Monday, April 22, 2019

IW on the web

Tech entrepreneur David Siegel has an interesting and thorough post up at that is a sort of reference point to cutting edge or wise thinking on a vast array of topics.  One reason I am posting a link here is because David kindly includes the work of Scott Sumner and myself on the financial crisis and the housing bubble, prominently, as work that should be read and understood.

Regarding our work, he begins with:

Understanding the Great Financial Crisis

A good way to see the storytelling effect is to look at the “common wisdom” of the Great Financial Crisis of 2008/9, an event that impacted every person on earth and destroyed a billion jobs. Almost everyone got the story wrong. Michael Lewis’s book and movie, The Big Short, was popular but completely missed the true cause and effect. So did Niall Ferguson and many experts.
In reality, two people — Kevin Erdmann, an investor and Scott Sumner, an economist — have shown that the “common wisdom” does not fit the facts. Using the scientific method and hard evidence, they show that the GFC was a result of bad reactions to scarce resources

I appreciate David's support and his willingness to consider this new point of view.  But, in addition to that, his post can be fruitfully used as a starting point into inquiry in a number of topics.

He lists my book "Shut Out" as an "advanced" reading.  For those visiting IW from David's post, if you don't want to dive into a long tome of "advanced" reading on the topic, here are a couple of shorter pieces that may get the ball rolling.  I'm not sure they are any more accessible, but they are much shorter, and introduce the basics.  (My writing tends to be analytical rather than narrative, but I don't think you will find any of my work to be nearly as difficult as, say, the typical academic article in an economics journal.)

Housing Was Undersupplied during the Great Housing Bubble

The Danger in Using Monetary Policy to Address Housing Affordability

Thursday, April 11, 2019

March 2019 CPI Inflation

Here are the updated inflation numbers.  Non-shelter core is down to 1.1%, shelter is still at 3.2%, and core CPI inflation is at 2.0%.  As IW readers know, the reason this is important is that (1) shelter inflation is largely an imputed figure of rental values of owned homes that involve no cash transactions and, (2) in the era of Closed Access, in some important markets, these transfers have little effect on production.

This is the setup that I worry will cause the Fed to be behind the curve.  They believe that merely stopping the rate hikes will be enough.  Of course, in this context, the inverted yield curve is also a bad sign in this context.

It's not so much that 1% inflation would be an automatic disaster.  I'm not even sure it's a great recession indicator.  It's more a problem of being shielded from timely cyclical developments because of misreading the measures that should lead to shifts in policy trends.

It seems that, along with the Great Moderation, has come a peculiar Fed behavioral tick, where the Fed Funds rate is held for some time at a plateau, which is followed by a contraction.

Monday, April 8, 2019

Real Phillps Curve Update

Here are a couple of charts comparing real wage growth and unemployment.  My contention is that the Phillips Curve is real, not inflationary.  It only appears to be inflationary when monetary policy is procyclical.  When unemployment is low, real wage growth is higher, largely because of better matching, fewer frictions in labor markets, and higher labor productivity.

If we treat the Phillips Curve as nominal, then the inclination is to reduce growth to prevent inflation, and unemployment will be invariably driven higher in a misguided attempt at moderation.

If we treat the Phillips Curve as real, then the inclination is to celebrate low unemployment unconditionally, and allow the benefits of highly functional markets to continue to accrue.

There is a relatively stationary long term relationship between real wage growth (I prefer using CPI less food, energy, and shelter as the deflator) and the unemployment rate.

We shouldn't be afraid of real wage growth.  And, in either case, wage growth is humming along pretty close to the long-term trend.  Celebtrate that unconditionally.

Wednesday, April 3, 2019

Housing: Part 348 - How Affordable Is Housing?

The other day, I looked at Price/Rent ratios over time in various cities.  In Dallas and Atlanta, by this measure, prices have been relatively similar in 1998, 2006, and 2019, but they were extremely low in 2013.  Other cities I looked at were more mixed.  Prices looked high in 2006, and generally looked higher in 2019 than they had in 1998 or 2013.

Here, I will look at the same cities, in the same way, but here, I am looking at mortgage payments/rent payments, which is another way to think about affordability.

A couple of caveats:
1) Here I am using the 30 year conventional mortgage rate, but I am applying it to a 100% loan-to-value.  I know that's not realistic as a mortgage product, but it's a way to get at the relative value here.  I don't want to give the buyer a 20% advantage just by assuming a down payment, but my point here isn't particularly to look at mortgages with higher spreads.  In any case, it is the relative values over time that are informative, so this shouldn't matter that much.

2) All data is from Zillow.  (Zillow rocks.)  But, they only have rents back to 2010, so 1998 and 2006 are estimated rents based on metro area level affordability measures and prices.  It probably doesn't matter that much for 2006, but 1998 might be taken with a few grains of salt.

Looked at in terms of mortgage payment/rent payment for a house at a given price level, housing affordability for buyers is well below levels of 1998 and 2006.  For instance, in Atlanta, in 1998 or 2006, the mortgage payment on a $160,000 house would have been similar to the rent payment.  In 2013 and 2019, the mortgage payment on a $160,000 house would be about 60% of the rental payment.

By this measure, even in 1998, mortgage affordability would have been pretty good.  Price/rent was near long term lows and mortgage rates had been high since the 1970s.  By this measure, home prices are well below historical norms, and for buyers of those homes, there is little interest rate risk, because the homes are already priced at a discount when taking interest rates into account.

People who are complaining that loosening lending standards will push prices up to levels that hurt affordability have a (somewhat) plausible point.  But, this measure is why I think that they are applying that point at a standard that reflects their own preference for tight lending more than it reflects a half century of American homebuying norms.

Also, of course, it is noteworthy that from 1998 to 2006, mortgage affordability improved in Dallas and Atlanta.  That's what an elastic housing supply does for you.  On the other hand, in this measure, it was only the high end of Dallas and Atlanta that became more affordable.  One could argue that this is a sign of loose lending.  On a Price/Rent ratio level, prices in Dallas and Atlanta rose fairly proportionately.  But, if we think of Rent/Price as a sort of yield on housing, the yield is higher on low end homes, which means that their prices should be less sensitive to changing risk-free rates.  (Going from 9% to 8% has less of an effect on price than going from 4% to 3%.)  So, prices should rise less because of falling risk-free yields in low tier homes than in high tier homes.  If these estimates are accurate, though, it appears that low tier prices in those cities were more sensitive to changing interest rates, so that falling rates increased prices as much as they decreased the mortgage payment. (Although, keep in mind, the 1998 data I am estimating here wouldn't be dependable at that level of granularity.)

Here is San Francisco, too.

One interesting pattern here is that mortgage/rent ratios are pretty similar in every city.  For instance, a $160,000 home in every city would have a mortgage/rent ratio of somewhere around 50-60%..That's the case from Dallas to San Francisco.  Of course, there aren't many $160,000 homes in San Francisco.  But, the pattern is curious.

Tuesday, April 2, 2019

March 2019 Yield Curve Update

Since the zero lower bound distorts the yield curve at very low rates, an inverted yield curve at low rates is worse than an inverted yield curve at higher rates.  This is a reason why the curve didn't invert in the 1950s.

The necessary adjustments here could be made either by just looking at the short end of the curve, recognizing that the long end of the curve will have a bias for a positive slope. Or, it could be estimated with a regression of the depth of the inversion in the various economic downturns that have happened since WW II.

Here are visualizations using each estimate.

In the Eurodollar market, the inversion of the short end is very deep.  Today rates bounced up a bit.  They will do that.  In the post-WW II era, though, even though rates seem to bounce around within the inversion, yield curve normalization has only happened when the short end rate has been lowered.  At this point, I have a fairly strong expectation that short term rates will be well below 2% before 2021.

The estimate using the 10 year minus Fed Funds spread also has us well into inversion.  As this graph shows, in 2006 and 2007, the 10 year rate moved up and down without normalizing for some time, then, when the Fed finally lowered rates, the entire curve came down.

The inversion from early 2006 to summer of 2007 was especially extended.  I think the Fed was already sucking cash out of the economy far to aggressively for that whole period.  The housing boom is what delayed contraction, because many households could access their home equity for liquidity.  Lending was still growing at something close to double digits even into mid 2007.

Today, this source of liquidity is not significant.  Mortgages are growing at low single digits, if that.  Home equity is still declining.  General bank lending is at around 5% annual growth.  So, I expect falling rates to come sooner this time.  But, admittedly, I was surprised by a rising yield curve after the Fed raised rates in 2015, so my credibility on this point is worth the monthly subscription price to this blog.

Whatever else happens, this is definitely an inversion event at this point.