Tuesday, July 30, 2019

Part 10 of my Housing Affordability series at Mercatus

Property Taxes Can Be a Tax on Monopoly Power.

"If politically maintained monopoly power is going to remain, claiming monopolist profits through taxes is an improvement. The fact that the tax doesn’t affect rents is a sign of efficiency. If rents must be elevated, better that they go to local public services than to the real estate cartel."

The series will continue each Monday with discussion of the effect of various regulations and taxes on housing costs.

Sunday, July 28, 2019

Housing: Part 356 - Black Homeownership

Here is a new Bloomberg article on black homeownership.  The title is:
"Black Homeownership Falls to Record Low as Affordability Worsens"

The headline, and the article, are wrong.  Affordability isn't bad and that isn't why black homeownership is falling.

Here is a graph in the article, which also has an incorrect headline.  It says, "Over 25 years, the gap between blacks and whites has widened."  What the graph really shows is that from 25 years ago to 15 years ago the gap was narrowing, and then for the past 15 years it has been widening.

Here is a graph that combines old decennial Census data with the more recent quarterly data to provide a little more historical comparison.

From the Great Depression to the late 1960s, white homeownership rose as a result of Federal programs that explicitly excluded black families.  Then homeownership for black families increased, but then fell back again in the 1980s, for reasons I am not familiar with.  Then, in the late 1990s, it recovered back to the levels of the late 1970s, relative to aggregate US homeownership rates.

Then, homeownership peaked in 2004 for black families as well as for the US in general.  And the drop in ownership since then has been stronger among black families than among others.

To describe these trends with "Over 25 years, the gap between blacks and whites has widened." obscures what is important.  Black homeownership was recovering and expanding when mortgages were more available.  Note, however, that the recovery in ownership peaked near the beginning of the private securitization boom that lasted from roughly 2004 to 2007.  Since then the relative homeownership rate collapsed.

The disparate impact of the recession, the crisis, and the subsequent sharp tightening of lending standards on black households has been strong.  But, who would dare to claim that the pre-crisis housing market was good and that post-crisis lending market has been detrimental?

Here are the Zillow measure of mortgage and rent affordability - the portion of the median household's income required to pay the rent or the mortgage on the median housing unit.  It would be more accurate to say that it has been especially unaffordable not to be a homeowner in recent years.

Wednesday, July 24, 2019

Housing: Part 355 - Homes and population growth

I noticed that housing units per adult has actually started to level off.  This is interesting because total permits and total starts are still below past averages.

So, maybe the new neutral run-rate for new units is less than 1.5 million annually.  Maybe the need for new units is less acute than I have been saying.

But, there is a problem of causation here.  More people means we need more homes, but also, a lack of adequate housing can lead to less people - both by limiting migration and by limiting family formation.

And, it is true that population growth has slowed.  Before the financial crisis, it tended to run at 1-1.2%.  Since the crisis, it's more like 0.7%.  So, in a way we solved the housing shortage, in part, by reducing population growth.  If this is the new normal, then maybe 1.2 million units a year isn't an unsustainably low peak.  But, if population growth, either through immigration or through family formation, returns to anywhere close to historical norms, then housing starts probably need to catch up a bit and then settle at something closer to 1.6 million units annually.  (Ignore the big drop in housing/adult in 2000.  I haven't taken the effort to try to account for the discontinuity in the data there.  I suspect that mostly that discontinuity comes from an overestimate of the housing stock in the late 1990s, but it isn't central to the main trends I am discussing here.)

Certainly an argument can be made that population growth through family formation has been naturally slowing, so that we shouldn't expect population growth to continue at historical norms.  On the other hand, there are many good reasons to counter that decline with more generous immigration policies.  And, while there is a long term down trend in natural population growth, there was a sharp downshift that appears to have been related to the economic turmoil of the crisis and to the lack of housing growth since then.  Even without immigration, it seems likely that natural population growth has declined more than it otherwise would have after the crisis.

Also, there is always the important signal here of rent inflation, which has persistently run high for the past 25 years and returned to high rates during the post-crisis recovery.  That is not a signal we would see in a country where housing was being depressed by natural declines in population growth.

Tuesday, July 23, 2019

The latest posts in my Mercatus Housing Affordability Series

The last two posts in my series were:

"Tight Lending Regulations are a Wealth Subsidy".  An excerpt:

Thinking in terms of rental value, public policies and market innovations that lower mortgage interest rates can be broadly beneficial to consumers, even if those benefits don’t accrue to the actual borrowers who use those low rates.  That is because higher mortgage interest rates have a similar effect on price as exclusionary lending standards.  Downward pressure on price creates a rental subsidy for home buyers who don’t require a mortgage.

"Property Taxes Are Rent to a Public Landlord" An excerpt:
If there is concern that the net effects of government policies, in total, favor housing and lead to market volatility, a return to higher levels of property taxation can be a useful tool for countering it.

Writing the series helped clarify my thinking on several issues.  I hope you find some nuggets of interest in it too.

Thursday, July 11, 2019

June 2019 CPI Inflation

Here is my monthly inflation update.  We continue along in the same pattern.  This month there was a bit of a bump in non-shelter inflation, but the trailing 12 month rate remains about 1.1% and shelter inflation remains about 3.4%.

Going forward, I think inflation may become a less important indicator.  The Fed has shifted to a more dovish posture and they are not insisting on holding the target rate at a plateau.  It would be a shock if they don't lower rates this month.  So, I am happy to say that my worst fears appear not to have come to pass.  Monetary policy is on the margin of neutral.  Unless the Fed reverses course, I suspect there will either be a slight contraction or a continuation of the expansion.  For now, I will call that a tentative prediction, but it seems to be where we have moved.

We are probably near the point in time where a tactical long position in fixed income should shift into more of a long position in equities and real estate, either now or over a few months as this plays out.

In terms of broader influences, I'm more worried about nominal growth rates in Australia and Canada than things like the tariff issue, but I'm no expert on those issues.  That's just my hunch.

Monday, July 8, 2019

Squeezing "Unqualified" Borrowers

The latest post in my Mercatus bridge series.

More on how recognizing the key importance of rent as the measure of affordability - for both owners and renters - helps clarify the issue.  Tight lending is making housing less affordable for renters.

Considering this set of circumstances, the idea that housing affordability is getting worse because prices are high and that the solution is even higher interest rates or tighter credit access is a disastrous misreading. It will lead to a vicious cycle of segregation between households that can qualify under today’s standards (and who then can buy ample units at favorable terms) and households that cannot qualify (and who must keep economizing while a large portion of their wages is transferred as rent to the ownership class).

There are two options. Re-opening credit markets to entry-level buyers will return the market to a more equitable equilibrium. Maintaining the market as it is will continue down the path of settling at a new equilibrium where certain households live in smaller, less adequate units, either because of size, amenities, or location.
Please read the whole thing.

Here is the link to the full series.

Sunday, July 7, 2019

Housing: Part 354 - Nashville follow up

I wanted to revisit one graph, because I think it tells the story so well about what's happening in many US cities while lending standards are tight.

In the process, I realized that I should have adjusted for inflation, and in the process of doing that, I realized I had a minor excel worksheet error.  Here is the chart with the error fixed and the dollars constant.

In the last post, the linear trendlines were pretty nearly lined up.  But, the things that would affect the price/rent relationship should generally scale with inflation, so this is probably a more accurate portrayal of the Nashville market.  There has been some recovery of price/rent ratios in the low-to-mid part of the market.

At the top end of the market, P/R ratios are up about 5 points since the bottom, which was around 2011.  The bottom should be up at least that much too.

Low tier prices have risen as much or more than high tier prices.  But, as I pointed out in the previous post, this is because of low tier rent inflation, and the positive feedback of units with higher rents moving up to higher price/rent ratios.

That is still evident in this corrected graph.  At the high end, adjusted for inflation, price changes since 2011 are generally due to a recovery in price/rent levels.  That is the part of the market where building is taking place.

At the low end, little building is taking place, and rising prices are largely from rising rents.  Here, we can see that, even adjusted for inflation, the bulk of zip codes have moved from rents typically around $1,100 per month to rents more like $1,300.  P/R ratios in that part of the market have risen by around 2x and the rise in rents led to an additional P/R expansion of another 2x or so.  So, the low-to-mid part of the Nashville market moved from $1,100 rents at a P/R of 10x to $1,300 rents at a P/R of 14x.  The combination of those things was enough to cause those areas to appreciate in price faster than high end Nashville where rents have remained about the same in real dollars and P/R has increased by about 5x.

Some of the increase in rents is due to gentrification and in-fill capital improvements, but the tendency to blame those capital improvements for the increasing problem of unaffordable rent completely misses the point.  Rents won't come down until those segments of the market get as much capital as the top end is getting.  And, the top end is getting a lot.

Friday, July 5, 2019

June 2019 Yield Curve Update

Rates have continued to dip.  Forward markets have already moved much of the way back toward zero.  This has been somewhat surprising to me.  I expected the Fed to maintain the Fed Funds rate at a plateau level, as they did in the last two cyclical reversals.  But they are almost certain to start to lower the target rate this month, and that is great news.

There is still some potential for trading gains in forward rate markets, I think, because short rates are highly likely to return to near zero.  I hope the newly dovish turn by the Fed is enough to give that move some oomph.  I think an important signal will be the long end of the curve.  If it remains low as the Fed lowers the target rate, this is a sign that the Fed is following the neutral rate down, and isn't really inducing nominal growth.  It will be a bullish sign if the long end of the curve moves up.

In fact, using the adjustment I make to the yield curve, at today's levels, the 10 year treasury yield would still be effectively near inverted rates even if the Fed lowers the target rate to zero.  My worry is that mistaken associations between low rates and loose money will prevent the Fed from being aggressive enough.  But, recent Fed communications have been more promising.

The first graph here shows the 10 year rate vs. the Fed Funds rate, and shows my modeled inversion indicator.  By this measure, the curve has been inverted for many months and moved much farther into inversion territory this month.  In some ways, that is a good sign, because it reflects expectations of near-term Fed rate cuts.  But, ideally, it would be better if long term rates held firm.  The fact that long term rates are declining along with short term expectations is a sign that frictions in credit markets were keeping long term rates high.  To me, this is the best way to think about yield curve inversion.  Some set of frictions in the market prevent long term yields from declining to unbiased forecasts of future short term rates when the yield curve is inverted.  I suspect that this causes problems with credit allocation that may be a causal element in the contractions that tend to follow inversions.  If long term rates decline when short term rates are lowered, that suggests that lowering rates has removed those frictions and allowed long term rates to move to a less biased level.  So, the good news is that Fed dovishness is helping to offer relief to markets, but the bad news is that this means an inversion is in effect and that usually leads to a contraction.

The market is currently priced to expect the dots on my scatterplot to move sharply to the left as short term rates are lowered.  But, the key to avoiding a recession is for the upcoming dots to also move up.  If they don't, then I suspect that we will be playing catch-up and economic growth expectations will remain subdued, which will continue to lead capital to safer assets instead of into riskier investments that can trigger productivity and employment strength.  It is hard to tell in real time, but it is beginning to look like real GDP growth peaked a year ago and that the 4 quarter real GDP growth rate will move back down below 3%.  One might expect real growth to level off as unemployment bottoms, but employment growth has been pretty stable since 2012 at 1.5% to 2%, and continues to move in that range as workers re-enter the labor force, so changes in employment growth don't point to a GDP slowdown yet.

The best thing that could happen is the Fed lowers the target rate aggressively, long term rates rise with new real growth and inflation expectations, and then FOMC members and pundits who incorrectly view lower rates as a stimulus to risky investments will interpret higher rates as less stimulative, and they won't pressure the Fed to stop lowering the target rate.

Wednesday, July 3, 2019

Housing: Part 353 - The Seemingly Strange Case of Nashville

There are two core constructed details that have formed a basis for much of my analysis about the 21st century housing market and the financial crisis.

  • Price/rent ratios tend to rise as rents rise, but at some point in each metropolitan market they reach a ceiling.  This means that (1) excessive price appreciation in low tier homes during the housing boom in cities like LA and NYC was mostly a product of rising rents. and (2) The core error of the FCIC and most analysis of the crisis was missing this fact and blaming rising prices on aggressive credit markets instead.
  • In most cities, rents were moderate enough that there was not an unusual rise in low tier home prices from this effect, but after the boom, when credit was greatly tightened, low tier prices were decimated, frequently falling more than 20% compared to high tier prices.
This first graph basically tells that story (PS: Many thanks to Zillow.com for making so much price and rent data public):
idiosyncraticwhisk.com 2019
Data from Zillow

LA is highly unusual, both for having such high price appreciation and for having such a divergence between the high and low end during the boom.  These are related.  They both come from the extreme shortage of supply relative to demand for housing in LA.

Seattle is more expensive than Atlanta because incomes are higher there and supply of housing is more constrained, though much better than LA.  So, you see a bit of difference between Seattle and Atlanta during the boom, but little difference between the top and low tier of each city.

Then, during the bust, bottom tier home prices in both Seattle and Atlanta collapse, to the point where low tier prices in Seattle had total appreciation that was no more than high tier appreciation in Atlanta.

idiosyncraticwhisk.com 2019
Data from Zillow
I recently had occasion to look up data in Tennessee.  I have gotten so used to seeing this pattern that running the numbers has become rote.  Almost every city looks something like Seattle and Atlanta.  So, I was quite surprised when Nashville looked like this:

Nashville looks like Atlanta before the crisis and Seattle after the crisis, and it doesn't have the lagging low tier price appreciation of either of those cities.  In fact, it is high tier prices that have been lower in recent years.

What gives?

It turns out that in recent years, Nashville has been on fire, economically.  Population growth, in-migration, rising incomes.  Things are going really well there.  Things are going so well that housing supply pressures are making it look more like a Closed Access city.  Well, it's more the case that there are two Nashvilles.  The top half of the housing market operates like an open access city before the crisis.  The bottom half of the housing market operates like a closed access city because new tighter lending standards are preventing owner-occupiers from buying homes in those sub-markets.  This has compressed price/rent ratios so that yields are high enough to induce buying by landlords.  This can happen through lower prices or by rising rents.  In practice, it can be a little bit of both.  In Nashville, it appears that economic success has led especially to rising rents, because pressure for residency in Nashville is pushing up demand for Nashville housing.  At the top end, this leads to more supply.  But, that demand pressure also appears to be seeping into the low tier, where it can only push up rents, because buying pressure is limited mostly to landlords and they are still mostly just buying up the existing stock, apparently at price points that still can't induce much new supply.

Here is a Fred chart of housing permits in Nashville.  The red line is single family homes and the blue line is multi-unit homes.  Both are very healthy.  Pre-crisis Nashville had strong rates of new home building.  It may be unique among cities where building was well above the national average before the crisis and has recovered to those pre-crisis levels.  You just don't see this in other cities.

I presume that eventually, rents will rise high enough to trigger even more building at the low end, putting a stop to excessive rent inflation.  But, it hasn't happened yet.  Though, multi-unit starts are very strong.  To the extent that investors will build new stock, it will tend to be multi-unit.

idiosyncraticwhisk.com 2019
Data from Zillow
Here is a graph of median rent and mortgage affordability in Nashville and in the US over time.  (Again, all hail Zillow.)  The national story here is that rent affordability has been high (though it has moderated recently) but that mortgage affordability has never been better.  There has never been more reason to loosen lending standards.  This is basically why the low tier of most cities is lagging in price and supply with rising rents, because we have financial gatekeepers preventing potential low-tier buyers from closing this financial arbitrage gap.  Price is not the moderating factor keeping mortgage expenses so low.

But, note what the Nashville story is here.  It has traditionally been an exceptionally affordable city, in terms of rent.  But during the housing boom and after, that gap has closed, and Nashville isn't particularly affordable any more.

idiosyncraticwhisk.com 2019
Data from Zillow
Because of these high-tier vs. low-tier supply issues, this affordability problem is especially pronounced in low-tier Nashville neighborhoods.  Zillow only has rent data from 2010, but here is a graph comparing aggregate median rent levels in each zip code in Nashville from 2011 to 2019.  The x-axis measures the starting median rent and the y-axis measures how much rent has increased in that zip code since then.

More affordable areas have experienced rising rents much higher than more expensive areas.  So, the median rent affordability measure above really splits a divide between top-tier areas where rent affordability has remained low and low-tier areas where it has moved up more.  In Nashville, this has been strong enough factor to swamp the compression of price/rent ratios.

idiosyncraticwhisk.com 2019
Data from Zillow
And, this brings us back to the bullet points at the beginning.  The counterintuitive issue at the core of the question of rising home prices is that rising rents cause price/rent ratios to rise.  Here is a comparison of rents and price/rent ratios in zip codes in Nashville in 2011 (blue) and in 2019 (red).  As we can see, the typical pattern holds.  Price/rent ratios rise as rents rise, up to a point, where they level out.  At the top end of the market in Nashville, price/rent ratios have increased since the market bottomed, and top end price/rent ratios now average around 17x or so, up from around 14x in 2011.

One would think that price/rent ratios at the bottom end would have to have expanded at least that much, because prices have appreciated at least as much at the bottom.  I have added linear trendlines here, reflecting the portions of Nashville that are not at the peak price/rent level.  As you can see, that relationship hasn't changed much since 2011.  A typical unit renting for $1,200 has a price/rent ratio that is right at the same level it would have been in 2011.  But, rising rents have pushed all housing units up this price/rent ratio incline.  This is basically the same effect that was happening in places like LA before the financial crisis.

The long and short of it is that there are zip codes in Nashville where rents might have been $1,000 per month and looser lending may have pushed price/rent ratios up from 10x to 12x.  The trend line in this graph would have moved up.  Instead, because of tight lending, rents in those zip codes are more like $1,200 with price/rent ratios around 12x.  The trendline hasn't moved at all, yet this doesn't make housing more affordable.  This is one of many reasons why the focus on affordability should be on rent, not price.  Rent is the coherent source of information for that question.

I have concluded that the relative rise in low-tier prices in cities like LA during the bubble was unrelated to loose lending markets.  That is a tough argument to make, because it coincided with loose lending markets, and it just seems to make sense that loose lending would create new buyer demand that might push prices up.  But, here, in Nashville, we can see the same effect, and here, the effect coincides with tight lending.  In both cases, however, rising rents and rising price/rents coincide with limited supply.

Tight lending standards have created the same context in the rest of the country that supply constraints in Closed Access cities had created before the crisis.  Any positive economic developments will create a side effect of pushing up the cost of living for families with the lowest incomes.  Eventually, I presume, Nashville will hit a rent level that pushes prices high enough to induce enough investor building to level off rent inflation.  There will be a new normal, where the level of rents will be higher for low-tier tenants relative to where they used to be, but once we hit that level, the rate of change in rents should level out.

There is no rule here that points us to the correct place.  Maybe access to mortgages should be tightly regulated and housing should be more expensive than it used to be for low-tier tenants.  An advantage of that market norm would be lower rates of mortgage defaults, etc.  It would be a safer equilibrium with less volatility and less punctuated distress.  But, the cost of that safety comes at the expense of low-tier tenants.  They replace less punctuated distress with more chronic distress.  And, if prices are going to be depressed by limiting access to capital, then that means, mathematically, that we are enforcing a system of inflated returns to those who happen to have capital.  Again, maybe that's ok.  We just need to be honest about the implications of these lending norms.

If this is the new normal, then most cities have a few decades to look forward to that look like Nashville today.  Economic success will mostly simply mean rising cost of living for households with lower incomes.  This will be blamed on all sorts of supposed problems with laissez-faire markets, but most of it lays at the feet of a national consensus that has supported an extreme regime shift meant to make real estate markets less volatile.  Supporting an economic structure that benefits all Americans will require coming to terms with the pros and cons of that consensus.

Follow up.

Monday, July 1, 2019

The next post in my Mercatus series on housing affordability

Here is where you can see the entire series as it is posted:

Here is the latest:
"The Myth About Bubble Buyers"

A lot of this particular post will probably be familiar to long-time IW readers.
(F)or households 45 to 54 years in age, the homeownership rate in 1982, when the Census Bureau started tracking it annually, was 77.4 percent. It bottomed out at 74.8 percent in 1991 and then recovered to 77.2 percent at the peak in 2004. By 2017, it was down to 69.3 percent!
Rental expenses as a proportion of incomes (Figure 1), belie the conventional wisdom. The rental value of owned homes was more stable as a portion of owner income than the rental value of rented homes from the late 1990s to the mid-2000s. In other words, if there was an increase in relative spending on housing, it was among renters. The rental value of homeowners was rising in line with their incomes. There is no sign of marginal homebuyers being induced into homeownership and overconsumption.