Thursday, September 4, 2014

Speculative Position in Housing

I have written many posts regarding my bullish position on home prices.  This early post on the topic probably gives a decent overview of my narrative.  So, I won't recap my thesis here.  My starting point here is simply that home prices in the 2000's were not as excessive as is widely imagined.  Low real and nominal interest rates led to a boom in nominal home prices and that led to excesses with regard to low down payments, subprime mortgages, bank speculation, etc. (not the other way around). Prices might have moved into an unjustifiable range toward the very end of the episode, but for the most part, price stickiness in real estate was preventing markets from clearing efficiently.  The large amount of speculation going on at the time was a product of sticky prices that were slow in catching up to intrinsic values.  (This is where we should expect to see speculators.  Part of what creates a false narrative about the housing market in the 2000's is the perverse colloquial belief that speculators are largely engaged in pushing prices away from intrinsic values, as opposed to collectively making markets more efficient.  People who understand markets tend to understand this common error explicitly, but the power of our aesthetic sensibilities on these matters is such that the error seems to sneak into common interpretations of events like this by informing priors without ever being stated explicitly.)

Here is a graph of Home Prices relative to Rents and Mortgage Rates.  I have discussed in my previous posts how most of the rise in prices could be justified by the low long term interest rate environment.  Average New Home prices and the Case-Shiller 10 City Index have markedly different behavior in the 2000's.  Please comment with any insights that you have about this or links to discussions.  I'm not sure that I completely understand the causes for these differences.  In either case, I am not that interested here in arguing for a specific value.  Only in pointing out that home prices have a lot of room to move up rationally.  Long term interest rates could rise nearly 2 points without moving above the low rate environment that was in place in the 2000s.

Here is a graph comparing relative mortgage payments to rent.  (A caveat: over very long time periods, differences in the underlying adjustments of these data series may cause some divergence.)  Mortgage payments are extremely low compared to past relative levels.  Keep in mind that while lower inflation premiums should lead to lower mortgage payments, lower real interest rates should actually increase mortgage payments, relative to rent.  Even if mortgage rates increased back to 6%, with no change in home prices, mortgage payments would still be lower than rent, indexed to 1987.

Current Home Price Trends
Rent shows YOY change, Home Prices show MOM change
Home prices have been recovering, generally over the past couple of years.  Some of the more sensitive indicators are showing some leveling out.  Here is the Case-Shiller 10 City Price Index (both seasonally adjusted and not) and the CPI for rent.  There are some difficulties with the decline of distressed sales and its effect on seasonal adjustments.  It generally appears that home prices are settling down to single digit annual increases.  But, it should be noted that all of the post-crisis growth in home values has come from institutional and cash buyers.  Mortgages have been stagnant since early 2008.  Nominal home prices, at least according to the Case-Shiller indexes, are roughly back to where they were in early 2008.  And, the long-standing typical balance between mortgages and equity values has roughly been re-established.

This is where QE3 was so important.  Its effects on bank credit, inflation, and real economic growth may have been muted, but, since the 4th quarter of 2012, household real estate market values have increased by about $3 trillion, and this was likely goosed by the added liquidity.  This greatly reduced household real estate leverage.  Without this boost, household credit markets would be dead in the water.  But, QE3 basically carried us back to a place where household real estate markets might be able to achieve sustainable growth without further infusions of cash.

Now that housing has recovered from the demand/deleveraging crisis, we should see a sort of self-healing circle as mortgage and equity grow together, which will allow traditional home owners to bid prices up to their reasonable levels.  To the extent that home prices have slowed their re-ascent, I think this is a temporary lull as mortgage growth kicks into gear.  It would be quite normal for mortgages and total real estate market values to enter a period where both are growing by about 1% per month.  Some of this will play out as new home production and some will play out as price appreciation.

I need to clean up my data a little bit on the homebuilder revenues,
but this is pretty close to the consolidated outcome of the public firms.
For the purposes of a position in homebuilders, I'm not sure that it matters that much what the divide is between price appreciation and new home construction.  Total home equity and total market value of household real estate seem to correlate fairly well with revenue growth among the home builders.  So, for the purposes of this proposal, I am going to avoid the problem of forecasting new home starts, rent inflation, and the price to rent ratio.  These factors will settle into an equilibrium reflecting the factors I have touched on before.  I am going to rely simply on the historical tendency for total household real estate market values to grow in the range of 10% to 15% per year during periods of expansion, and I will use this value to forecast home builder revenues.  This range probably isn't accidental.  This probably reflects the limited ability of the housing market to reach new value levels because of issues like production bottlenecks and sticky prices.  I am convinced by my previous housing analysis that the price pressures on homes are strong enough to continue to push market value growth to that level.

I had originally tried to conceptualize the home builders in our current volatile context as a sort of stable building operation sitting on top of a big pile of real estate assets, so that, in this context, they would effectively be land speculators, even if they don't tend to operate or view themselves as such.  But, I could not find any systematic way that large fluctuations in land values were moving into their bottom lines.  The fluctuations in the real estate market came to the home builders' bottom lines through higher volumes and some margin expansion.  As much sense as it makes to me to look at a firm, like Hovnanian for instance, as an incredibly leveraged option on land appreciation, for some reason that I don't understand, it seems like the bottom line really is a product of operations, and profits seem to come from revenues in a pretty straightforward way.

So, as a basic framework for looking at a position in the industry, I would start with an expectation that by 2016-2017, total homebuilder revenues will be 75% - 100% higher than they currently are (which correlates to total household real estate 30%-40% higher than the current level).  Keep in mind, while I think a 30%-40% rise home price-to-rent ratios from where we are is not outside the realm of reasonable expectations, the total increase in household real estate is a combination of rising rents, rising price-to-rent, and new home production.  So, this forecast could play out even if we just see an increase in the teens for Price-to-Rent ratios.  I would say that, given current trends, something like a 4% increase in the housing stock, a 10% increase in rent, and a 15% increase in Price-to-Rent, over a 3 year period is actually a fairly conservative forecast.  And, that would get us into a 30-40% range for total real estate values.

I think this is more optimistic than the typical forecast for homebuilder revenues.  I am basically forecasting a positive outcome fairly far out on the probability distribution represented by current market prices.  This forecast can be very specifically targeted by taking a highly leveraged position.  Looking at a firm like Hovnanian, which has very high financial leverage that in many cases is funding claims on land that are, themselves, options, and where I can take a position in out-of-the-money call options on the firm's equity - I can take a position that is leveraged to the third power, and each layer of leverage is really high leverage with insurance on the downside.  I love these sorts of positions, where I can isolate my speculative idea and expose myself to tremendous upside with very little downside.

I will follow up with a second post tomorrow.


  1. TravisV from TheMoneyIllusion comments section here.

    Fascinating stuff, thanks a million! I find myself drawn a bit more to companies like Lennar and D.R. Horton. Their trailing and forward P/E ratios look a lot more reasonable. D.R. Horton's stock in particular has significantly underperformed the S&P 500 over the past two years.

    1. Funny, I think those may be the worst two home builders in my final analysis. But that's probably because they are fairly strong and stable, relative to the other builders, and my framing of the position basically favors a "low quality" firm. So coming out badly in my analysis is probably a good sign for a decent safe investment. I'll have more tomorrow.

  2. Not sure I agree, but then I don't disagree. How is that for resolve?

    I do think this: As America continues to urbanize, and as incomes rise, you see real estate values rise. Ergo, housing is a good investment in densifying cities, such as Los Angeles.

    Of course, housing was a honking buy in 2009, and I am on record as saying so. Now, still a buy in the right places.

    For the longer run, a lot depends on both national and local policies. Immigration is big, and so is land-use zoning and schools in urban areas. Really, urban cores should be the most valuable land, but this model got turned on its head in 1960-80s, when middle-class Americans fled cities due to crime and schools. Now, in a city like L.A., the lower-class is being pushed inland. From downtown to to the sea, and then well inland, will be middle-class or above in 20 years. You can't miss.

    Of course, zoning can create scarcity but also cramp growth, so that cuts both ways, city by city.

    But, given the tax code, I think most people are compelled to buy a house. If you can buy a lot or house and carve-out a "secret" rental, or trailer, then it is a very good investment indeed.

    Given that stocks and bonds are fully valued, buying a property that can generate tax breaks and rents may be the best avenue for many.

    And I salute Kevin in one regard: He is right. The USA told people that buying a home is the best thing they can do; we supported with the tax code and Fannie and Freddie, and then we tell buyers they were speculating.

    Of course, anyone who ever buys a home is speculating, if they need monthly income to make the mortgage payments.

  3. Great points, Benjamin. I should probably make it more clear that I'm strictly looking at this subject from a top down perspective. Individual real estate investments are mostly dependent on local factors. I am mostly looking at this for insight into the broader economy and for potential speculation on national builders with location diversification. I don't have any insights on local factors, so I will tend to ignore them, even though they would be very important for many investments.

  4. TravisV from TheMoneyIllusion comments section here.

    Why has the 10-year U.S. treasury suddenly surged to 2.54%?……

    1. It looks like it's due mostly to the slope of the yield curve in years 2 to 4 or so. It's about time. I've been shorting Eurodollar futures in that range for a while, so I guess my question is, what took so long.... :-)

  5. TravisV from TheMoneyIllusion comments section here.

    "Homebuilders party after Lennar results"

    "Homebuilder Sentiment Soars To 9 Year High (Mortgage Apps 14-Year Low)"

    1. Thanks Travis. Interesting. I have a related post coming up in the morning. Housing starts and flow of funds data tomorrow could be enlightening.