New York | Count the numbers of markets and economic indicators currently at extreme valuations and positive correlations. Financial markets showed a glimmer of normality last week when stock prices fell and bonds rose. The surprise expressed by market participants is an illustration of just how long investors have been dealing with a market where all manner of assets are correlated and at multi-year highs. We talked about this last week with our friends at BNN in Toronto in a TV hit from Bloomberg TV in New York.
To our readers: We are ending our weekly email notifications via our third-party listserve. If you’d like a reminder from Wix when new items are posted, please register again at www.theinstitutionalriskanalyst.com and you can manage your subscription directly. As before, new items will be posted on Twitter, Google and other social media. Apologies for any inconvenience. – The IRA
Markets were abuzz last week over the prospect of a trade war, but we see President Donald Trump’s latest outburst over tarrifs as the opening salvo in the 2020 election cycle. Yet a number of observers are openly wondering if the latest upward move in interest rates is essentially finished and if the next leg for the 10-year Treasury is a bull rally back down to the low 2% range in yield. As the chart below illustrates, the 10-year popped to 2.8% yield following the November 2016 election, then re-traced back down just shy of 2% yield on September 8th of last year.
Despite the latest upward move in rates, we remain in the bond bull, flattening yield curve camp for the simple reason that there is still far too much liquidity – call them dead presidents -- chasing scarce assets. Even with credit problems emerging around the world following eight years of irrational easing by the Federal Reserve and other central banks, the markets and particularly credit spreads remain remarkably calm – almost too tranquil. It’s as though that last big dose of monetary thorazine from former Fed Chair Janet Yellen still has not quite worn off.
Should long-term interest rates continue to rise, however, we think it is fair to ask whether the period of low credit costs and artificially boosted economic activity in the US also is ending. It is clear that the proverbial party is over in autos, with rising credit standards and falling sales incentives more than tapping the brakes on industry volumes. The particularly nasty chart below shows monthly light vehicle sales in the US through January 2018.
Default rates on prime auto loans held by banks have essentially doubled since 2015 and now stand at 1%, which is really not a big deal compared with unsecured loans. But the rate of change is notable. The Wall Street Journal reported last week that residual values for cars coming off lease are falling, one reason why dealers and the automakers have pulled back on costly sales incentives.
By no coincidence, loss given default (LGD) for auto loans, which is calculated as charge-offs less recoveries divided by charge-offs, has risen 10 points over the past three years to just under 70% of the original loan amount. LGDs are essentially the inverse of asset prices in a given loan category, thus rising LGDs for auto loans is another way of saying that prices for used cars are weakening.
Our pals at the Federal Deposit Insurance Corp reported last week that provisions for future credit losses for all loans and leases have increased nine quarters in a row, albeit from very low levels seen in the trough of 2015. But the key question facing investors is whether the end of the period of extraordinary ease by global central banks will now see a decline in economic activity and an increase in credit costs for US banks and bond investors.
The secular increase in asset prices for stocks, bonds and particularly real estate is exhibit number one in this analysis. As we’ve noted in past issues of The Institutional Risk Analyst, the credit metrics coming from US banks in asset classes such as 1-4 family home loans and multi-family real estate are anything but normal and suggest an adjustment down the road. Will home sales and even prices follow the bearish example of autos?
It is interesting to see, for example, that past-due 1-4s owned by banks actually rose to 2.7% in Q4 ’17 after falling steadily for the past five years. More thought-provoking, however, is that fact that record low LGDs for bank owned 1-4s have stabilized at just 24% of the original loan balance vs the 25-year average of 65%, as shown below. Is the next leg up?
Like the chart for auto sales above, the LGDs of 1-4 family loans exhibit a large degree of skew from historical norms, something you’d expect to see after years of experimentation by Chair Yellen and her colleagues on the Federal Open Market Committee. The researchers at the San Francisco Fed (hat tip to Rosie) put it very nicely in a January 8th comment:
“Current valuation ratios for U.S. equities and household net worth are high relative to historical benchmarks. The cyclically adjusted price-to-earnings ratio reached its third highest level on record recently, and the ratio of household net worth to disposable income, which includes a broad set of household assets, stands at a record high. Such extreme values of these ratios have historically been followed by reversions toward their long-run averages. However, other current factors, such as low interest rates, caution against bearish forecasts.”
So if Yellen and company have front loaded a decade of growth into the past five years, what does that say about prices for stocks, bonds and particularly real estate? While the price increases illustrated in the world of 1-4 family loans suggest a considerable deviation from long-term averages, the degree of skew in the world of multifamily real estate is even more pronounced and suggests a proportionately great degree of asset price distortion.
The chart below show LGDs for the $400 billion in bank owned loans backed by multifamily residential properties. Like 1-4s, the default rates on this asset class are extremely low – in large part because distressed debtors are often taken out of these exposures short of a formal default. Last quarter, LGDs for bank multifamily exposures fell to minus 109%, meaning in cash terms that recoveries exceeded charge-offs 2:1.
This chart describing relative change in collateral valuations in the multifamily sector suggests that these assets are overvalued and must, eventually, revert to the LT mean. Also, if the weakness in autos suggests a more general slowing of economic activity, particularly among consumers, how much should banks and bond investors expect loss rates – and LGDs – to rise in coming months and years in related asset classes?
The good news is that both 1-4s and multifamily assets are quite solid historically, with the 2007-2010 period of extreme losses standing out. Losses on bank owned 1-4s peaked at the end of 2009 at a whole 2.47% for $2.5 trillion in loans. Multifamily loans held by banks saw net charge-offs peak at just 1.77% at the end of 2009. The chart below shows loss and past due rates for both 1-4 family and multifamily loans.
Compared with unsecured consumer loans, the real estate exposures of US banks have performed quite well over the past 25 years. For example, the charge off rate on the $850 billion in credit card receivables held by all US banks at year end was 3.77%. Of course, credit cards are a much more profitable product for banks than making loans on real estate.
But the unusual behavior of the credit loss experience of these asset classes today suggest that prices are very toppy indeed. Both the low loss rates and LGDs displayed by real estate portfolios illustrate the high value of the collateral behind these bank credit exposures. The key question for investors and risk mavens is when or even if these valuations actually adjust downward.
When you price credit at zero and compress spreads via brute force methods like quantitative easing, it is easy to conceal a lot of sins in credit terms. For example, the dearth of assets engineered by the Fed has also led to a deterioration in credit covenants and other investor protections.
Yet even today, few economists inside the Fed system have publicly stated that QE was overdone, that is distorted markets, and has perhaps created the circumstances for the next financial crunch. Fed Chairman Jerome Powell brushed aside a new paper by two Wall Street economists and two academics questioning the effectiveness of QE, but the next couple of years may reveal some significant real world costs of this ultimately speculative policy.
The bad news is that the enormous skew in asset prices engineered by the FOMC may hold significant credit risk in the future. The twin shocks of rising interest rates and widening credit spreads could significantly increase loss rates across the board in US and global banks. Looking at the body language of US regulators, it is hard to avoid the conclusion that the supervisory community is bracing for a repricing of risk.