Reporting on the Housing Market: Do We Need Drug Testing for Economists?

We all know that economists aren't very good when it comes to understanding the economy. This is exactly what economic theory predicts.

When economists completely mess up, for example by missing an $8 trillion housing bubble, they do not risk any sanction, such as being fired. This means that they have little incentive to get things right. If waiters had no incentive to get orders right and bring food to customers quickly, then people would have to wait a long time for their dinners and frequently get the wrong dish. Since economists aren't held to same standard as waiters, we typically get wrong economic analysis.

Thus we had numerous stories telling us about the bad news on existing home sales. That is not what the data show. May sales were down 1.5 percent from April levels, but this is hardly bad news. We had just seen four months of relatively strong sales, which were spurred in part by unusually mild winter weather. (Remember sales typically take place 6-8 weeks after a contract is signed. The homes sold in May were mostly contracted in March and early April.)

If people buy a home in January or February, they will not rush out and buy another one in March and April. In other words, the relatively strong sales through the winter should have led us to expect weaker sales in the spring. The 1.5 percent falloff is relatively mild. If we look back over a longer period, May sales were 9.6 percent above the year-ago level.

Looking further back, May's sales are roughly 30 percent higher than average monthly sales from the mid-90s, before the housing bubble distorted the market beyond recognition. Since population has only risen by a bit more 10 percent over this period, we are actually seeing a very high rate of sales.

Furthermore, May sales indicated a sharp rise in prices. Monthly price data on existing homes sales are always erratic and are typically driven much more by a change in the mix of homes being sold rather than a rise in the price of homes. Nonetheless, this is the third consecutive large monthly rise in the price of existing homes. The median price of a home sold in May was 17.3 percent above its February level and 7.9 percent above its year-ago level. The price increases show up in every region indicating that this is not a story of high-priced regions seeing an increase in sales relative to low-priced regions.

This report should have been seen as very positive news on the housing market. Unfortunately, the economists who missed the bubble still don't seem to know much about the housing market. While house prices are not going to return to their bubble levels (which no one in their right mind should want), they have bottomed out and will likely be rising modestly through the rest of 2012 and beyond. Furthermore, we are seeing higher rates of construction as the backlog of unsold homes has been whittled away in many areas. Housing is now making a positive contribution to the recovery.

Economists are Credence Goods Who Require Drug Testing
written by Last Mover, June 22, 2012 7:21

Unlike search goods for which information is easily available, or experience goods for which information is available only after consuming them, economists are like credence goods for which information is not available even after consuming them.

Like welfare recipients, testing economists for drugs beforehand would sharply reduce misuse of that received for doing nothing by converting them into experience goods if not outright search goods.

Once economists know they will be outed after the fact, appropriate incentives will kick in to shape up the forecasts into something uselful like a used car that actually works well after buying it.

" a very high rate of sales" - Really?
written by Paul, June 22, 2012 7:45

According to the chart at Calculatedriskblog.com existing home sales are below 1998 levels. The sales trend line has been moving sideways for the past three years.

http://www.calculatedriskblog.com/2012/06/existing-home-sales-in-may-455-million.html

This is the dawning of the age of Aquarionomics
written by david, June 22, 2012 10:05

I always suspected that Fama, Cochrane, Mankiw, Rakan, et alia were under the influence of peyote or LSD. How else to explain their distorted view of reality? But now it dawns on me: money from wealthy donors/sponsors is one of the most powerful hallucinogens on this planet. And they're addicted: the addict always defends their 'connection' (supplier) and usually denies that they are addicted.

existing home sales
written by david, June 22, 2012 10:25

Paul, if you compare the number of units sold now to the middle of the 90's and prior (not to the late 90's when the stock bubble was forming), then it seems the historical trend-line has been reached there as well. Factor out the booms and busts, and one sees existing home sales have reached a healthy level, implying that the housing sector has for the most part recovered and corrected from it's deathly, irrational exuberance of the late 90's and mid-2000's.

But there is one problem
written by EMichael, June 22, 2012 11:02

Somewhere near 40% of home sales in the last quarter were cash sales. That says to me more than half of all sales are going to investors.

Not a good thing.

Inventories are tight
written by Lord, June 22, 2012 12:12

so reduced sales are positive, or at least expected.

Response to EMichael
written by Oarboar, June 22, 2012 12:50

How are cash sales a bad thing? Investors tend to seek investments that will pay off for them down the road. If investors didn't think house prices would go up, they would not be buying houses.

If anything, I think it speaks to the irrationality of lenders: After finding out that lending money willy-nilly is not a good strategy, they've overreacted by tightening lending standards to ridculously high levels. They're like the person who got burnt in a relationship and reacts by staying at home, drawing the curtains, and watching movies in the dark.

-1

...
written by MacCruiskeen, June 22, 2012 2:01

The CR graph shows a SAAR of about 4 million in 1994, and 4.5 million now. That doesn't sound like a 30 percent increase to me. Also, in the early 90s US population was appx. 255 million, now it is 309 million. That's more than 20 percent growth.

Maybe economists should be subject to math tests as well as drug tests.

cash sales don't help if going to investors
written by mel in oregon, June 22, 2012 2:17

the housing sales are similiar to what happened in the 1930s. a lot of people lost their homes, & investors bought $5000 homes for $300. so a lot of people that own lots of property today got their start when old great granddaddy fleeced the poor out of their homes. that's not an improving housing market. interesting that romney says just let the housing market alone, let those that are behind on payments go live on the street. he's basically telling the same story he tells students, i'm not going to help you one goddamn bit. the economy will never recover, there are too many wrong fundamentals that cannot be overcome. to think otherwise is not to have done enough research as to what is really wrong. you won't what you need to know in the ny times or wa post.

Increase
written by david, June 22, 2012 2:36

written by MacCruiskeen, June 22, 2012 2:01 PM
The CR graph shows a SAAR of about 4 million in 1994, and 4.5 million now. That doesn't sound like a 30 percent increase to me. Also, in the early 90s US population was appx. 255 million, now it is 309 million. That's more than 20 percent growth.

Maybe economists should be subject to math tests as well as drug tests.

If you compare sales from 2011 to 1991 (which is closer to the long term trend line) then you get 35.43% increase in sales. If you compare to 1994, you get only a 9.62% increase. Note that this does not take into account the increases over the past two quarters, yet there is still a definite upturn in the graph of sales. Another thing this points out is that one has to be careful to pick meaningful reference points. Dean was comparing to early 1990s figures, not mid-1990's figures, where I think there is evidence that the already-forming internet bubble probably blew a little bubble into existing home sales, particularly near Silicon Valley.

In any event, my guess is that a sleep Dean got his years mixed up, but I think his general argument still works, though it would be useful to factor out the paid-in-cash sales.

Here is my plot of existing home sales since 1969, with a link to my data source: https://dl.dropbox.com/u/8967303/ExistingHomeSalesHistorical.png

underwater average?
written by garth sevdalis, June 22, 2012 3:18

Is there a moving average number for homeowners who are underwater in their loans?

Unstopped Stopper
written by diesel, June 22, 2012 3:20

Glad to see you back, Jonah.

...
written by PeonInChief, June 22, 2012 5:49

A "good" market for economists and realtors would be one that brought the bubble prices back. They don't seem to understand that those prices aren't coming back for a long time. Of course, if they admitted that, a lot of underwater homeowners would decide to walk rather than wait 15 or 20 years.

Response to Oarboar
written by EMichael, June 22, 2012 8:24

Perhaps because investors picking up rental properties shows that average Americans cannot buy houses?

And that the economy is not driven by investors buying properties, but by average Americans buying properties?

Half of the bubble was caused by investors, and the other half ended up paying way too much for their properties because of those investors.

Need a "VIX" for New Single Family Home Sales Series
written by TVeblen, June 22, 2012 8:47

I just did a quick calculation of the growth rates for new single family home sales back to 1963 (see FRB FRED database, "HSN1F"). I understand that these types of homes account for about 1/5th of all sales (can somebody check this). Here are the decade averages: 1963-1969, -3.2%; 1970-1979, 5.9%; 1980-1989, 0.9%; 1990-1999, 3.6%, 2000-2009, -6.5%. Now for the entire period (1963-2011), the average rate of growth is 0.1% with a standard deviation of about 16 points (talk about volatility!). I'm not sure we're going to return to an avg. run-rate of 900K new units, but 550-650K may be the new norm. The point is that new housing sales are, and have been even more volatile than GDP around a basically flat long-term rate of growth. On the other hand, since 2006, the rate of decline has "improved" from -37.3% in 2008 to - 4.4% in 2011. So pushing 30-year rates to below 3.5% is having a positive impact, but until real wages and debt relief improve, I'd say 350K annual home sales will be around for awhile.

...
written by MacCruiskeen, June 24, 2012 7:14

Increase--one could easily argue that 1991 is no closer to the trend line than 1994, but a dip cause by the early 1990s recession. A comparison to the early 90s is just convenient cherry-picking in either case. Also, if one wants to compare the health of the housing market to home sales, then you should also include new home sales, which are still at very low levels, well below what they were in the 1990s.

Reporting on the Housing Market: Do We Need Drug Testing for Economists?

About Beat the Press