I knew I shouldn’t have. I read Kevin Williamson’s article in National Review about the recent housing report from the NLIHC. The report compares each county’s average rents with their minimum wage, and finds that housing is beyond the means of the poor in every county in the US. Williamson used the report as a launchpad for his regurgitation of economic tropes, and he managed to fumble basic mathematics in the process. Yes, ‘someone at the National Review is asphyxiating mathematics while buggering economics’ is already a trope, too. I got this one:

Percentage vs. Percentile

Williamson rests his argument that the NLIHC’s report is “basically useless” on a claim that the report is measuring apples to oranges — specifically, that it is inappropriate to compare average rents to minimum wage, and a comparison between respective minimums would show that rents are actually affordable.

He explains, as if he is truly educating his readers, that the NLIHC report measured the price of rent for the 40th percentile of the rent market, for families who were just moving into housing. Williamson wishes to equate this to the 50th percentile, by observing that new rents cost more than average, because the average includes people who have been renting prior to price increases. He finds data showing that, compared to the set of all rentals, new rentals are “…typically about 6 percent more. So the 40th percentile of rents for families paying a 6 percent premium — that won’t be the dead median, but it will be in the neighborhood.”

Ouch. Williamson forgot that the 40th percentile doesn’t work the same as 6 percent! If Williamson’s conflation were true, then the 10th percentile would have rents that were only a quarter the cost of NLIHC’s data, (‘because 10th is a quarter of 40th’) while the absolute highest rent would only be 2.5 times higher than what NLIHC quotes for each county (‘because 100th is 2.5 times larger than 40th’). How convenient!

In reality, the mid-range rent prices are clumped together. That’s the whole ‘bell curve’ thing that they teach in math class. So, the 40th percentile price is likely already very close to the 50th percentile; each county may have a larger or smaller variation in prices, but not by much. And, while there is plenty of room for high-end prices to soar, the low-end cannot fall below zero. So, the rent paid by the 40th percentile can be much less than rent at the 70th percentile, while the 10th percentile does not see as significant a change in price.

In concrete terms: here in the East Bay, rents for one bedroom are around $800 (yikes!) Let’s call that price the 40th percentile. Now, moving to the 70th percentile does NOT “increase the price by 3/4ths” — here, it is likely a bigger increase. Meanwhile, the 10th percentile isn’t “one quarter the price” — that would be just $200, wow. I wish that Oakland still had rooms so affordable!

So, with the skewed distribution of housing prices, the 40th percentile is not paying significantly more in rent than the 10th or 20th percentile, most of whom are the minimum wage workers mentioned by NLIHC.

That leads to the deeper confusion in Williamson’s write-up. Williamson fails to consider that both housing and pay are heavily skewed distributions, such that matching 10th percentile earners with 10th percentile rents, etc., does not translate to a fixed portion of income for all percentiles. As you near the bottom of rents, the portion of income spent on rent grows. And, for the highest rents, you find occupants who are spending only a fraction of their income renting. As a result, low-income renters have less money to save, and are more likely to need credit after any financial pitfalls. Falling into debt exacerbates the financial squeeze on low-wage earners.

(Rent as a portion of wealth would be a damning metric, displaying the portion of our country that would be homeless or in debt after just one month without income. Williamson’s fudged percentile nonsense couldn’t turn those data into support for his thesis. I would love to see the calculation of rent as a portion of wealth, in an update to NLIHC’s report!)

Williamson later quotes a statistic which exactly refutes his claims: “Household income has not kept up with the rising cost of rental housing. From the housing crisis of 2007 to 2015, the median gross rent for a rental home in the U.S. increased by 6 percent, after adjusting for overall inflation, while the median income for renter households rose by just 1 percent.”

If median costs are rising faster than median incomes, then the median is getting hosed, too. Those are apples to apples. Williamson doesn’t seem to notice: his argument, that the NLIHC study was “basically useless” because it compared median rent to minimum wage, falls apart. Housing is actually becoming unaffordable for most of us.

ROI vs. Gross Sales

Williamson digresses in his argument, hoping to prove that our six-year low in the construction of multi-family housing is not to blame for rising rents. He explains that Walmart makes a lot of money serving predominantly low-wage earners. He then argues that Walmart makes more money than Louis Vutton, and thus, an increase in low-rent housing would be in the market’s interest.

Unfortunately, Williamson missed the business memo: Louis Vutton sees a much higher profit margin than Walmart. And Walmart is huge, not because of the profits available to companies that serve the poor, but because it has comparative advantages to the smaller retailers. If a company seeks profits, it will look for areas with higher rates of return; increasing gross sales while diminishing profit margins is not a winning strategy. That is the problem at root of low-income housing: contractors who build housing for higher-incomes stand to gain a larger percentage profit. In short, low-income housing is not as profitable.

Williamson elaborates that, if developers were not hamstrung by liberal city-dwellers’ zoning laws, then contractors would happily build more low-income housing. No, Walmart cannot save us, Kevin. Walmart’s comparative advantage let it gain market share, by pushing-out other retailers. For a similar behemoth to provide low-income housing, they would need to push-out smaller contractors and pinch pennies with cost-cutting practices, in order to see a worthwhile rate of return. The fundamental point is that contractors already see a lower rate of return on low-income housing, hence the housing shortage that drives up rent! Williamson bleets: “ Why aren’t we building more housing for low-income people? It’s not because there’s no money to be made selling goods and services to low-income consumers…” No, it’s because there is a lower margin for low-income markets, Kevin.

Inflation vs. Allocation

Williamson veers into specious economic policy, arguing against higher minimum wages with his assessment of “ how inflation happens: We value what we value just the way we value it, and introducing more money into the system does not change those value judgments; it just makes money worth less…” On its face, that is true. An increase to the money supply (i.e. printing extra dollars) would cause inflation (i.e. everything costs more dollars). The increase in money supply is at fault. But, Williamson equates a higher minimum wage with inflationary forces — in essence, he argues that ‘paying poor people more would just raise prices.’

No. Pay is not ‘more dollars printed’ — it is ‘dollars that would have been profits.’ When a company earns money, that money is allocated between profits and payments. If pay was kept low, that money would be going toward profits, instead. The rich, following Williamson’s argument, would see an increase in prices for all their goods! Instead, paying workers more means that they constitute a larger share of spending, compared to profiteers. It’s a problem of allocation, not inflation.

And, if workers represented a larger share of earnings, then their demands would become a larger share of market activity. In housing construction, this would play out as a higher volume of housing being built for low-income earners, while the rich would see a lower volume. No goods were added or lost, and no extra dollars were printed. It’s an allocation problem, not an inflation problem. Higher minimum wages make the market listen to the needs of the poor more than the needs of the rich.

Value vs. Cost

The most destructive folly in Williamson’s argument is a conflation of value and cost:

“ You could pass a law that says we have to pay 15-year-old baby-sitters eight times what we pay hedge-fund managers or brain surgeons, but that is not going to change how we actually value their respective labor.”

People are not paid according to the value of their work. They are paid the cost. So, if you can fill a teaching position when you offer $20/hr, the cost of that work is $20/hr. Yet, the value of that work is measurably much greater. Meanwhile, hedge fund managers’ activities have been involved in financial turmoil that caused a loss of value to many — that is called a negative externality, because the loss is borne by people who do not pay the hedge fund. Yet, those hedge funds make money for their clients, and thus, their incomes are far above the ‘value’ they create.

On a deeper level, if workers are more productive, then their work is more valuable. Yet, wages have been stagnant for decades, even as productivity has increased. That is possible because companies are taking a larger share of sales as profits. Workers’ wages are lower in real terms, not because “ They have labor that is lightly valued in terms of everything for which money can be traded,” as Williamson would like to believe. Workers see flat or declining wages because their value is not being translated into their cost.

Williamson argues that neither affordable-housing plans nor higher minimum wages can help the housing crisis; only de-zoning will cure it! He misses the value of the average worker, the lack of incentives for businesses that would make larger margins selling to the rich, and the basics of macroeconomics and arithmetic. I don’t pretend that National Review has ‘stooped’ to such inane arguments; they have always wallowed there.

Written by

Easily distracted mathematician

Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store