of like the integral of income inequality. Because those with higher incomes save more, even if the GINI remains largely constant over a period of time, a lot of wealth will accrue in the hands of a few

But there’s another big, structural, reason why a wealth tax makes more sense for the 21st century:

I graphed what economist’s call the “wage share of GDP” out of curiosity. Take a farm that sells wheat. Before the industrial revolution, there might be many farmhands who would each earn a wage to work the farm, producing n kilograms of grain. Assuming this farm has no capital, the wage share of total income is 100% (distributed in some manner among the farmers). Now consider the landlord buys a tractor, allowing him to save 60% of his outlays by laying off the worst farmers. Because it’s a nice tractor, he can still produce n kilos of grain.

Economists would say the wage share of income has fallen to 40%, and the capital share of income has increased to 60%. You don’t need to be Nancy Drew to figure out that this is broadly what’s happened to the American economy with the advent of ever-more-powerful technology.

This is a good thing, but needs to be handled with care. A worker is by definition the owner of his labor (and hence the income derived thereof), but the capitalist is the owner of the machine (and hence the income derived from technological advances).

Where does this tie in with wealth inequality? By saving money in mutual funds, companies you think will succeed, and even a bank you are accumulating capital. And yet, in 2010, the top 1% owned 35% of all mutual funds and 61.4% of all business equity. 

Wages will continue to fall as a percent of GDP. If they didn’t, it would imply that natural evolution can somehow beat human ingenuity. It took millions of years for nature to create sentient life, and yet glimpses in the long years of human history for us to create computers.

In light of this trend, income taxes will become increasingly irrelevant. But what are the numbers behind a wealth tax? Is it sustainable? Would it erode wealth? Could it really replace income taxation?

These are questions that I can’t answer, but I can provide a glimpse of why I firmly believe it will work. Using FRED, I graphed the total financial assets held by households and nonprofits as well as the % year-on-year change:

Several things are pretty clear. Wealth has almost always increased year-on-year, with the exception of the dot-com bust and the subprime mortgage crisis (a quick aside here to direct interested readers to The Banker’s New Clothes, which explains why the latter created a long recession and the former was a blip. Hint: equity is safer than debt).

The data looks pretty noisy so I generated a histogram of year-on-year changes:

It’s not really a normal distribution, but one can see pretty clearly that wealth usually grows by more than 3% a year. In fact, it’s a very smooth exponential curve, with an average growth rate of around 7.3%. The ten year average on wealth-growth has remained largely unchanged so taking this estimate, I generated this projection of wealth through 2023:

Please note that value is measured in billions of US dollars. To determine the wealth tax rate needed to replace the income tax, I borrowed the CBO projections through 2022:

Total income tax revenues is projected at 21.3 trillion US dollars. At an average wealth tax rate of 2.5%, the corresponding figure would be 20.6 trillion US dollars. Pretty similar. (And it’s growing faster, as well).

There should be two modifications to a wealth tax:

• Although IRS should aim for an average tax rate of 2.5%, it should be progressively implemented. One way to do this would be to exempt the first 50% of all wealth generated for society as a whole, and provide those exemptions on the first k dollars of overall net worth, and tax everything above at 5%. This is not purely for egalitarian reasons, but to incentivize saving. If all wealth is taxed, there is an implicit subsidy of consumption, discouraging long-term savings. For example, if k was 250,000 US dollars, the only people who would face a disincentive to save are those with already high savings rates, with a low marginal propensity to consume.
• Wealth, as estate, taxes are a tricky business. Especially when much of one’s net worth is in form of home equity and similarly non-liquid assets, it might be difficult to pay an immediate 2.5%. Government can introduce the option to amortize the cost over 5 to 10 years, removing the immediate burden on a family.

I hope this post has demonstrated that a wealth tax is fiscally feasible, and socially egalitarian in the face of an ever-mechanized economy. Incidentally, the United States is one of the few nations that can really succeed with this kind of a taxation system because of its international jurisdiction on assets. France tried, and failed – but to be fair, France also had a highly-progressive income tax scheme which doubly-burdned the affluent. I’m proposing a complete replacement of a highly complex system.

A wealth tax would also be well-supplemented with a progressive consumption tax on carbon, sugar, and cigarettes – balancing out any non-market incentives to save or consume, while at the same time ridding the country of associated external costs.

I truly believe this reform can achieve a bipartisan consensus. Much of the country would be grateful to be rid of an awful income tax system. Indeed our tax code is so complex, that accounting costs represent 4% of overall revenue (enough, for example, to achieve the Simpson-Bowles goal without increasing tax rates). Of course, since I hear very little noise in the news world of any such reform, I doubt it will come to pass – but I would love to hear some debate.