Looking in the Rearview Mirror

LDI entered the lexicon of many people during the last couple of weeks, following the tradition of other acronyms in years past.  It made the jump from being a topic mostly discussed by pension plans (and the investment providers who sell things to them) to being the focus of market attention.

Forced selling, margin calls, fears of insolvency, and (eventually) government intervention will do that.

The mechanics of it all have been explained elsewhere, especially in the Financial Times.  It has published a great many helpful articles about the developments, which have been centered in the United Kingdom.

The backdrop

Let’s take a broader perspective — and a step back.  Alan Greenspan said this in a 2003 speech:

The use of a growing array of derivatives and the related application of more-sophisticated methods for measuring and managing risk are key factors underpinning the enhanced resilience of our largest financial intermediaries.  Derivatives have permitted financial risks to be unbundled in ways that have facilitated both their measurement and their management.  Because risks can be unbundled, individual financial instruments now can be analyzed in terms of their common underlying risk factors, and risks can be managed on a portfolio basis.  Concentrations of risk are more readily identified, and when such concentrations exceed the risk appetites of intermediaries, derivatives can be employed to transfer the underlying risks to other entities.

As a result, not only have individual financial institutions become less vulnerable to shocks from underlying risk factors, but also the financial system as a whole has become more resilient.  Individual institutions’ portfolios have become better diversified.  Furthermore, risk is more widely dispersed, both within the banking system and among other types of intermediaries and institutional investors.

This was during a period when Greenspan was unironically called “The Maestro.”  Five years later everything came apart and, in Congressional testimony, he said that he had “found a flaw” in his previous beliefs.  (Really, the market had.)  Representative Henry Waxman asked whether he “found that your view of the world, your ideology was not right.”  Greenspan’s reply included this:

That’s precisely the reason I was shocked, because I’ve been going for forty years or more, with very considerable evidence that it was working exceptionally well.

And that’s the point of this posting — the nature of evidence and belief — as befits our focus on investment practice in organizations.

Risk management

Most “risk management” follows the script as laid out by Greenspan:  Use historical evidence to formulate beliefs about how the world works and act accordingly.  That makes perfect sense much of the time and is a huge problem the rest of the time.

When you think about it, in spite of the amount of data we now have at our fingertips, we often have remarkably little real evidence from which to work.  Lots of what passes as evidence doesn’t go back far enough to be statistically reliable (even though it might be anecdotally helpful).  Most of the data we count on comes from the last four decades, which has arguably been all one economic regime.  Plus, the structure of today’s markets are vastly different anyway.

In most investment organizations, risk management tends to be almost exclusively backward-looking and statistical.  To repeat:  That works overall but — oh, boy — when it doesn’t, it really doesn’t.

“Stress tests” are popular, but they often represent five or ten landmark events — what would happen if we relived a past spasm.  What’s missing?  A qualitative dreaming of what might go bad in a big way because of the particular excesses in valuation, vehicle structuring, or economic (or geopolitical) events of the moment.  And not much attention is given to the extremes of the projected distribution, even though markets are creatures with fat tails.

Because history seems to do that rhyming thing, we can learn some lessons.  Along that line, Seth Klarman’s observations about the financial crisis are worth reading in full; here are a few excerpts:

Things that have never happened before are bound to occur with some regularity.  You must always be prepared for the unexpected, including sudden, sharp downward swings in markets and the economy.  Whatever adverse scenario you can contemplate, reality can be far worse.

Nowhere does it say that investors should strive to make every last dollar of potential profit; consideration of risk must never take a backseat to return.

Do not trust financial market risk models.  Reality is always too complex to be accurately modeled.  Attention to risk must be a 24/7/365 obsession, with people — not computers — assessing and reassessing the risk environment in real time.

We keep living through “teachable moments” like that of 2008-9, but do we really learn?

The issue of leverage crops up in a number of Klarman’s twenty lessons, and it led to the sticky wicket of LDI.  A risk that is thought to be managed by means of leverage can become unmanageable.

A Bloomberg editorial, “UK Pension Funds Shouldn’t Be This Exciting,” called the latest episode “a novel variation on a well-known theme:  Leverage meets unforeseen events.”  (In an indirect rebuttal to Greenspan’s idea that dispersing derivatives away from banks was a good thing, the editors also wrote:  “The more leverage held by nonbanks such as pension funds, hedge funds, and insurers, the greater the chance that dislocations will proliferate and threaten the broader system.”)

Inflation

The culprit this time around is a sharp increase in (apparently non-transitory) inflation.

It’s quite understandable that investment people thought a rise of this magnitude was impossible since — over two generations — it first declined in a methodical way and then went dormant.

But, while the odds against inflation rearing its ugly head again might have been judged to be vanishingly small, the impact that it would have if it ever did should have been judged to be significant.

Rising inflation begets rising interest rates, which means that the pricing of almost everything gets affected, mostly not in a good way.  Starting out with historically low rates and high multiples on risky assets, the potential impact of inflation has been substantial for years, even if it never arrived until now.

Unfortunately, when it does show up it might hang around for a time, as it did during the last cycle:

There were superb returns on stocks and bonds during the long disinflationary time charted above, but the starting valuations — very high interest rates and very low multiples — were the opposite of where they were when the current bout started.

Looking backward

As noted, understanding how markets have worked is important, but extrapolating that behavior forward — and tightly basing strategies upon it — ignores the reality of a complex adaptive system.

In When Genius Failed, Roger Lowenstein wrote that, at a critical moment, “as usual, the partners driving Long-Term had their eyes on the rearview mirror.”  With leverage.  Massive amounts of leverage, because the things they thought would mean revert had always done so before — and they had bet big and won before.

Given the demise of Long-Term Capital Management, other high-profile calamities, and the history-based modeling that drove the financial crisis, you might think that people would be cautious in their use of tools that attempt to quantify risk, but it seems that the opposite has happened.

From Wall Street to Main Street, investment ideas are judged by the numbers first and foremost.  Forecasts and policies are strongly influenced by historical record, and they are normally proposed and adopted without the qualitative context and caveats that ought to accompany them — and without much attention being paid to those fat tails.

In addition to asset allocation decisions, the affinity for statistical models drives investor demand for the very long list of products created and sold by the industry on that basis.  There is faith beyond reason in the methods.

Still dancing

Another thing that you might have thought would go away after the financial crisis was the need to keep dancing as long as the music was playing.  But that is thoroughly baked into the investment culture.

We saw that in spades a year ago, as asset managers kept throwing money at the huge winners of the post-pandemic period — and investors kept throwing money at those same asset managers, who also had been huge winners.  Woe be to you if you gave someone money to allocate in September of 2021, since it probably was put to work all at once, in stuff bunched toward the top of the performance charts.  You got the market move down and then some, as the music stopped not long after you started dancing.

To requote Klarman, “Nowhere does it say that investors should strive to make every last dollar of potential profit,” but up and down the investment chain that’s what’s most common.  True risk management is hard to find, because everything works against countercyclical actions due to the fear of walking away too soon or of being accused of unwise market timing.

Where we are

LDI is spawning lots of media attention.  (There were some early warnings, such as the July article in the Financial Times by Toby Nangle, and even a few postings about potential problems going back a decade on Risk.net and elsewhere.)  There is much speculation about where we go from here, specifically in regards to LDI, as well as the possible ripple effects from it and whether it is a harbinger of more trouble spots to come.

LDI has moved to the top of the agenda at the pension plans who have implemented it using leverage (including those who haven’t made the press yet), and will be a topic of conversation at this quarter’s review meetings for other asset owners as well.  Investment providers are busy filling the inboxes of their clients with analyses, predictions, and recommendations regarding a problem that they hadn’t otherwise highlighted.

One line of questioning for the debates to come is identified in another FT piece, titled “Liability-driven pension investing is still sound, says man who brought it to UK,” in which Dawid Konotey-Ahulu is quoted as saying:

What happened this week was the gilt market got hit by the equivalent of a category-four hurricane and the LDI system wasn’t built to withstand a weather system of that ferocity.

The immediate task ought to be to examine those strategies that will fail during certain kinds of market storms — and those that will do much worse than the base-case predictions due to an extended change in climate driven by higher interest rates (should they persist).

A recent Robeco report on five-year expected returns is called “The Age of Confusion.”  Interest rates have already blown through the end points found in most of the ten-year capital market assumptions published by consultants and asset management firms going into 2022, so the confusion is not surprising.  From the expectations of the asset-liability studies of pension plans to the financial plans prepared by investment advisors, things are off kilter in a big way.

A change in practice

Beyond the reactive firefighting, organizations need to reexamine and reorient their risk management practices.  The first step is identifying those areas in which statistical analyses and historical information are used as the foundation for high-impact decisions.

Models should be respected for what they can do but should not be relied upon for answers.  Therefore, the second step is the addition of regular and systematic qualitative risk management activities designed to trigger examinations of the potential impacts of out-of-norm events — and the maintenance of a threat matrix that isn’t static but that evolves as markets (and prices) change.

That seems obvious, and most investment professionals probably think that they do that in their heads as a part of their normal decision making.  But for the most part they don’t.  They are too busy dancing.

The leaders of organizations need to ensure that risk management gets taken to a new level, by communicating the need for change and building a culture that identifies and prepares for emerging risks before there is panic on the dance floor.