Monday, November 30, 2015

Housing, A Series: Part 91 - Blaming "financialization" for economic stagnation gets the causation wrong.

A category of academic work has developed around the idea that "financialization" - a bloating of the finance sector - leads to economic malaise.  I think this sort of idea is built on a misinterpretation that relates to the relationship between housing and low interest rates.

The problem is that home ownership is somewhat unique as a financial security.

Equities tend to have values that are highly variable because the cash flows are highly variable and uncertain.  Bonds tend to have certain cash flows, but changing discount rates.  This causes the face market values of individual bonds with fixed rates to change over the life of the security, but it doesn't systematically change the total amount of borrowing.  If businesses were borrowing $2 trillion when interest rates were 4%, they don't suddenly double their borrowing when rates drop to 2% or halve it when rates rise to 8%.

Homes differ in that home ownership is basically a perpetuity with stable cash flows.  Especially if we consider the aggregate housing stock as a diversified basket of real estate holdings, rents on that basket of holdings will rise roughly with the rate of inflation.  And, the housing stock itself is very stable.  Only a small change in the total real amount of housing outstanding can be effected in a given year.  A perpetuity that grows with inflation is very sensitive to changes in long term real interest rates.  So both the rent cash flows and the stock itself are very stable, and the valuation is very sensitive to changing long term real interest rates.  This means that the total market value of the housing stock is uniquely sensitive to changing long term real interest rates in a way that equities and bonds are not.
Let's imagine a context where the causal factor is some sort of increase in demand for deferred cash flows, or maybe a decline in growth expectations.  These sorts of changes would lead to a falling long term real interest rate.  These causal factors themselves would lead us to expect lower growth ahead.  And the related fall in long term real interest rates would also invariably lead to rising home values.  This is unavoidable.

A measure of aggregate Price/Rent: Source
The idea that home prices are dominated by demand factors as opposed to basic universal tendencies toward intrinsic value as defined by cash flows and discount rates is betrayed by both the experiences of the late 1970s and the present situation.

In the late 1970s, nominal interest rates were very high but real rates were low.  This gives us a window into the relative power of demand vs. supply in home values.  Demand was extremely constrained because households had to qualify for mortgages at double digit interest rates.  On the other hand, low real long term interest rates meant that homes should have higher intrinsic values relative to rent.  And, we see that the supply factor dominated.  The Price/Rent ratio rose sharply in the late 1970s as nominal interest rates soared.  This should be very surprising if you believe that demand dominates.

Home Mortgages, log scale: Source
Likewise, in the current context, rates are low, but mortgage availability has been sharply curtailed, such that total mortgages outstanding have not grown in a decade, and are well below any realistic stable trend.  Yet, because long term real interest rates are very low, Price/Rent has been climbing strongly again.

During periods of slowing growth, real estate values will naturally rise, and given our conventions regarding home ownership, mortgage levels are bound to follow.  But, this is a result of the broader phenomenon, not the cause.

In our current context, this problem is even more pronounced.  The constriction of housing supply in our highly productive cities has caused the cost of housing to rise.  Instead of earning returns on invested capital that increases the available housing stock in valuable locations, urban real estate owners earn capital gains due the economic rents they gain because there are limits to invested capital in housing.  This means that those added housing costs are simply a transfer to rentiers.  The costs reflect nothing of value.  They are a drag on economic progress and equity.

In our current context, I believe that these housing supply problems are the root cause of all of these problems - stagnating economic growth, low interest rates, income inequality.  And the manifestation of these problems is multiplied in home prices, because, not only are housing costs higher as a result of the supply constraint, but the transfer of incomes to real estate owners and the drag on economic growth both cause interest rates to fall, leading to home values that rise as a proportion of the already inflated rent levels.

The core solution is to remove the obstacles to housing expansion in the Closed Access cities.  But, until that happens, economic growth will necessarily translate into higher transfers to real estate owners.  Currently, our mistaken attempt at a solution, which is to dampen credit and nominal expansion, puts a lid on real economic expansion.  Real economic expansion in this context will naturally lead to excessive debt levels and very high real estate values.  We cannot have one without the other.

This is really a conjunction of Krugman's work on path dependent geographic advantage and the work of Douglass North, who recently passed.  Cities like New York City, San Francisco and Silicon Valley, Boston, Seattle, etc. have gathered very highly productive workers into networks built on innovation and reputation in industries like communications technology, bio-tech, and finance.  These networks happened to develop at a time where some combination of factors common in Anglosphere central cities have created severe obstacles to housing expansion.  So, at the same time that these cities provide significant opportunities for highly skilled workers, they also limit access through housing, so that the normal process of migration into centers of opportunity has been closed off.  Krugman's geographic advantages meets North, Wallis, and Weingast's "Limited Access Order".

Normally, limited access order societies are built around stagnant economic systems, where rentiers capture a large portion of a small and stable level of production.  What we have created is a limited access order built around highly innovative, growth oriented, super-achievers.  Instead of having limited access to cultivated land, we have created limited access to the most vibrant forces of innovation and growth ever created.  This urban limited access order creates two classes of rentiers - the classic class of landlords, pocketing the traditional rents on property, and the new class of high skilled laborers.

We see high capital income because of these landlords.  We see high variance of incomes within labor compensation because of these new labor-rentiers, who tap into these highly productive, highly innovative networks.

The housing supply problem is key to both of these issues.  Battling against the financial sector as if it is the cause of this problem is the opposite of what we need to do to solve it.  Now we especially need to support the financial sector, either as a second best solution where it at least supports new housing stock in Open Access cities, or as part of the optimal solution, which would be to support massive housing expansion where it can boost innovation, lower costs, and ultimately reduce these returns to rentiers.  We must support finance in order to shrink it.

The distinction between income to existing capital and income to new capital is key.  Preventing the development of new capital as a reaction to high capital incomes is, at once, rhetorically gratifying and pragmatically futile.


  1. Not sure if you've seen this:

    Love the blog by the way.

    1. Thank you. It is nice to see this issue gaining support from many quarters.

  2. Certainly, "highest and best use" should become the de facto zoning law of the land, and that would lead to higher living standards and GDP.

    That said, organized militaries, religions and Wall Street are obviously parasitic.

    Wall Street makes money FROM investors, not FOR investors. As most investors cough up their money of their own free will, I see nothing wrong with this (as long as there is full disclosure, and perhaps seniors need some protection for practical reasons, like senility or extreme gullibility).

    Organized religion is fine, as long as they keep their hands off the public till. Maybe they should pay taxes like any other business.

    Organized militaries---long the bane of productive people. Keep as small as possible, down to spartan skeleton crews except in actual wars, and largely volunteer militias, no pay (not mercenaries, like today. They become a lobby group, effective too.) Most people do not know the bulk of the income and capital-gains taxes are eaten up by "national security," rural subsidies and debt.

    And, yes, build tons of condos and apartments on the Pacific Ocean! I suggest "Project 5 Million." That's how many units should be built in SoCal.

  3. The first four paragraphs of this post try to explain why bonds and equities are affected by interest rate changes differently than housing is. You are more of an economist than a financial services type, perhaps. Those four paragraphs are mostly extraneous, and including them increases the likelihood that investment management industry readers will get caught up in relative minutiae that are incorrect, (e.g. the part about face value of bonds changing) yet overlook the right-ness, er, veracity of your central findings:
    "housing supply problems are the root cause of ... stagnating economic growth, low interest rates, income inequality"
    "Instead of earning returns on invested capital that increases the available housing stock in valuable locations, urban real estate owners earn capital gains due to economic rents... This means that those added housing costs are simply a transfer to rentiers. The costs reflect nothing of value. They are a drag on economic progress and equity."
    That section, along with the charts from FRED, stands well, on its own merit. The paragraphs that follow, about "the new class of high skilled laborers" and vibrantly innovative super-achievers who physically reside in a few coastal urban centers is mostly techno-utopian speak. Cities like New York City, San Francisco, Silicon Valley and Boston have been hubs of productivity and innovation for the past 50 years or longer, in industries like communications technology and finance. In fact, one could make a case that due to the Internet and recent communications technology innovation, there is LESS (not more) need than in the past for the innovative super-achievers to occupy housing in a particular geographical housing market. I don't believe this to be correct, not yet...maybe never. Perhaps you don't either, which means that you're not such a techno-utopian after all! I suspect landlords extract economic rents on urban housing much more than city-dwelling, highly-paid workers do. The landlords have the upper hand due to being entrenched for decades (first mover advantage?) and despite having the financial means for home ownership, the workers often require unexpected and broad geographic mobility, especially when younger.

    Some cities do not have much choice in being "Closed Access," due to geographical constraints. That situation applies to both San Francisco and New York City's borough of Manhattan. San Francisco is hemmed in by mountains, although there is still plenty of housing development space for now, if it were to be allowed by policy makers in government. Manhattan is more seriously constrained, and even a transition to "Open Access" would provide limited expansion opportunities.

    Is Open Access in Manhattan more than geographical in scope? Rent control probably constitutes a form of Closed Access. Rent-controlled housing in New York City is a benefit that increasingly accrues to landlords and crony capitalists, rather than to workers and middle or lower income families, despite original intent when established in the 1940s or thereabouts. That further suppresses economic growth. It also increases inequality, while dampening rates of upward socioeconomic mobility.

    1. San Francisco looks pretty low-rise to me. New York is more seriously constrained, although there's an amazing amount of low-rise there too - and a surprising number of lots that lie empty for many, many years. Transportation could become an issue in both: mobility already barely exists in Manhattan - average speed of public transportation is often barely 2-3 mph. The evils of rent control (in both places) have been much discussed but the worse the problem gets, the more support they have.

    2. Since much of the transportation is related to commutes, higher density may actually create less demand for it, though certainly a modernized subway would be helpful. Our cities are much less dense than most foreign cities. The limits aren't logistical.

    3. Oh gosh, please forgive me, Whisk! I read your About page. I was 100% wrong in suggesting this in my prior comment:
      "You are more of an economist than a financial services type."
      In fact, you are more of a trader and money manager (i.e. a financial services type) than an economist! You were kind, and did not highlight my silly error. Thank you! I am an Arizona resident at present. My brother attended University of Arizona, just as you did, for a master's degree, and undergrad too.

      Getting back on topic, you observed that our cities are much less dense than most foreign cities. This is true, yet the relationship between high density cities and more affordable housing is non-trivial. Compare the cost of average residential real-estate per square foot (meter?) in NYC or San Francisco with the same measure in large western European and coastal Asian cities. US cities are much more affordable, sometimes by a factor of three. Shouldn't higher-density foreign cities have lower costs for housing than lower-density American cities?

    4. No problem. I tend to wear an economists hat a lot, though I don't have the credentials to justify it. Interesting question on international differences. I don't know the answer, although the Anglosphere seems to follow a common pattern - London, Sydney, Vancouver, NYC, SF, etc. I see a lot of the same complaints. I don't know much about housing globally, though.

  4. I'm sorry for the lengthy comments. I wanted to add one more thing.

    I think you need to make a distinction between financialization and banking. Increasing financialization of the Wall Street variety is quite different than residential mortgage banking expansion. Your concluding lines, which are the tie-in with the post title, are
    "Battling against the financial sector as if it is the cause of this problem is the opposite of what we need to do to solve it. Now we especially need to support the financial sector..."

    Yes, we DO need to support retail banking, insofar as it provides funds for home mortgages! The other part of the financial sector, specifically merchant bankers, do the following: facilitate mergers and acquisitions, lead complex (non-mortgage) securitization deals and manage assets for pension funds, among other things. These activities don't have a direct beneficial impact on the stock of housing, unlike the activities of non-investment banks, whether small community banks or huge banking institutions such as Chase, Bank of America and Wells Fargo.

    1. Thanks for your input Ellie. A lot to think about here.

      1) What is incorrect in my early paragraphs?
      2) Revealed preferences suggest that it is extremely valuable for certain tech firms to be geographically clustered.
      3) Manhattan had more residents in 1900, before the expansion of the modern skyscraper, than in 2000, and millions of people come in every day, but have to sleep in New Jersey, etc. These cities aren't very dense, residentially. They aren't even close to hitting geographical limits.
      4) The statistical expansion of finance is strongly related to the expansion of mortgages outstanding. I think this premise is accepted by those who blame financial expansion for economic contraction.

    2. Oh, and thanks for the input on writing. I tend to write the blog without thinking too much about audience. I usually find some obscure technicality to think about which leads me to a conclusion about the broader issue, and I just spill it out as a stream of consciousness. I am actually kind of surprised and grateful that I have any readers, and especially that I have a number of readers who have muscled through the entire housing series enough that they find my narrative compelling. I don't know if I would have had the patience as a reader to do it.

      I hope to get this into a medium for a wider audience, and that really will be a process mostly of going back and editing all of these thoughts into a manageable and readable story. So, I appreciate your feedback about what seems to work and what doesn't.

    3. Hello Idiosyncratic Whisk! I am trying to remember how I found my way here. You might have been a click removed from Arnold Kling's blog or possibly Trotsky's Children (that one isn't about communism, rest assured ;o)

      Regarding your first point, in response to my comment, the following passage is what struck me as incorrect:
      "Bonds tend to have certain cash flows, but changing discount rates. This causes the face values of individual bonds with fixed rates to change over the life of the security, but it doesn't systematically change the total amount of borrowing."

      The first sentence of the quoted excerpt is correct. I don't understand the second sentence, stating that varying discount rates causes the face values of bonds with fixed rates to change over time. Bonds have a face value, which does not vary. Face value is not the same as market value. Current market value, i.e. price to purchase, is what varies over the life of a fixed income security.

      For example, the semi-annual coupon payments on a fixed-rate corporate bond are based on a static par value, usually $1000 per bond. Even zero-coupon bonds, which are issued at a discount to face value, do not have varying face values, but rather, varying prices at any point prior to maturity.

      What kind of a bond are you thinking of, whose actual face value changes? Does its coupon payment change too?

      Next, I'll jump to your fourth bullet point, "the statistical expansion of finance is strongly related to the expansion of mortgages outstanding". I am curious what you mean by the "statistical expansion of finance". Are you referring to growth in financial services as a percentage of GDP, the financialization trend that your article begins with? If so, that makes sense. If not, please explain?

    4. Geez. You're right. That's a typo. I meant market value. Thanks for catching that.

      On point 4, yes. I'm being a little sloppy and vague. But, yes, generally the articles that blame financial expansion for recessions or stagnation seem to base it on total bank assets as a proportion of GDP, or finance value added, but these things seem to generally move in the same directions. "Finance" has a sort of slippery definition, in any case. I would tend to generally view debt and equity as different forms of claims of ownership over productive or durable assets. When that ownership is deemed to be facilitated or captured by the finance sector seems to be somewhat arbitrary to me.

      When real long term interest rates fall, safe assets like homes gain nominal value. Since a large portion of this type of asset is deemed to be facilitated by finance, we are tempted to see "finance" as the central mover, but really it seems like a secondary factor to me. If instead of mortgages, our homeownership system was based on, say, diversified institutions that sold call options on properties to their tenants, they would all just be considered "owners", and there would be no association with financial expansion, even though the basic ownership process would be similar. If semantics can make that big a difference, I think it is dangerous to draw sharp conclusions about the size of a sector.

  5. A common problem, too, is that as real estate values rise, so do property taxes - and this reduces the state's interest in adding more to the capital base, which would naturally be the force that would help to offset the rentiers' lobbying against new construction. Not as true in California, of course, because Prop 13 requires waiting until there's a change of ownership, but the trend is the same.

    1. I don't think this is key. As you point out, California is the worst place, and they have limited this effect. Also, oddly, it is not the rentiers who seem to be the obstacle. It seems to be the populists.

    2. Populists?
      Maybe so. Also "existing property and homeowners."

      Can you name a single-family detached housing district that

  6. It's interesting to compare real estate to New York taxi medallions, and consider that relaxing development restrictions could have a similar effect to the advent of Uber. Or, it might not: New York already had black cars, yet taxi medallions were worth $1 million and more.

  7. This is one of the most wonderful blog, this is work is tremendous. Thanks
    Opt In Leads

  8. Kevin, this is up your alley:

    1. Thanks. Interesting article. I suspect that macroprudential policies make things worse, in part, because they simply add another obstacle between savings and investment. If home prices are high because real interest rates are already low because of a surfeit of investment, then that will just make matters worse.
      It's strange to me how the presumption of efficiency that finance academics would normally apply to capital markets doesn't remain in housing. Yet the evidence, such as what your link discusses, seems to support that presumption in housing.

  9. This is a really interesting topic that I have never really thought of. Kudos to you for bringing it up and thank you for the link to the paper. It is a very intuitive concept once you make sense of it.

    I do feel that i disagree with some of your contentions and solutions to the issue. I believe that interest rates do have a significant effect on the value of both stocks and real estate over long periods of time. I would contend that low nominal interest rates are the primary measures of investment flows and therefore determine value. For example, in the current environment, interest rates remain low which creates crowding and forces investors into risk assets, creating or leading to higher equity valuations. A similar phenomenon imo is evident in real estate. Crowding creates more volatility which again is evident in both asset classes. This valuation crowding is related to crowding within communications technology, bio-tech, and finance in terms of both workforce and overall investment. The question is whether this crowding is justified. Even if this crowding is justified, where do you find more money to pump these industries? Either you must find other parties to invest (and given that the commodity slowdown is taking money away from savers/governments in EM, the market for US securities is shrinking), introduce more leverage (in an economy that is already over-leveraged), or manufacture it (print it or whatever, which is essentially a method of wealth redistribution which only adds to the problem of inequality). And even if you succeed in doing so, price/rent will continue to rise because as long as more workers and money moves into these sectors, there will be someone to profit off economic rent. Look at some of the gold bubbles or oil bubbles in US history. Rents in the Dakotas now are extremely steep around the wells. If you're making a lot of money, someone will find out and you can bet they'll find a way to do the same or rip you off. Its part of capitalism and momentum. Some areas of finance are just as useless as what a landlord does and I guarantee you there are plenty of hoaxers inside biotech (Martin Shkreli for example) that do nothing but rip people off. Why continue pumping industries that have too much investment already? In the long run, we are all dead anyway I guess.

    1. There are useless and useful people everywhere. I think our perceptions of reality, especially when it comes to broad aggregates, get clouded by dwelling on these characterizations.

      On the other asset classes, I haven't gone as deeply as I have with housing, but my preliminary findings have been that implied real yields on equities are surprisingly stable, and corporate leverage does not rise with falling interest rates - it's the opposite, if anything. Here's a series I did on the topic. On leverage, you need to be careful. For instance, the expanding global footprint of US firms will make debt levels look high if you compare them to US GDP.

  10. Kevin, have you seen this paper? It seems highly relevant to the topic of skyrocketing rents in the richest urban cores of the country:

    1. Thanks for the link. I had seen a summary of it, but I hadn't looked at the complete paper. There is a lot to chew on there. It really does seem like the grand story of the past 40 years has to do with the interrelationships between technology, globalization, tax rates on wage income, gender rights, and urban housing constraints. That paper plugs into a lot of those factors. Great stuff.