Housing: Part 266 - Tax subsidies as annuities

Let's imagine a home purchase in an environment with no tax distortions.  Rental income taxed after expenses.  No mortgage interest deduction.  No capital gains exemption.

A house that rents for $20,000 (annually) might sell for $200,000.

Now, we add in these tax benefits that are only captured by owner-occupiers.  No tax paid on rental income.  Mortgage interest deducted when taxes filed.  No capital gains on most sales.  Now, the house might sell for $300,000.  (I am using round numbers for ease, so this might be a slight exaggeration of reality, but it probably isn't too far from the actual premium.  White House estimates put the value of these subsidies at about 25% of annual operating profits for owner operators, and those subsidies are focused at the top end of the market.)

I think it helps to look at property taxes as a silent partnership.  If property taxes amount to 20% of net rental income (either cash or imputed), then the government has become a 20% silent owner of the property.  This is not that different than having an adjustable rate, negative amortizing mortgage with an LTV (loan to value) of 20% with no prepayment option.

Likewise, we can think of tax subsidies as a sort of annuity.  When you buy the house, you are basically buying the future cash flows of a house with $20,000 annual rent.  But, tax subsidies mean that when you buy the house, you also are required to purchase a $100,000 annuity.  That annuity amounts to the forgiven taxes on the net tax income as they are earned, the tax deduction on mortgage interest when taxes are filed, and the forgiven capital gains tax from when the home is sold in the future.

Since the value of tax subsidies are capitalized into the price of the house, then it is best seen as a legal requirement that when you buy a home, you also must purchase an annuity.  The payment for the annuity goes to the previous homeowner, and the annuity income comes from the government, in the form of forgiven taxes.

So, the question about the effect of tax subsidies on homeowners really comes down to the price of the annuity compared to the value of the annuity.  If the price of the annuity is more than the value, then home buyers will decrease their investment.  If the price of the annuity is less than the value, then home buyers will increase their investment.  Of course, the value of the annuity is complicated, itself, since it is largely determined by discount rates on those future tax savings.  These are difficult, if not impossible, to quantify.

But, we can make some generalizations.  And, instead of thinking about how this affects "the housing market", there is probably a lot more going on in the differences between sub-markets.  Low priced homes or homes owned by households with low incomes would not have very valuable annuities, because those owners would not capture many tax benefits.  But, in neighborhoods that would be dominated by those owners, prices are much lower (in price/rent terms).  The prices of the annuities are low, too.  The distortions in those markets are probably not great.

I think one of the things that happened in the Closed Access cities during the boom was that home prices in low-tier neighborhoods were rising high enough to make those annuities valuable, and the in-migration of high income buyers to the Closed Access cities also might have made the marginal new buyer also more sensitive to the tax advantages.  So, in Closed Access cities, during the boom, low tier Price/Rent levels rose up to nearly the same level as high tier Price/Rent levels.

At the other end of the spectrum, in the highest tier neighborhoods, households would value the untaxed rental income, but those neighborhoods tend to be less leveraged.  Another mitigating factor here is that at the top tier, rent tends to take a lower percentage of household income.  In order to capture the annuity, the household must purchase a more valuable home.  They may simply not demand more shelter, so even if the annuity has value, it may not have enough value to induce them to consume more housing than they already care to consume.

So, as in the low tier, the annuity may not have much of an effect on owner-occupier demand.  This may be why Price/Rent ratios level out in zip codes above $400,000 or so (plus or minus, depending on local conditions).

In the middle, households would value untaxed rental income and capital gains exemptions.  Here, there would be a lot of variation.  In a single neighborhood with a single price point, there would be less leveraged older families who would value that benefit less, and younger, leveraged families who would value it more.  So, if there is a section of the market that might value the annuity the most, it is young households with above-average incomes.  They would react by increasing their demand for housing because housing would be a complementary good with the annuity that legally comes attached to it.

But, these families are probably among the most credit constrained households.  In other words, they may be in a position to gain the value of ownership (in and of itself, before any value from the annuity), but their limit to gaining that value may be limited by their access to mortgage credit.  So, their demand for housing may be very elastic at the limit of mortgage payments they can qualify for.

This means that the households who might be most likely to increase their housing consumption because of the tax subsidies may be least able to pay for the annuity.

Consider: without the subsidies, those households would be provided with the opportunity to purchase $20,000 in rental value for $200,000.  With the subsidies, those households would be provide with the opportunity to purchase $20,000 in rental value for $300,000.  What if their credit constraints mean that they are basically willing and able to purchase a house with a price of $200,000.  Mortgage constraints are based on price, not rental value.

That would mean that, without the subsidies, they would purchase a home with $20,000 in rental value, but with the subsidies, they would only be able to purchase a home with $13,333 in rental value.  For households with high incomes but with credit constraints, the subsidies might reduce their housing consumption, and it is those households who would most value the subsidies.  This seems like it would especially be a factor in Closed Access cities, where higher rents mean that potential homeowners have both higher incomes and more credit constraints.

Where new homes are built, the important margin is the comparison between price and the cost of a new home.  The payment for the annuity goes to the homebuilder, and the builder doesn't care if you're paying for the home itself or for the legal annuity attached to it.  It's all profit to them.  So, we should expect new residential investment to decline if subsidies are removed.  Price/Rent of existing homes would decline.

But, the important margin for housing consumption is nominal rent expense as a proportion of incomes.  Households would be willing to spend a similar amount of income on rent, but with less supply coming online, the normal increase in demand (expressed as total rent paid) would continue to rise.  There would be rent inflation, which would cause home prices to increase somewhat, mitigating the decline in value because of the withdrawal of subsidies.  In other words, over time, households would tend to live in homes with lower cost of construction (smaller, etc.) but with no change in rental value.  This would mainly happen in the mid- to upper-tier markets where the subsidies had been the most valuable.  (This would not be the case in the Closed Access cities, because cost of structures is not particularly relevant to home values there.  In Closed Access cities, values would fall and rents would be generally unaffected by this effect.)

Now, we need to think again about those credit constrained buyers.  Without the subsidies, their demand for housing (in terms of rental value) might actually rise.  This would put more upward pressure on rents, and probably upward pressure on residential investment and housing starts.

There are many studies that seem to suggest that tax credits get capitalized into home prices, so that they make homes more expensive, but they don't lead to an increase in homeownership.  That is not surprising to me.

In the end, under certain circumstances, I wonder if eliminating homeowner tax benefits might simultaneously be associated with strong residential investment while at the same time creating some negative wealth shocks among the households who would lose the annuities which they had paid for when they had purchased their homes under the tax-preferred regime.  These effects would probably not be significant in low tier markets, where the price and value of the annuities is low.  It would be significant in the high tier markets.

This seems counterintuitive, because we generally think of tax subsidies as a handout.  But, if tax subsidies are capitalized into home values, which they clearly are to some extent, then they could be conceived of as a regulatory burden.  New home buyers are required to purchase an annuity, frequently entirely with borrowed money, which would rarely be considered a wise investment strategy.  And, their buying decision is partly based on the effective value of that mandated annuity to them.

Housing: Part 265 - Canadian edition

I am keeping an eye on Canada.  The central bank has begun to raise rates.  Home prices, especially in Vancouver and Toronto, have reached very high levels, and concern about high prices seems to be affecting public opinion about monetary policy, as it has in the US.

They seem to be in danger of doing the same thing we did in 2007.  Prices peaked and have fallen back, especially in Toronto.  They seem to be in a sort of 2006 place right now.  There is a sense that "the bubble is over", but general growth still seems reasonable.  They could manage it well.  But, declining home prices are probably a sign of disequilibrium.  If they treat it as a return to normalcy, they will underestimate the contractionary level of their policy.  It will be interesting to see how things proceed.

Here is a post at seekingalpha that reflects the broader problem.  I don't mean to single this author out.  This is the typical reaction I see.

The thing about economic expansions is that people are supposed to work, pay down debt and build up savings, so that they can ride out the economic downturn that follows. Canadians have done the opposite over the past 8 years, and now approach 2018 with the highest debt and lowest cash savings in decades.

This is understandable.  It seems like excessive demand is the problem - excess lending, speculations, foreign investors, etc.  So, it looks like Canadians (just like Americans, Australians, Brits, etc., etc.) are profligate and short-sighted.  And, this subtly works its way into the public consensus about what needs to be done, how much growth or contraction we will demand, how much sympathy we should have for bankers and homeowners, etc.

The problem is the story is backwards.  In a Closed Access economy, growth naturally leads to rising debt levels, because at the center of the economy is regulated access to opportunity.  It is inevitable - unavoidable.  A bidding war ensues for residential access to lucrative labor markets.

That debt isn't from profligacy.  The debt is a transfer of wealth.  The debt is the result of economic rents that must be paid to the owners of the restricted asset.  So, inevitably, when the economy grows, earners have to take a haircut.  They have to pay off the landholders.  This creates debt and economic stress.

I am not as well versed on the empirical evidence in Canada, so I can't say that I am certain about the story there, but it seems to have all the relevant inputs.  In 2006 and 2007, America became infected with a sense of vengeance, or at least a lack of sympathy.  This is probably the core question about the Canadian economy in the months ahead.  Do they approach housing contractions with sympathy, with resignation, or with bitterness?  Sympathy might save the day, even with the wrong model of what is happening.

Upside-down CAPM, Part 1: Why interest rates are a poor indicator of monetary policy

I have previously written about an Upside-down CAPM, meaning real total returns on corporate assets are generally quite stable (with some noise) at around 7%.  Those returns are divided between equity and debt owners.  Total real returns of the total stock market can be considered the sum of the long term real risk free rate (estimated by long term TIPS bonds) and the equity risk premium (ERP).  Total real returns of an individual firm are shared between its equity holders (with expected returns equal to some multiple of ERP, depending on the firm's "Beta", or sensitivity to market volatility) and its creditors (with yields equal to risk-free returns plus a credit spread).

Thinking about market returns in this way helps to see how thinking about monetary policy in terms of interest rate targets is not helpful.  Thinking about monetary policy through interest rates seems to lead to the idea that low rates induce leveraged investing, and that this is how monetary policy affects spending and economic growth.  But, interest rates don't systematically affect corporate leverage, at least in terms of inducing speculative cyclical investments.

Source

Here, the red lines are the Fed Funds rate and the 10 year treasury yield.  Generally, when the Fed Funds rate is well below the 10 year yield, this is when monetary policy is generally viewed as being loose or accommodative.  The dark blue line is nonfinancial corporate leverage as a proportion of enterprise value (debt + equity), based on historic cost.  The light blue line is leverage based on the market value of equity, instead of historic cost.

We can see that most of the cyclical shift in leverage is based on collapsing equity values during a contraction, which then recover.  Based on book value, there is usually little cyclical shift.  And, leverage has been declining as interest rates have fallen, both in real and nominal terms, for several decades.  And, when the Fed Funds Rate is low relative to the 10 year yield, there is no noticeable shift in leverage.

The Modigliani-Miller thesis says that debt financing is advantageous for firms because profits are taxed at the firm level, but interest expense is not.  I don't see any reason to doubt that this is true, to an extent.  The counterintuitive result of this is that higher interest rates provide a tax advantage, so that higher rates actually could lead to higher leverage.  In fact, if the total required returns on equities are stable (which I assert that they are), which means that a 1% increase in the risk free long term interest rate will generally be matched by a 1% decrease in ERP, then declining interest rates should lead to declining leverage and declining firm share value!

Note, the Modigliani-Miller effect is nominal, not real, so that it would be related to high leverage in the 1970s and 1980s because of the high inflation rate.  Mostly, this seems to flow through lower equity values, because high inflation increases the de facto tax rate on firm profits.

Interest rate induced cyclical corporate leverage simply isn't a thing.  For as much bandwidth gets used up talking about it, you'd think there would be something there.  There isn't.

This makes sense, if we view markets through the upside-down CAPM model.  First, firms don't borrow based on the overnight rate.  In fact, the iShares Core U.S. Aggregate Bond ETF (AGG) currently has an average effective maturity of nearly 8 years.  This is a mixture of various forms of debt, but the point is that long term yields really are a more important factor in aggregate borrowing costs than short term yields.  And, certainly, a firm will try to match the duration of its borrowing with the duration of its investments, to a certain extent.

Total operating profits to firms don't change because of changing interest rates.  Changing interest rates change how operating profits are shared between equity holders and creditors.  Real interest rates mostly reflect a discount taken by creditors compared to the return claimed by equity holders, because creditors avoid cash flow and market price volatility.

The idea that interest rates would effect corporate leverage is especially suspect when we think about the standard way in which any CFO would manage a firm's balance sheet.  Firms don't leverage up specifically based on each individual project.  They don't say, "Oh, we can now borrow at 3% and here is a project that has come across my desk that returns 5%, so let's borrow cash to make this project happen."  Firms generally use the weighted average cost of capital ("WACC") and they compare the full range of projects to that WACC.  The leverage that the firm targets will be based on a number of preliminary factors of risk and its effect on credit spreads.  In most cases, WACC is largely a product of the returns to equity.  If a firm with any substantial amount of risk was leveraged enough for the cost of debt to be dominant, they would have a high credit spread, which would overwhelm any effects of lower risk free interest rates.

The riskiest firms and the firms with the most pro-cyclical risk tend to be the firms with the highest credit spreads.  They tend to be more equity financed, so that in those cases WACC will especially be immune to changing short term interest rates.  It will mostly be a function of the cost of issuing equity.  In that case, the cost of capital will be related to short term interest rates, but contrary to how people seem to normally think about it.  Let's say you have a highly cyclically risky firm, with a Beta of 2 - twice as volatile as an average firm.  Then, if investors are feeling safe, ERP might be 3% while real long term rates are 4%.  So, the risky firm has a real cost of equity that is 10% (4% + 3% x 2).  But, if investors are risk averse or afraid of cyclical volatility, ERP is 5% while real long term rates are 2%.  And the risky firm's cost of equity would be 12% (2% + 5% x 2).  Risk aversion increases the cost of capital.  The Fed moving around overnight borrowing markets just isn't going to do much to change those long term borrowing costs.  But, if cash it injects into the economy improves NGDP growth expectations, then ERP will decline, long term real interest rates will rise, and WACC for cyclically sensitive firms will decline.

In fact, if we think about it this way, to the extent that monetary policy affects the cost of capital, it would happen mostly as a result of NGDP growth expectations.  NGDP growth expectations would increase expected corporate profits, increasing share prices, decreasing WACC.  But, this would happen through equity-financed investment, not debt.  And, in fact, that is what we see in the graph above.  During recoveries, market-based leverage declines because the value of equity rises.

We can imagine this in an extreme example where a firm in distress has financial leverage above their optimal target.  If equity becomes so cheap that the firm's enterprise value is dominated by its debt, then WACC would be determined mostly by the firm's interest rates.  In those cases, the firm will be credit constrained, and marginal investments will be limited to generated cash.  So, in that case, lower interest rates will be unlikely to generate leveraged investments, but accommodative monetary policy might lead to a stronger economic recovery in general, which would boost revenues and profits, and for a firm like that, those improvements could lead to a sharp recovery in equity value.  Enough equity recovery might eventually allow them to draw on credit markets again.  But, the recovery plays out in the value of the equity.

Loose monetary policy is frequently blamed for leading to a build up of risky borrowing.  But, cyclically accommodative monetary policy actually leads to systemic stability as equity financing grows.  If monetary policy is loose in a secular sense - over the entire business cycle - so that it does lead to increased inflation, like in the 1970s, then leverage tends to rise because the de facto rate of tax on corporate earnings is higher, and interest expenses bloated by inflation serve to defer taxes.  So, we might say that loose monetary policy does lead to destabilizing leverage, but this happens through high interest rates, not low interest rates.

The neutral overnight rate was probably already less than zero in September 2008.

I have written about how interest on reserves was a contractionary policy implemented in October 2008.  It seems plausible to me that much of the damage to nominal economic activity happened in November and late December 2008 when the Fed had the interest rate on reserves set at 1% along with the Fed Funds target rate, but that the effective Fed Funds rate was less than half that.  This induced banks to deposit hundreds of billions of dollars in excess reserves at the Fed.

Source

During the period between the disastrous September FOMC meeting where the Fed Funds rate was maintained at 2% and the initiation of interest on reserves, banks had already deposited about $150 billion at the Fed in excess reserves.  I had interpreted this as a sort of mitigating fact regarding the contractionary nature of the policy, since this meant that some excess reserves would have accumulated even without interest on reserves, so that the entire $800 billion in reserves accumulated before the Fed finally dropped the target rate to near zero couldn't necessarily be blamed on interest on reserves.

But, I don't think I pointed out the other implication of that early buildup of reserves.  The other implication of those pre-IOR reserves is that already by September 2008, the neutral Fed Funds Rate was below zero, because banks were already willing to park a tremendous amount of reserves at the Fed.  To see the scale of this, note that the measure shown above (in black) isn't just of excess reserves.  It's of all reserves.

It is true, I think, that the gross positions of sales and purchases of overnight Fed Funds are larger than the net reserve position, but that being said, this was a tremendous amount of reserves banks were willing to set aside.

I don't really know anything about the microstructure of the Fed Funds market, so I don't know how the Fed maintained a Fed Funds rate above zero.  Maybe those were reserves lent to banks that other banks considered to be credit risks?  Maybe the defensive draw down of credit lines that happened in late September (blue line) was causing some banks to run short of reserves while others were flush.  Once that draw ended, the Fed Funds rate does appear to have fallen to the level of interest on reserves.  Help me in the comments if you can.

The stated reason for implementing interest on reserves is that it would help the Fed keep control of interest rates, so they could maintain their target rate without having to sell all of their treasury bills.  Not only is that stated reason strange, the policy seems to have failed to achieve even that strange goal.

My main point here, though, is that this is evidence that even while the Fed Funds rate was at 2%, then 1.5%, and then 1% for three months, the neutral rate was surely below zero for the entire period of time, and would have been, with or without interest on reserves.  That neutral rate was surely partially due to the Fed's policy target, so maybe if the Fed had pushed the target rate to 0.25% in September, the neutral rate would have been high enough to require the Fed to inject cash through treasury purchases.  That is, strangely, what they were trying to avoid.

The Upside Down CAPM and Lawrence Summers

I think there is a lot of value in turning the CAPM upside down.

Normally, we think of it as:  ER = R(f) + B*ERP

The expected return of the broader stock market is the risk free rate of return plus the equity risk premium - the premium investors demand for having exposure to cash flow that is volatile over the business cycle.  The expected return on a given equity is its "beta" - its sensitivity to broader market volatility.  The broader market, by definition, has a beta of 1.

I prefer to think of equity market expected returns this way:  R(f) = ER - ERP

It's a simple algebraic reformulation, so it seems like it can't make that much difference.  But, the key is that (1) expected returns (ER) seem to be pretty noisy (or unknowable) around a stable mean, so that ER can generally treated as a sort of mysterious constant and (2) the real risk free rate is generally negatively correlated with expected real growth and does not have a stable mean.

When secular growth expectations decline, R(f) tends to decline and ERP tends to rise.  Expected returns aren't a product of R(f).  It's more accurate to say that ERP is a product of R(f), so that, lower growth expectations cause equity holders to demand higher returns relative to R(f).  Comparing equity yields to real bond yields gives us little information, and comparing equity yields to nominal bond yields is less than useless.  (We are currently in an odd time, because what amounts to financial repression in mortgage markets has pushed real long term interest rates lower than what we might call broadly the "natural" rate.)

I was recently listening to Lawrence Summers on David Beckworth's great podcast series.  Here, I think Summers has presented an example where this framing would help.  He is talking about whether it would be beneficial now to have a US sovereign wealth fund investing in equities:

If it were all about risk premiums then you'd think this was a particularly attractive moment to invest in stocks relative to bonds because there was an extraordinarily high risk premium . I think most sophisticated people in markets kind of think the opposite. Some people think the stock market's a bubble, some people think it's not. I don't know many people who think it's unusually cheap, which would be the corollary of the view that the risk premium is extremely high. So I don't particularly relate to the safe asset paradigm. If you had that paradigm, then you would say that if the spread between bond yield and stock yields was super wide. This was a good moment for the government to issue bonds and buy stocks. But I don't find that a plausible guide to my own investment behavior. And so I'd be hesitant to inflict it on the government.

Now, in ERP terms, clearly there is a high equity risk premium.  Clearly, expected returns on equity are much higher than risk free returns.  Summers is speaking about risk premiums, but he seems to be thinking in terms of absolute valuations.

If equities perform poorly, though, it will be the result of unforeseen real income shocks.  It will have little to do with valuations, because valuations, in terms of total expected returns, don't actually change that much after taking account of cyclical changes in growth expectations.  Thinking of equity valuations as a fairly stable long term variable with a lot of short term and mid-term variability, then it is much less confusing.  Clearly the equity risk premium is high, and we don't need to know much more than that real long term interest rates are low to realize that.  After cyclical adjustments (which are large for equities, make no mistake), it will take some sort of epic regime shift in the American economy to keep equities from vastly outperforming real bonds over the next 30 years or so.

--------------------------

Later in the interview, a couple of other items caught my attention.  On QE:

I'd also say, whatever you thought a few years ago, fact that we stopped QE and interest rates kept falling, it has to lead you to think QE is less effective than whatever you thought before we did that experiment.

This comes from thinking of monetary policy through interest rates.  Clearly, to me, the fact that interest rates fell after the QEs were ended is a sign of how QE was effective and had kept rates from falling while QE was on.

But, then, he follows up with this comment about raising rates moving forward:

And I don't see the financial stability rationale either. I'm not sure that markets are extraordinarily overvalued. If I believed with confidence that markets were a bubble, I'm not sure that would be a reason to tighten policy. It might be a reason to ease policy. 'Cause when the bubble bursts, we're gonna have a real problem. And so, I might as well get some stimulus into... Behind the economy. So, I don't see the case for tightening.

Can I nominate Summers for Fed chair in 2006?

Housing: Part 264 - Rent is how we consume housing services!

Chris Dillow at Stumbling and Mumbling has a post up about housing where he claims that added supply in London won't make housing much more affordable.

This post is an excellent example of how confusion between owning and consuming is such an important source of error in housing markets.  To be fair, his treatment of the issue is conventional here.  As he writes: "Changes to the flow of the asset are generally too small to have much effect. For this reason, many economists have traditionally modeled house prices as if only demand matters".

This is absolutely, sadly, true.  As with his post, pick up any random academic article about the causes of the housing bubble and it is likely that "rent" is not mentioned in the paper at all, even as a factor to be dismissed.  It is widely treated as a factor not relevant enough to even bother dismissing it.  Not only is housing treated as if only demand matters, but it is treated as if this is demand for owning, rather than demand for consuming.

I wonder if the single most useful policy we could enact regarding housing markets would be to force all homeowners to actually write themselves a check each month for the market rental value of their homes.  I think the fact that we don't do this, so that many homeowners equate their cash expenses with the cost of housing, is a significant reason why this error is so widely committed.

The cost of housing is rent.  Period.  The crazy thing is that about half of the residents of the large urban centers are renters!  For them, this is not complicated.

Dillow writes:

If supply doesn’t affect prices, what does? Lots of things: demographics, incomes, debt levels, expected incomes and the availability of credit. My chart shows another influence: real interest rates.

Now, in order for interest rates to affect home prices, rental value must be an input, as Dillow's chart implies.  But, by removing rent as an explicit input, notice how much easier it becomes to dismiss supply as a factor in price.  Supply won't affect any of the remaining factors Dillow cites.

Imagine if we conceived of other markets the way we conceive of housing.

Families are finding it increasingly difficult to feed their families.  But, the proposed solution - to harvest more grain - is a non-starter.  There is only so much land, and adding to the available acreage is a slow process that can't begin to keep up with the problem.  And, acreage continues to be very expensive.

Acreage is expensive mostly because of low interest rates, loose lending to farmers, and high farmer incomes.  So, the question is, how do we reduce the value of acreage to make food more affordable?  Maybe we could pass laws so that farmers must allow gleaning.  Maybe we can put more obstacles up so that banks aren't so eager to lend to local farmers.  Maybe we can regulate the harvest so that farmers only sow the grains that we have determined are most in need.

PS. It could be that Dillow's claim about supply not bringing down prices is somewhat accurate in practice, which I discuss here.  It depends on whether high prices in constrained cities are more sensitive to interest rates or to expectations of rising rents.  If constrained supply makes prices more sensitive to low interest rates, then the rate of new supply in the worst cities would, indeed, need to double, triple or more in order to stop the flow of low-income out-migration and to introduce enough supply to create a regime shift in rent expectations so that they are actually expected to decline.  I have been moving toward this point of view.

Housing: Part 263 - The correct answer is not in the set of choices.

Actual footage of economists answering IGM Questions

The IGM forum polled  economists about Factors Contributing to the 2008 Global Financial Crisis.

In order of importance, the choices are:

1. Flawed financial sector
2. Underestimated risks
3. Mortgages: Fraud and bad incentives
4. Funding runs
5. Rating agency failures
6. Housing price beliefs
7. Household debt levels
8. Too-big-to-fail beliefs
9. Government subsidies: Mortgages, home owning
10. Savings and investment imbalances
11. Loose monetary policy
12. Fair-value accounting

Some of these are actual problems, or at least policies which are arguably not optimal and feed instability.  Numbers 1, 2, 4, 5, 8, and 12 are basically issues that will always be up for debate and that will always appear, in hindsight, to be the cause of financial crises.  When there is a run up in defaults or a run on assets at financial intermediaries, for any reason, in hindsight, it will appear wise to suggest that financial intermediaries took too much risk, held too few assets, or underestimated risk, and balance sheet accounting that forces pro-cyclical buying and selling will make those things worse.  All of these things could possibly be addressed to help avoid future instability, and they certainly count as contributing factors, as posed in the question.  But, they don't really help much in terms of finding foundational causes of a financial crisis that occurs after national real estate markets have lost a quarter or more of their nominal value.  In the respondents' defense, the question was about "Factors Contributing to the 2008 Global Financial Crisis", so these answers aren't wrong regarding the question at hand.

Numbers 3, 7, and 10 are outcomes of the bloated housing market.  These outcomes are the result of bloated asset values, regardless of the cause of the bloat.  Again, as "factors contributing" to the crisis, they aren't necessarily wrong.

Number 9 legitimately contributes to the capitalization of real estate asset values, and thus higher prices, and also contributed to price volatility, though in a way that I don't think is widely appreciated.

The problem is in the list of questions.  In all of these cases, there is a presumption that the cause of both the boom and bust was the financial sector.  And, there is a presumption that the bust was inevitable.  Some of the respondents note that some of these factors are mechanisms or effects more than causes.  Of 74 respondents, only one noted that the list was tilted toward issues in the financial sector, but even that respondent, Nicholas Bloom of Stanford, makes clear that he considers home prices to have been too high, presumably making the bust inevitable.

Numbers 6 and 11 reflect the confusion that comes from the presumptions behind the rest of the list.  Loose monetary policy is given low importance, which is heartening.  What is interesting about that is that tight monetary policy isn't even presented as a potential cause, and none of the respondents seem to have a problem with that.

That leaves Housing Price Beliefs.  This one little factor in the heart of the list.  The rest of the factors really hinge on this factor.  Surely, limiting ourselves to this list of potential factors, none of the other factors would have been notable if we hadn't started with the core problem of overpriced housing that was bound to collapse.

Here, I present a series of scatterplots.  Each point represents the median home price of a given metropolitan area.  The x-axis is the price at the beginning of the time period.  The y-axis is the change in price over the given time period.

The graph these economists appear to have in mind is the first graph, of home prices from 2006 to 2011.  During the specific period of the crisis, there was a clear negative correlation between home prices and subsequent changes in home prices.

This period of time does appear to suggest that beliefs about home prices were optimistic.  It was also characterized early by an inverted yield curve, which is a broadly accepted sign of tight monetary policy, followed by a collapse in investment and GDP growth, a rise in unemployment, and a persistently low rate of inflation - also all signals of tight monetary policy.  It was also characterized by a near nationalization of the mortgage market and probably the single most extreme publicly imposed tightening of credit standards in US history.  Also, not available as an option on the poll.

Regarding those home price expectations and credit standards, we might take a step back and look at the broader picture.  What if we step all the way back to 1998 and all the way forward to today?  I think that it can be stated uncontroversially that, regardless of how one regards credit standards in 2006, in the time since then standards have remained tight - tighter than in any recent period of time.  After all, the private securitization market that funded subprime and Alt-A loans is basically non-existent.  So, I hope we can agree that home prices in 2017 are not part of a credit-fueled housing bubble.  Surely, collapsing rates of homeownership and a decade-long period of depression level residential investment are confirmations of this fact, if confirmation is required.

So, how about price expectations of home buyers back in 1998?  How did buyer expectations pan out?  In this case, there is a strong positive correlation between starting price and subsequent price growth.  Keep in mind, this period of time includes the period shown above.  This period of time concludes with a decade of public policy explicitly aimed at limiting access to mortgage credit where real estate is expensive, keeping home prices low.  Yet, for each 10% premium homebuyers paid in 1998, they could expect to earn about 4% in additional profit over the next 19 years.

This isn't just a refutation of the idea that price expectations were too optimistic.  In high priced markets in 1998, home price expectations were extremely pessimistic.

Now, let's move up in time, to the peak of the boom at the end of 2006.  Here is the scatterplot of home prices and subsequent changes from then until today.  Remember, again, during this period of time there was an extreme directional shift of tightening mortgage credit market standards that included the collapse of an entire conduit for mortgage originations.  In fact, today, total mortgages outstanding have still not grown to a higher level than they were at the end of 2006.

There is a bit of a V-shape here, because the Contagion cities, which tend to fall in the middle of the valuation range, did experience a bit of a bubble and bust as a result of the housing refugees fleeing the expensive cities, which I have addressed at length.  But, the notable thing here is that there is little relationship between home prices in 2006 and subsequent price changes, between MSAs.  A $690,000 home in San Francisco gained about the same as a $130,000 home in Lincoln, Nebraska.

The difference, overwhelmingly, between 1998 and 2006 is that price expectations in 2006 appear to be much more accurate across MSAs than they were in 1998, because home prices in the highest priced MSAs had been finally bid up to levels that would make future price gains insensitive to the starting price.

Maybe the 30% lag in relative valuations among those mid-range cities is the tell-tale sign of a generation defining bubble?  Then, what does that say about the 100%+ gap in relative valuation shifts after 1998?  Can someone please point me to the section in the library with 50 books explaining how extremely underpriced housing in coastal cities was in 1998?  Or, maybe price expectations in every city were too optimistic in 2006.  Maybe the $130,000 buyer in Lincoln was just as blinkered by optimism at a Price/Rent ratio of 14x as the $690,000 buyer in San Francisco was at a Price/Rent ratio of 28x?

Or maybe, none of this was inevitable.  It is a shame that tightening credit standards and tight monetary policy were not available as choices.  A few respondents did apply a low score to price beliefs and monetary policy.  I count 6 from the US panel and 9 from the European panel that gave both a score of 2 or less, on a scale from 0 to 5.

Robert Hall of Stanford and Chair of the NBER Business Cycle Dating Committee applied a 1 to loose money, home price beliefs, debt levels, underestimated risks, fraud, and savings and investment levels. (He gave himself a confidence score of 1.)  On the European panel, Lucrezia Reichlin of the London Business School gave monetary policy a 0 and price beliefs a 1, with a confidence of 8. Richard Portes of the London Business School gave them a 0 and a 2, and noted that not being the lender of last resort for Lehman would get a 4, with a confidence of 9.  Per Krussell at Stockholm University and Rachel Griffith at the University of Manchester gave both a 1.

I wonder what these respondents would have answered to the missing questions.  The correct understanding of the financial crisis lies outside the Overton Window.  The correct answer to the poll was not an option.  Would any of these respondents have given tight money or tightening credit a high score?  We can't know if we don't ask them.  I suppose that's why the question has been left for a doofus like me to answer.  Not that anyone asked.

Housing: Part 262 - Self-fulfilling prophecies are highly reliable.

From Calculated Risk:

"Usually near the end of a recession, residential investment picks up as the Fed lowers interest rates. This leads to job creation and also household formation - and that leads to even more demand for housing units - and more jobs, and more households - a virtuous cycle that usually helps the economy recover.

However, following the 2007 recession, with the huge overhang of existing housing units, this key sector didn't participate for a few years."

I'm sure I'm a broken record on this, but it really is amazing how much power our priors have in what seem to us to be empirically derived conclusions.

If the overhang of existing housing units was inevitable, then the protracted recession was inevitable, and there was nothing to be done.  If the overhang of existing homes was a result of our resignation to contraction, and, in fact, our insistence upon it, then the protracted recession was collective self-immolation.

The existence of unsold inventory in 2007 or 2008 tells us nothing about which of those interpretations is correct.

There were millions of willing buyers for that inventory, and those willing buyers ran headlong into a national obsession with preventing them from doing so.  Would allowing that to happen have been a self-fulfilling prophecy, too?  Sure.  The fact that this nation overwhelmingly approves of the self-fulfilling prophecy we chose, even as it bankrupted so many of us, threw workers into unemployment, and destroyed the balance sheets of many young and middle class homeowners tells us just about everything we need to know about the mystery of our ailing economy.

The American populace, paraphrasing "A Few Good Men", barks, "Economic stability?! You can't handle economic stability!"

Sentiment is counter-cyclical. Policy views are pro-cyclical

Quite the chart/poll from today’s @barronsonline pic.twitter.com/gUTLdyfnWu

— NoonSix (@NoonSixCap) October 14, 2017

I thought this was an interesting and telling chart.  And it is a nice piece of evidence against the various macroeconomic theories that are built on the idea that pro-cyclical investor sentiment drives business cycles.

This sure looks like counter-cyclical sentiment to me.

This is related to another great chart that I recently came across.  Sentiment had turned south in the housing market well before 2004.

This counter-cyclical sentiment is, I believe, the reason why we have macroeconomic theories that project a pro-cyclical sentiment.  If sentiment in some group is counter-cyclical, then that group will be convinced that their aggregate sentiment is pro-cyclical.  This is why I am only slightly flippant when I suggest that a reasonable monetary policy rule would be to conduct surveys about what policy shift is most favorable, and then do the opposite.

Think about it.  The reason that it seems like housing markets were out of line in 2005 is because practically everyone knew that prices were too high, yet prices kept rising.  When prices kept rising, lo and behold, there were buyers and sellers who were thumbing their noses at our communal sentiment.  There were even some buyers who were pretty excited about it.

Now, there were certainly some types of buyers and some types of speculators who were more active in the market than usual, at the time.  But, our perception of those buyers is skewed by our counter-cyclical perceptions.  Consider that the housing stock is pretty stable.  We don't suddenly double the number of units in a bull market.  And, think about all the anecdotes from 2005.  Small time speculators who owned a half dozen homes in the sand states.  Janitors in California talked into buying a home with a mortgage that would claim 90% of their take-home pay.  Etc.  Add up all the buyers in your mind.  Think about the fact that for that small time speculator, there had to have been a half dozen former homeowners who sold their homes to the speculator.  There are only so many homes to be owned at any time, with only marginal changes, so those homes had to be bought from someone.

How many articles were in periodicals at the time about all the former homeowners who exited the market?  How many books line the shelves at your local library about how many former homeowners sold out in 2005 or 2006?  For every naïve speculator with a half dozen leveraged homes there must have been several.

Our point of view controls these perceptions as much as facts do, and our point of view is counter-cyclical.  We really notice when market activity betrays our point of view.

We can see this in the stock market in the image above.  As prices rise, sentiment sours.  Yet, the price keeps rising.  How can that be?  It's because price is surprisingly immune to sentiment.  Markets take in a huge amount of primary, secondary, and tertiary investor activity and input.  Market participants may not bid up stock prices because they are in a speculative mood.  Maybe, they bid up prices because prices are cheap, in spite of sentiment, and that fact is manifest in prices, even though our primary market sentiment is weak.

How do we reconcile the chart above with the growing number of anecdotes like this one?

---

Uber driver in OC says she and husband were in real estate, lost all in 2008. Only now venturing back into "investment real estate." Hmm.

— Bethany McLean (@bethanymac12) October 14, 2017

---

Do the 4% of potential investors who are feeling bullish and speculative all happen to be Uber drivers and such, or does our counter-cyclical sentiment cause these anecdotes to become more palpable when we are bearish?

We interpret our perceptions as a confirmation of pro-cyclical sentiment.  But, in reality, prices are not particularly sensitive to sentiment.  Intrinsic value of assets is more stable than our sentiment is, but we treat our perception as fixed and we dismiss the efficiency of markets.  So, as our sentiment swings above and below the market, as it generally follows intrinsic value with some small amount of deviation, we come to a conclusion about prices that is the opposite of reality.

And, what is the result?  Where we have the ability and resolve to impose our "theory by attribution error" * on markets, we impose cyclical instability.  When sentiment is bearish, we become convinced that stubbornly bullish sentiment is pushing prices too high, and we demand policies that will dampen economic activity.  When sentiment is bearish, policy becomes bearish, and our sentiments are confirmed.


Update:

Almost 1/4 of those surveyed by E*TRADE expect the market to rise 10% or more during Q4. pic.twitter.com/hduE7Pi3ap

— NoonSix (@NoonSixCap) October 16, 2017

This survey data appears to contradict the data in the above tweet, and suggests that investors expect gains, even though they don't personally think the market is undervalued.  This could be due to speculative fervor, or it could be a realistic expectation of rising real production that causes values to increase along with economic growth.  Even in this survey, the average respondent's expectation is for something just under 5% for the quarter, with a negative skew.  That doesn't seem particularly unreasonable.  But, the negative skew of expectations has definitely declined over the past year.




* "We" base our personal point of view on an assessment of market conditions, the business cycle, and hard earned wisdom.  "Others" are creatures of fear and greed, chasing after every trend.

September 2017 CPI

We continue in a holding pattern.  YOY core inflation is about 1.7%, core inflation ex. shelter remains about 0.6%.  Shelter inflation continues to remain above 3%.  There hasn't been a single month of core ex. shelter inflation above 0.1% since February.

Mortgage growth and bank lending seems to be growing at a rate of 2-3%.  I suppose things could just keep moving ahead indefinitely like this.  But, I think the obsession with asset prices and the tendency toward viewing asset prices as a reason for tighter monetary policy will prevent us from easing when random economic shifts cause a downshift in these patterns when there is no buffer for a downshift.