Austrian business cycle theory continues to make a comeback

Just read the abstract from the latest NBER working paper by Òscar Jordà, Moritz HP. Schularick, and Alan M. Taylor:

Is there a link between loose monetary conditions, credit growth, house price booms, and financial instability? This paper analyzes the role of interest rates and credit in driving house price booms and busts with data spanning 140 years of modern economic history in the advanced economies. We exploit the implications of the macroeconomic policy trilemma to identify exogenous variation in monetary conditions: countries with fixed exchange regimes often see fluctuations in short-term interest rates unrelated to home economic conditions. We use novel instrumental variable local projection methods to demonstrate that loose monetary conditions lead to booms in real estate lending and house prices bubbles; these, in turn, materially heighten the risk of financial crises. Both effects have become stronger in the postwar era.

The piece is called “Betting the House,” and you will find some non-gated copies here.  And to my Austrian-oriented readers, please don’t let this evidence push you away from more synthetic accounts of what is going on…

Here is FTAlphaville coverage of the paper.  And here is an interesting paper on interest rates and equity extraction.


It's ironic that this post is right after the post about Scott Sumner's work. The framing of interest rates here is questionable. The fact that they find long term interest rates falling along with short term rates suggests that in many cases, natural rates are falling and the Fed isn't lowering policy rates fast enough to remain neutral. Right off the bat, the idea that housing "bubbles" happen because low long term rates are a product of loose money is very debatable.

All of this poor framing leads to a narrative that is 180 degrees wrong. Rent is consumption. But home purchases are investments. I find it very strange that homebuilding booms are treated as some sort of unsustainable consumption bubble. It is the product of households taking exposure to relatively stable income producing long term, inflation protected assets. Quite the contrary to debt fueled consumption, this is a flight to safety and long term asset commitments. The evidence of overinvestment in homes in the 2000s is questionable, itself. Mortgage/Equity levels weren't rising until the bottom dropped out, and rent inflation was fairly strong through the period. How does rent keep rising if we are overbuilding?

The problem is that we don't frame home prices like we do other investments. Homes are basically like a TIPS bond that receives payment adjustments on local rent inflation instead of the CPI. But we don't all flagellate ourselves about bond bubbles because we use a simple mathematical framework to conceptualize bonds in terms of yields. Low real bond yields are also a good sign that poor economic outcomes are around the corner. But, we don't have piles of research papers out there claiming that households are dangerously speculative in TIPS bonds because monetary policy is too loose.

Here's a recent post I did where I was thinking through this problem of framing prices and returns, in case anyone is interested.

and here:

If home purchases are investments they are crappy ones. Same home sales show no appreciation over long stretches of time and after taxes, insurance and repairs housing is a net loser.

30 year TIPS bonds have spent most of the past few years with yields under 1%. Long term home returns don't come from price appreciation any more than bond returns come from increasing face values. And, be careful comparing mortgage payments to rent, because rent isn't inflation-protected.

Past few years =! decades of data. Yields under 1% =! negative nominal yeilds.

Your post linked (skimmed) there makes a few conceptual mistakes. You extend analysis of rental properties to occupying your own home, in short this is akin to stating that because some cars are purchased as investments (taxis) that all car purchases can be considered investments. Your specific case does not extend to the general.

Aren't all car purchases investments, unless you drive it off a cliff or something?

@Brian- how so?

A car produces a stream of value over time, typically measured in years.

Mistake number 2 is not acknowledging substantial differences between owning and renting. Probably the most important is the ease and speed a person can get out of most leases relative to selling a home (especially when selling costs are considered). There are many people who value "liquidity" in the housing market like this- those who might change jobs soon or frequently, or those who expect their family size to change are two obvious ones.

Lastly houses do not function like commodities the way bonds do. There is far more personal preference in the housing market, not only in the physical house but in the location of that house and its proximity to work, schools and social opportunities. Considering a house purchase as paying rent to yourself is a poor assumption as it is frequently difficult to impossible to rent an identical property.

Surely you could rent your house at some price. Opportunity cost!

I've long suspected that the median homeowner return greatly lags the Case-Schiller index. Housing is a highly illiquid market with a lot of adverse selection. When you go to buy a home as a normal consumer, you're almost certainly overpaying. Professional investors watch the listings like hawks, know the comps backwards and forwards, and quickly close on any undervalued properties before you have a chance. If you suddenly find the neighborhood you bought gentrifying or getting trendy, almost certainly savvy developers will realize it before you know to raise the price.

When you make any improvements to your home, you very rarely see positive NPV. Home flippers and their contractors know how to make cheap low-quality improvement that look just as good as yours. Home inspectors very rarely have the one-time buyers interest at heart, and will most likely overlook any problems so the realtor (repeat business) closes the commission.

Are you comparing them with other investments *you can live in*?

A purchase does not become an investment by virtue of being a cheaper version of a different purchase. Buying a case of soda and then returning the empty cans for 5 cents each doesn't turn your consumption into an investment.

An investment is a good that is used to produce other goods or services over a lifespan. A house produces a place to live and the homeowner can either use that service themselves or sell it in the market.

Cars can be considered investment but their useful lives are much shorter and a case of soda is shorter less. Generally imagine that all investments eventually 'decay' over time into consumption. If that decay period is very short it's usually easier to treat them as simple consumption.

The food I eat gives me energy to gotowork, the clothes I wear keep me warmso I have to eat less fod, the golf I play on the weekend keeps me fitter/healthier which leads to higher productivity, going drinking with collegues creates bonds that might lead to future opportunities and better working conditions. Oh my god there is no consumption, only investment!

A house does not produce goods and services. A house IS a good, it does not produce baby houses. Saying a house produces place to live is like saying a piece of cheese is running a perfect simulation of cheese.

Or we can put it this way- the arguments that purchasing is an investment apply to renting. A rented house also produces housing, and now there is no distinction at all and we are talking time preferences only between renting and owning.


I get your gut feeling on the subject but the accounting just doesn't work. A house itself is not a good like a steak dinner. You consume the steak and it is gone, if you want another one, you have to go buy it. You don't consume a house in this manner. You do live in it and that type of consumption is like a steak dinner in the sense that it can only happen in the present moment. The House is an investment asset and the good produced by that house has to be providing somewhere to live. What is the value of that good? Whatever market rents are in the area.

Now if you don't consider houses investment assets, then what do you consider them? Consumption goods? But when does the consumption happen? When you buy the house? When you sell it?

Or we can put it this way- the arguments that purchasing is an investment apply to renting.

Last year my wife brought a condo, we stayed in it for a few months as a weekend getaway. As the year came to an end, she rented it out. So how do you compute this into GNP? If we used the condo for two months, let it sit 8 months, and then collected $1K per month rent for the last two months of the year does the house swtich from asset to non-asset and back again?

And consider the problem that you have to live somewhere. You have a house and you live in it. Your neighbor makes the same income and he has two he lives in and the other he just keeps empty...meaning to use it for vacations but he doesn't get around to it. Does it make sense to consider you both as making the same income really? Even if he doesn't use it, doesn't he in a sense get something from owning that 2nd house that you are missing? The way to square this accounting circle is to treat the house as spitting out 'inputted rental income' each month. You enjoy that income but then 'pay it back' in the sense that you are renting your house from yourself.

The way to square this accounting circle is to treat the house as spitting out ‘inputted rental income’ each month. You enjoy that income but then ‘pay it back’ in the sense that you are renting your house from yourself.

IOW, it's a wash.

Houses are durable consumer goods, like personal autos. It's your realtor and car dealer who would like you to think otherwise.

Great comment and post, Kevin.

"Rent is consumption. But home purchases are investments."
This is a completely arbitrary distinction which makes no sense in the real economy.

"The evidence of overinvestment in homes in the 2000s is questionable, itself."
Have you seen Spanish ghost cities?

"Homes are basically like a TIPS bond that receives payment adjustments on local rent inflation instead of the CPI."
Have you ever heard the term 'vacancy rate'?

Umm, Kevin, you seem to think you're contradicting somebody, but it's not clear who you think you're contradicting.

Austrian business cycle theory is all about the central bank artificially lowering interest rates below where markets would set them, thus sending an errant signal that there is ample savings and light consumption demand, thus encouraging excessive long-term investment and insufficient investment in meeting short-run consumption demand.

That's not exactly what Jorda et al are writing. I bet they hate being associated with von Mises, who actually had in mind excessive investment in lengthening industrial production cycles, not real estate. But hey, he was writing 100 years ago, the world has changed. If what he wrote then kinda sorta fits the story of modern real estate investment cycles, why not give him some credit. That's what I understand Tyler to be saying.

As for what housing bubbles have to do with consumption bubbles and how investment in housing can be excessive despite rising rents -

Housing bubbles are spiraling credit bubbles in which growing credit volumes drive rising housing prices which increase the value of collateral driving growing credit volumes. The growing volumes of credit aren't only spent on home-building, they also leak into other investments and into consumption. The McMansion as ATM story, you know it. Meanwhile easy credit drives more household formation than incomes can sustain, goosing housing demand. Nevermind Spanish ghost suburbs, look at the collapse of US home-building since 2006. What's your theory, that Kevinonomics could have sustained the 2000s housing demand surge?

There was no bubble. Prices weren't especially too high. We didn't build too many homes in the US. Monetary policy was relatively tight during the "bubble".

Here is today's post at my sight.


As for there being no bubble, you are indeed crazy. Certifiably koo koo for cocoa puffs. Rental yields went negative across wide swathes of the country. Loose credit standards put millions of people into homes they would only be able to afford if credit continued to loosen and further drove price appreciation. Once the credit standard loosening maxed, it crashed.

We built way too many single family homes and that is evident from the very slow pace at which they have been built since.

As for monetary policy, you have a point that the Austrian focus on central bank policy is overstretched. Interest rates were low initially but not during most of the bubble. But credit was extremely loose. For the real economy, what matters is the availability of credit from lenders to the real economy - commercial banks, subprime lenders and so on. Low interest rates and rapid base money expansion are likely to loosen commercial bank credit, but it's not an exact relationship. Still even in this case you could say central bank policy was loose in the sense that the Fed is the regulator of commercial banks and the Fed was supportive of the subprime boom.

Just noting that there are now 2 posters named "tom" replying to Kevin, to try to avoid some seemingly inevitable confusion.

More points

"Mortgage/equity weren't rising until the bottom dropped out"- this is just a bizarre metric to use- M and E are both related to price, the deviation from historical pricing (case-shiller) is identified as the root problem, and the return to mean exposed that E was inflated. Saying M/E was fine until the price decline is like saying "My cars speed to my speed stayed constant until we hit that tree". The important ratio is your cars speed to the road, not to you, if your car goes fast enough a tree is inevitable.

The US market appears very mature and tame compared to the other developed countries included. Not much to see here for Uncle Sam.

^Scott would also say interest rates aren't a good indication of the stance of monetary policy.

As for the article (I haven't read it yet, but will), I am slightly biased in that whenever I hear someone say that they have a novel instrumental variable design I get a little concerned. However, this does seem like an interesting approach.

Someday I gotta learn how to read an NBER abstract like Tyler does, having declared that he hadn't read underlying paper, so I too can make a declaration as strong as the one headlining this post that "Austrian Business Cycle Theory Continues to Make A Comeback."

Here is a link to what I think is an earlier version of the paper by the Fed:

I havent scanned the paper for the word Austrian. Yet. But, it does contain the word "macro prudential tools." which I find hard to believe are in the Austrian toolbox.

sometimes astrology is popular, too

I prefer astrology to almost implying the attacks from last week were Obama's fault (like you say in your blog).... Shame on you sir...

Off topic a bit, but what is interesting is whether the crash due to housing and finance in 2008 was different than the crash from the dot-com boom in 2001. Blurb below, which suggests the amount of money lost --about $7T in the USA--was not that different in both episodes. This means that either (1) financial crashes are different in kind from ordinary crashes (the Rogoff et al thesis), or, (2) the funk we are in is related to a lack of AD or confidence rather than fundamentals. - RL

Stuff I've scraped off the net: "Schiller told a late-August 2007 conference that home prices in some cities might fall by as much as half if the gathering bust could not be contained. If so, losses by U.S. homeowners could reach $10 trillion, more than the $7 trillion lost in the 2000-2002 stock market bust led by the decline of the technology-heavy NASDAQ index.” “At the peak of the housing bubble in 2006 Americans had $13 trillion in equity in their residential real estate. At the peak, total residential mortgage debt stood at $9.8 trillion. Today, American households have $6.2 trillion in equity while mortgage debt has grown to $10.3 trillion. In other words American households have faced a real financial loss of $6.8 trillion. ”

Dude, it is called "oil". A funk, we are not in. You don't get it and never have.

The US economy did very well in 2003 growing at 4.4%. That rate should have been continued through 2006 with the blowoff from the housing boom, but it didn't........because of the expense that oil drove through the economy.

Thus, the bust was made worse and made growth tougher afterward.......until now.

The problem with housing prices dropping is that you still have to pay the money that you borrowed back. Similar to buying stocks on margin. Losing the value of your investment is one thing; you adjust to simply having less wealth. But if you owe more than your house is worth you are stuck. That kind of mess doesn't resolve itself easily.

Of course, the measures adopted in the financial crisis (in particular monetary stimulus) have recovered the losses in financial assets, and then some. As for housing, not so much. The (monetary) tools available to central banks work very well in financial markets, but not so well in other markets. It's those "other markets" to which fiscal stimulus is directed. The ongoing debate about monetary stimulus and fiscal stimulus is important, but not so important as the identity of those who are benefited. In an economy in which the financial sector is relatively large (absolutely and historically), central banks will necessarily focus on it in times of financial crisis (for the same reason Willie Sutton robbed banks - because that's where the money is). If a relatively large financial sector correlates with financial instability, however, the central bank will be preserving that which contributes to financial crises.

Hard to see how this is Austrian rather than post Keynesian.

Interestingly, though the paper shows

Loose Monetary Policy -> Lower Long-term interest rates + Mortgage & Housing Increases
Mortgage & Housing Market Bubble -> Financial Instability

but it never shows

Loose Monetary Policy -> Financial Instability

I think this is because exogenously low interest rates increase the discounted value of future imputed rent, and also increase the value of rent-inflation hedging. I write more on this:

Comments for this post are closed