David Wessel of Wall Street Journal wrote that housing had bottomed. He is wrong. The housing bulls base their claims on past high water mark metrics which are irrelevant because fundamental structural changes occurred to corrupt the value of those metrics.
During 1984-2009 borrowers used “Easy Qualifier” loans to qualify if they could not otherwise qualify for a loan. Those loans have been outlawed so much of the past data of consumer behavior is not relevant. This means that historical ratios like rent to price or rent to income have fundamentally changed and can’t be used to compare with the past. In the past 30 years the labor market has changed workers’ pay from a straight salary to a hodgepodge of unreliable bonus, overtime, profit sharing, stock options, independent contractor work, etc. This unreliable income is often not acceptable by banks for a mortgage, especially during an era of extreme economic volatility. So even if someone’s income in real inflation adjusted terms is as good as it was in previous decades the problem is he may not qualify for a bank loan. And the loss of potential buyers who can’t qualify because they were using now unavailable “Liar’s Loans” to borrow more than they could afford is another permanent problem for housing. These potential borrowers who can’t get a loan will simply have to rent and push up the ratio of rent to home prices beyond normal metrics.
Housing is very difficult to get good data on because when a home sells it could be a better than average home in top condition and that would push up average prices. It is very hard to get data that adjusts the price for amenities such as recently upgraded homes, homes in good school districts, homes adjusted for age or for quiet streets, etc. So simply saying the average home sales price in a certain zip code went up is meaningless. My own house was appraised by an online appraisal with a 20% error because the online appraisal didn’t take these factors into account.
The recovery in the job market is skewed towards the higher income professionals, so these people can afford to play the housing game disproportionately to the moderate income people. Thus upper-middle class homes will sell more than homes in moderate income areas. This skews prices.
Housing economists are pilots doing an instrument landing in the fog with defective instruments
Suppose you are a pilot coming in for an instrument landing in fog. If some of your instruments have suddenly become unreliable and you absolutely have to land at your destination then you must rely on the ones that are reliable and ignore the now unreliable ones. What is still reliable in the housing market is to look at personal earned income. It is stagnant for those in the bottom 90% of the population. Until they start getting significant increases of earned income they simply can’t play the housing game (except by buying a lower quality house than they would normally want) and prices can’t go up. Also the huge overhang of underwater houses whose owners can’t play the move up game and foreclosed homes not yet put up for sale is a gigantic headwind against any potential appreciation. So which is stronger, the headwind of future foreclosures, or the power of moderate income job holders to get an inflation-adjusted raise? I think moderate income people are very weak in terms of economic power and no amount of tax credits, ultra-low rates or improvement in confidence can offset that weakness. The broken instruments are simple comparisons of past ratios of the bubble debt era to today’s no nonsense loan market. It is wrong to say that because housing starts are at record lows that they must go up. They can only go up if a buyer has a better job than his old job that prevented him from buying. The question to ask is has the job market for moderate income middle class people firmed up and improved?
When is a 38% reduction only a 10% reduction?
Another misunderstanding about the benefit of low mortgages rates is that when rates drop the amortization of principal goes up so that the total payment doesn’t drop as much as if it were an interest-only loan. Also the tax refund is less when interest expense is less. The property tax and insurance remains the same even if interest cost decreases. A $100,000 house with an 80% loan at 6% for 30 years would have $479.64 payment, plus about $167 monthly tax and insurance (in California) and roughly $80 monthly in maintenance costs, for a total of $726. If the rate was cut to 3.75% (a 38% cut) the mortgage would be $370.49, and the total payments would be $617, a 15% reduction. To this we should add $37 monthly increase in income tax (because of lower interest cost), so the adjusted cost is $655 versus $726 before refinancing, which is a 10% reduction in after-tax cost. (Note: the amortization of principal is not a cost from an accounting standpoint, but since you have to write a check for it until you sell the house and get back your principle then it “feels” like a cost. Also the amortization example assumes the new loan would be at the same $80,000 balance as the old loan, in reality the old loan balance was probably a few thousand lower due to amortization, but then one must add in closing costs, unless one got a higher interest rate no-cost loan). The point is that a 38% reduction in a mortgage rate may only provide a 10% economic betterment (ignoring the fact that some extra principal is being paid).
This essay doesn’t apply to special places like Silicon Valley, or Manhattan and is for general economic purposes.