The details of anything are tough. Sure, an omelette is just eggs’n’milk’n’butter’n’stuff and yet the number of people who can cook a good one is pitiably small. Baking is entirely a matter of chemistry, get the quantities right and it will happen – yet many cakes still fail to rise. The details of economic statistics are such that having a stab at playing with them is something that doesn’t really work.
As here with Matt Bruenig’s attempt at dissecting wealth in America. There’re a couple of thing’s he’s forgotten.
Here’s a statistic to get your class rage going: since 1989, the top 1 percent’s net worth has skyrocketed by $21 trillion. And the bottom 50 percent’s? It’s plummeted by $900 billion.
To start with, let’s recall the basic fact about wealth distributions anyway. The bottom 50% never do have very much of it. We can get this from Saez and Zucman and they’re not exactly right wing apologists. This is a reflection of the manner in which we measure wealth plus lifetime cycles. We are measuring *net* wealth, meaning assets minus liabilities. Then we’re also seeing the effects of the life cycle. Most of us enter adulthood with assets, sure – a living family, and education, human capital – all of which we don’t count as wealth. We then purchase assets along the way and a natural outcome is that our wealth peaks at about the time we retire. Pensions and mortgage free housing are indeed wealth.
Even if wealth were entirely equally distributed over the population’s lifetimes we’d still have perhaps half the population with none – all those under 30 maybe. OK, that’s too extreme but you get the point. An inequality in wealth without taking account of life cycle effects means precisely nothing in and of itself.
OK, then Bruenig takes us through how he’s done his calculation:
To derive this, I initially take the nominal net worth aggregates for each wealth group that are provided by the Federal Reserve and subtract out consumer durables. Consumer durables are things like cars and fridges that many academics who work on wealth distributions do not consider wealth. The top 1 percent owns around thirty-two times as many consumer durables (in dollar terms) as the bottom 50 percent owns. So the subtraction of them reduces the inequality between the top 1 percent and bottom 50 percent. From there, I adjust the 1989 figures to 2018 dollars using the CPI-U-RS price index. This is what the Federal Reserve also does to adjust wealth figures over time in its Survey of Consumer Finances reports. What the final product reveals is a 2018 where the top 1 percent owns nearly $30 trillion of assets while the bottom half owns less than nothing, meaning they have more debts than they have assets. This follows from thirty years in which the top 1 percent massively grew their net worth while the bottom half saw a slight decline in its net worth.
We expect the bottom 50% to have little net wealth anyway – S&Z tell us it’s unusual for them to have more than 5% in any known distribution.
Bruenig’s result is worse than that. Why?
Because he’s taken out consumer durables from that measurement of wealth. But he’s left in there the debt used to buy consumer durables. Wealth is a net concept, right? No, a car and a house aren’t the same thing, obviously. But if we took houses out from our measure of wealth while leaving mortgages in our measure of debt then our net number is going to be – well, politely perhaps pretty screwy – useless as a guide to anything, isn’t it?
Bruenig’s number just doesn’t work. It’s as silly as measuring the wealth of young people by including their student loan debts but not adding in the value of the degree they’ve attained with it, nor the future higher income they’re going to gain as a result. Oh, wait, we actually do that, don’t we? OK, so it’s as absurd as adding in all the social security taxes people pay over a lifetime as costs to them but then refusing to recognize the value of the SS pension at the end as an asset, as wealth. Oh, wait, we do that too, don’t we? OK, so it’s as foolish as demanding that we have more welfare state as that makes us richer, truly so, then refusing to include the value of the welfare state in a calculation of how rich we all are. Hmm, no, we do that too.
So Bruenig’s doing something quite common then but it’s still not something that illuminates, is it? Slicing out some of the assets while keeping in the debt used to purchase them just isn’t going to give us a useful measure of net wealth. Even if everyone does do it.
And no, sorry, we can’t say that whatever the problem is it was the same in 1989. Because we do worry about auto debts these days, don’t we? In a manner we didn’t used to? That is, more of the auto fleet is bought on credit these days – if worrying about that sub prime auto debt is something realistic to worry about that is. Consumer debt is greater than it was in 1989? Much of which will have been incurred to purchase consumer durables?