Monday, November 2, 2015

Why are low interest rates such a mystery?

I've become a bit obsessed with the whole housing thing.  But, so much of the time, it has a real one-eyed man in the land of the blind sort of feel to it.

The mystery of low natural interest rates has become the topic of the hour, and everyone is perplexed by the mystery.  Why are interest rates so low?  Has anybody mentioned housing?  This virus infected everyone's brain so that we can only believe houses are too expensive or too plentiful, but not the other way around.  So, nobody seems to notice that the return on real estate investments is very high.

I have posted some version of this graph several times.  In the 1970s & 1980s, its tough to get a read on it because there weren't markets in inflation-adjusted treasury bonds at the time.  But, there is clearly a relationship between real estate returns and real interest rates.  Why wouldn't there be?
And this relationship broke down at the end of 2007 when we shut down real estate credit markets.  The lack of access to home ownership made real estate returns go up while bond yields were going down.  This is an important signal of financial breakdown, and nobody seems to notice.
Source

So, no.  Natural interest rates are not low.  The real long term natural rate right now is about 2.5%.  If you have tons of cash or you can run the gauntlet and get a mortgage, or if you are an institituional investor going through the difficult organizational process of buying up billions of dollars of rental homes, you get the preferred rate of 4% real returns.

If you aren't, then you get the "class B" shares of low risk fixed income, which pay about 1% real (3% nominal).

The real estate market is much larger than the treasuries market.  This is a big deal.  This should be just about the only thing anybody is talking about regarding natural rates.  Household real estate is worth about $25 trillion.  If we hadn't stopped building homes a decade ago, and if home prices did not contain an access premium, there would be more than $40 trillion in real estate.  It dwarfs the size of the treasury market.  That's why rates have not reacted to large deficits.  The federal government couldn't realistically accumulate enough debt to make up for the gaping hole we have created in real capital.

We need to make some mortgages and build some houses.  We will not be doing that, because of the virus.  So, it looks to me like we will be wondering about the big interest rate mystery for another decade.

22 comments:

  1. Monetary policy, always and everywhere, should be tighter.

    ReplyDelete
  2. Hi,

    one question: why your theory doesnt take into considerariom about reit risk premium, term premium and liquidity premium? what i am seeing on this chart is just this three variables after 2007. Treasuries for instance are less risk, and reits more risk, less liquid and more term, so thats the difference between interest rates (1% vs 4%). Ehats your opinion? Thanks! :-)

    ReplyDelete
    Replies
    1. Those premiums are all pretty stable over time. There has never been a dislocation like this before.

      The irony here is that even though bubble proponents are typically described as being positioned against market efficiency proponents, bubble proponents like Robert Shiller depend on an efficiency argument for the bust. The idea is that irrational speculators push the price up to unsustainable levels, but they don't account for the fact that all the new supply leads to falling rents, which collapses the bubble. It's a great theory. The problem is that there can't be a supply response in the cities with the most price appreciation, and, thus, a fall in rents did not precede the fall in home prices. This is one of the signals I rely on in my general thesis about the recession. When homebuilding collapsed, rent inflation shot up. That is a sharp challenge to the bubble narrative that I haven't heard anyone address.

      In bonds, the price has a direct mathematical relationship to the yield, and the cash flows are fixed. In the aggregate, cash flows to housing rents are practically fixed. There may be some minor vacancy rate issues during downturns, but a temporary vacancy rate of a few percentage points has a much lower effect on present value than a large shift in yields. And, the thing is, rents didn't collapse. There was no cash flow shock. Homeowners defaulted, but those were a result of the falling prices. The causality went the other way - from falling prices to defaults. Notice that commercial real estate collapsed along with owner-occupied real estate. Their rent cash flows weren't affected by defaults. In fact, they were probably enhanced, because foreclosed households were moving into rental stock.

      So there was no cash flow shock, and yields should have followed bonds. The reason they didn't is because the crisis was caused by public policy and monetary policy that specifically sent real estate markets into disequilibrium.

      Here is a post where I address the idea of a volatility premium after the crisis. But, keep in mind, this premium couldn't have caused the crisis:

      http://idiosyncraticwhisk.blogspot.com/2015/05/housing-tax-policy-series-part-34.html

      Delete
  3. How are you calculating real estate returns and then the natural rate?

    ReplyDelete
    Replies
    1. (BLS table 7.12 imputed rent minus expenses and depreciation, which I create in the graph by adding net rental income to interest expense to arrive at the same figure) divided by total value of owner occupied real estate from the Fed's Financial Accounts of the US report.
      Rental housing tends to have the same behavior, but it is harder to get a number on total property values.

      Delete
  4. What I don't understand is how housing prices can stay so high when the ordinary Joe can't afford to buy a house (wage stagnation, student loans, lack of down payment) and lending standards have become more strict. With less demand, wouldn't one expect the price to come down? A decade ago lending standards were less strict and more people were working, so housing prices went up. Now lending standards are more strict and fewer people are working, so why do housing prices also go up? I don't get it.

    ReplyDelete
    Replies
    1. There is a lot to unpack there. First, it's important to separate home ownership with housing consumption. In terms of rent, spending on housing is very high. That's my point. Home prices are very low relative to rent. If we expanded lending and building, the added supply would bring down rents and intrinsic values, and the added liquidity would push home prices up to their intrinsic value. The net effect would probably be prices slightly higher than they are today, but to the extent that prices are high today it is because of severe supply constraints. We have had a decade long housing depression deeper than any contraction we have ever had before. We need a lot of houses.

      I think it is more helpful, in general, with a financial security, to think in terms of intrinsic values rather than in terms of supply and demand. And I think it is useful to think of home ownership as a financial security where the coupons are in the form of perpetual rent payments. If Apple has a good quarter and their price rises $50/share, thinking in terms of intrinsic value is much more helpful than trying to add up the buyers and sellers that are engaged in the price movement. Houses are the same.

      Also, keep in mind that it is real incomes that are stagnant. A big reason that real incomes are stagnant is that the housing shortage has led to persistent rent inflation. So, real wages are low because we are spending more on housing, not in spite of it. Building houses would increase incomes and it would be disinflationary, IMHO.

      Delete
    2. Thanks for the response. So why have few houses been built in the past decade? Prices were high a decade ago, which induced lots of building. Now prices are high again, so why no building? In my area the housing bust did not last long at all, but the building never came back. I don't get it.

      Delete
    3. Nobody trusts private securitizations any more, punitive regulatory shifts have banks' hands tied, and the public mortgage agencies are timid because they get blamed for the "bubble".

      FICO scores for denied applications are higher now than approved applications were before and during the boom. (Graph at end of this post:) We don't currently have a mortgage system that is capable of funding historical homeownership rates.
      And everyone believes the reason people can't buy is because wages are low. Wage growth is only low because of 3% rent inflation.
      http://idiosyncraticwhisk.blogspot.com/2015/08/housing-tax-policy-series-part-51.html

      Delete
    4. Kevin, do you have any good posts on housing consumption vs ownership? Im generally interested in how best to model housing prices/rent etc.

      Delete
    5. In the last couple of months I have done some posts on Price/Rent ratios and rent levels in different cities. Just click on the archives in the right margin and you'll see some of those posts.

      Way back in January, when I started the series, I thought this was a tax policy story. Several of the first posts were about the various factors in valuing a home. I haven't looked at those in a while. Maybe I would say they are all wrong now. :-) But, I was kind of thinking through the conceptual stuff in those early posts in January and February.

      I think this confusion is one of the key sources of error in the public consciousness about the housing market. Real housing consumption, as a proportion of total spending, has been declining steeply for 25 years even though housing expenses have been stable in nominal terms. It's all inflation.

      Delete
  5. Just read an interesting article in the National Post, here in Toronto, where demographic trends will lead to decades more of low interest rates. Basically, the growing cohort of older folks tend to have spare capital, while the young, who are a falling percentage of the population lack capital and need to borrow. As many of the young are having some amount of difficulty finding jobs that have some security and pay enough to create a viable credit profile, they are constrained from accessing credit. Thus borrowing demand is suppressed while seniors are providing credit desperately seeking yield.

    ReplyDelete
    Replies
    1. I agree that those sorts of factors account for the slight drop in real interest rates from the late 1990s to the 2000s, but since 2007 long term real rate movements have been much more volatile than normal. The timing and scale of the movements point to the problem of the crisis as the cause. But, even if the housing issue was fixed, I think you are right that real long term rates would still be lower than usual.

      Delete
  6. I'm not sure I understand the answer to your main question...why are interest rates so low? You are saying that many people are unable to buy houses so they put money into fixed income securities instead, depressing income yields??

    ReplyDelete
    Replies
    1. Yes. But, this is an emergent phenomenon. You can't think of it in terms of a family making allocation decisions. The various low risk asset classes that form the bulk of developed economy capital tend to move with pretty regular spreads, reflecting risks, liquidity, etc. We tend not to look at home equity this way. I think that is partly because it is a real asset (like TIPS) and before TIPS bonds there weren't any real market yields to use for a benchmark to home equity. Plus, we just tend to think of homes in terms of prices instead of yields. But, if you do think of homes as securities with a yield, which is very easy to do with BEA and Federal Reserve aggregate national data, they have behavior very similar to these other classes. Since they are long term real assets, their yields tend not to move much. Since WW II, they have fluctuated in a range of about 2%. Since the crisis, real bond yields have collapsed. I don't see enough appreciation in the public conversation about how outrageous these levels of yield fluctuations in long term real bonds are. But, this has happened to coincide with the collapse of the mortgage market - which I consider to have been wholly self-inflicted and unnecessary. Since that time, there has been a sharp divergence with home equity yields rising sharply, while all the low risk yields that aren't characterized by an overbearing gate-keeper context have collapsed in an unprecedented way. This is a huge red flag. And, there is no mistaking the various egregious obstacles to access to mortgage markets that have developed since 2006. The causes seem obvious to me. And the size of the housing market is large enough to dominate and create this effect.

      Delete
  7. Hi Kevin, I've just come across your blog and the housing/depressed yields links. This comment this is a helpful overview - can you point me to one or most posts or papers that develop your argument more fully?

    ReplyDelete
    Replies
    1. Kevin, my comment just now refers to your November 5, 2015 at 9:48 AM comment.

      Delete
    2. It's such a complex, subtle, and large issue, I don't know if there is a single post that covers it. I am hoping to be able to publish it in old-school dead tree form, because that's the only way I think I can organize it and illuminate the whole picture.

      I have been doing an extensive series on housing - part 79 just posted today. This idea really comes out of that work. Your best bet is probably to just work backwards from part 79. I didn't really figure out the supply problem aspect until around the mid-30s, so if you get that far, you've probably covered most of the evidence. Part 65 is probably a good narrative version of the problem and how we were fooled. Parts 8 through 12 were the first recognition I had that the bubble wasn't all it was cracked up to be and that the data don't match the narrative that says financial excess led to a bubble that led to a crash.

      One aspect that I didn't get into in this post is that rent inflation has been high since the mid-90s and is clearly the product of local supply constraints, not of monetary expansion. So, monetary policy has been tighter than we thought it was. In 2008 it became disastrously tight, and capital markets since then have been in disequilibrium.

      Sorry, that's not the answer you were probably looking for, and really this point about interest rates is simply a tangent from the deeper project of looking at housing. But, I think most things now rely on a proper understanding of housing, which is rare IMHO.

      Delete
    3. There is a great deal of demand from governments and central banks for explaining low post-GFC interest rates. Your idea connecting housing supply constraints to depressed interest rates has some intuitive appeal, but the argument has to be made out. There is a major prize if you can succeed. Think Friedman and Schwartz 1962. Think Nobel prize.

      Delete
  8. Hey Thanks for sharing this informative blog, i was looking for same kind of content about Home Loan

    ReplyDelete