In my bi-weekly missives I usually discuss a topic germane to financial markets. This time around I’m going to throw you a curve ball and discuss the US housing market. We know the mortgage markets and housing itself were very much the locus of excess in the prior cycle and in very good part responsible for the actual economic and financial market downturn. Many of the wounds from the downturn are healed or on their way. So why is the current housing cycle important?
First, the current housing recovery is closely linked with behavior we are seeing in the financial markets at present — a potentially very important overlap. Secondly, housing was a major contributor to GDP growth in prior economic cycles. The health of the housing market is very important to the tone of the macro economy. So, is that still the case in the current cycle, how is this cycle different than anything we’ve experienced in a half century, and what can we expect ahead?
Let’s start by reviewing what we have been seeing and hearing in the media. Listening to the media, one definitely gets the impression that housing is on fire. This is punctuated when looking specifically at geographies like my home perch in the greater San Francisco Bay Area. It was only a few years ago that we heard stories of millions of foreclosed homes in the “shadow inventory”, just waiting to be liquidated. The stories of today are of inventory shortages. All cash offers now make up close to one-half of all residential real estate transaction activity — an unprecedented number. Bidding wars are now the norm as opposed to being something that only happened at the housing peak in the middle of the last decade.
Below is a look at the history of the Case-Shiller home price index, which measures the change in housing prices in major US metro areas. From the depths of 2009, we are now solidly in positive rate of price change territory with the recent trajectory pointing straight up. As you already know, what has been occurring in places like the SF Bay Area makes the growth rate numbers below look tame.
The major mainstream media message is clear — housing is back and it’s on fire! Right?
Let’s switch gears for a second away from the “sound bite” media characterizations of the US housing market and look at what have been a number of the key factual historical housing market metrics. In other words, these are factual numbers that characterize current US housing activity. I’ll cut right to the punch line; the character of the residential real estate market in the current cycle is completely different than anything we’ve seen in US economic and housing cycles of the last half century. A bold statement? It really is different this time? Let’s look at the numerical facts.
The following chart is the recorded history of new home sales in the US stretching back literally 50 years. Looking at this data, one might naturally ask, “What housing recovery?” New home sales today in the US are consistent with the bottom of historical recessions of every economic cycle over the last five decades. In fact, four and a half years (length of current economic cycle) into each economic expansion/housing cycle, new home sales were twice what we see today.
In the next chart we see exactly the same rhythm of recovery when we look at the history of new housing starts, again stretching back five decades. We are at a level consistent with recession and cycle bottoms of the last half century. Is this what you expected to see based on the mainstream media coverage of the moment?
Now this is interesting. Historically, the average household has financed housing purchases via longer term mortgage debt. As economic cycles have accelerated, increased jobs and wages have triggered increased housing activity and a natural increase in mortgage applications. The chart below courtesy of Calculated Risk shows us the monthly moving average of mortgage purchase applications from 1990 to present. Not only economic cycles, but also longer term demographics (baby boom generation) explain the upward bias to this historical pattern.
The anomaly in the chart is self-evident — there has been virtually no pickup in mortgage purchase applications since the cycle bottom was seen in mid-2010. This is in spite of a meaningful recovery in home prices. There has not been a housing cycle in memory that wasn’t accompanied by a meaningful increase in mortgage applications. Sheer increase in population alone would support an upward bias to these numbers, but that has been absent over the last three years.
Listening to the mainstream media would lead us to believe this is simply a normal housing recovery. Yet the actual data alone set against historical context would lead us to believe housing is in a recession. Why this dramatic juxtaposition? I believe the answer is very simple. This is not an economic cycle for housing, but rather an investment cycle. Normal housing cycles are usually characterized by increased employment, an uptick in wages improving housing affordability, first time buyers taking on mortgages and forming households. Not this time. The cycle has also been importantly distorted by Federal Reserve money printing and interest rate suppression. Completely different than anything we’ve seen in the recorded history of the data. We have a very meaningful increase in home prices, but a recessionary environment for actual home construction and financing (mortgage applications). We’ve never experienced anything like this before.
So if new housing activity and mortgage finance are not driving existing home prices up, what is?
First, the Fed’s suppression of interest rates has forced investors to seek alternative and riskier assets in order to achieve rate of return. The Fed has systematically removed rate of return from safe investments such as CD’s, Treasury bonds, corporate bonds, etc. So investors have taken their cash and purchased rental properties where 6% and 7% cash on cash returns can be achieved, albeit with increased risk. This is exactly why 40-50% of current housing transactions are for cash. Housing has become an alternative investment to safe bond and broader fixed income securities.
Secondly, we know that institutional investors such as Blackstone (largest private equity firm in the US) and Colony Capital have purchased large portfolios of residential real estate as rental investments. We’ve known this has been occurring for years now and this was my investment rationale for buying mortgage backed bonds a number of years back. Interestingly, institutional buyers of the last three years are now issuing long-term bonds backed by rental payments from these properties — in other words, they are starting to shift risk away from themselves and to a point cash out.
Finally, we have seen increased demand from abroad. Remember, central banks globally are printing large amounts of money and openly attempting to debase their currencies. Capital is fleeing these areas and coming to the perceived safety of the US dollar. Residential real estate is a dollar denominated asset. Moreover, when we look at areas such as Singapore and Hong Kong where real estate can easily sell for $1,000-$2,000 per square foot, US real estate looks like a comparative bargain.
I think it’s absolutely crucial to realize and remember that it is investment demand that is driving the current US housing cycle, not a better economy that leads to job and wage acceleration, thus providing the means for families to afford and purchase homes. I believe this next chart exemplifies this. The homeownership rate in the US has been declining in linear fashion since the middle of the last decade — it is reverting to the mean. We don’t yet know if we’ve found a bottom in the current cycle or will continue to decline.
I’ll conclude with a quick look at how housing has influenced US GDP over time. When homebuilders purchase land and build, they provide new construction jobs, new home sales influence household appliance sales, home improvement retail, etc. Housing construction in past cycles has driven 5+% of total GDP. Not this time. As of now the contribution of residential real estate construction has returned to a level that is a half century low for US experience. Housing is not a meaningful contributor to US GDP in the current cycle.
What are the key takeaways from this discussion?
First, this is an investment cycle, not a normal economic/housing cycle. What we are seeing is yet another distortion of Fed money printing, forcing investors into ever riskier asset classes. As the Fed lowered interest rates to zero, it forced investors into alternative investments such as real estate. So naturally we need to ask ourselves, what happens when the Fed can no longer print and what happens when interest rates ultimately increase? For housing investors or those even considering selling their houses at some point, this is a key question.
Secondly, real estate is “acting” very much like US stocks — one of the last asset classes providing a positive rate of return this year. I personally believe residential real estate and US equities will act very coincidentally ahead. Capital is now “herding” or concentrating in these two asset classes as prices rise ever further. In past cycles, this type of herding has meant risk is increasing.
Finally, we continue to see historically subdued macro-economic growth in the current cycle. One of many reasons for this is subdued new housing construction activity. Until this changes, housing will not be a key contributor to US GDP acceleration.
This is not your fathers’ (or mothers’) housing cycle. Ours is something completely out of step with historical experience. I hope looking at the historical facts help put into perspective just how different the current cycle has played out relative to past cycles. It also tells me its ultimate conclusion will likewise be different. It will end when investment demand changes, not necessarily when real world economic circumstances change. What has primarily driven investment demand for residential real estate is the Fed sponsored artificial suppression of interest rates in the US. When the day ultimately arrives that the Fed no longer will or can print money to keep interest rates in check, will that also spell change for residential real estate? It’s hard to see how this will not be a catalyst for change. It’s different this time — really.