The First Priority of Risk Control

In early 1994, after the Fed’s first rate hike in what would be annus horribilis for the bond market, I was behind in my quota for selling Guaranteed Investment Contracts [GICs] at Provident Mutual.  Worse, my liabilities were running off faster than my assets.  My bosses gently nudged me about sales, and I told them that spreads were not justifying sales at present.  They let me be; they did not want to compromise underwriting in order to get sales.

But as my cash position got worse, and went negative, I eventually got a call from the Treasurer, one of the few people at Provident Mutual who could have worked at AIG and thrived.

“What are you doing?”

“What do you mean?”

“We are into the banks [DM -- borrowing money] because your GIC separate account is forcing us to borrow money!  What are you going to do about it?!”

“I’ll sell some GICs, and soon.”

“You better, or the CEO will hear of it.”

“You got it.”

After that, I reviewed my options, and looked for the cheapest way to raise cash.  The stable value industry at that time had yield illusion when the yield curve was steep.  GICs the would mature half in one year and half in five years were considered the equivalent of a three year GIC, even though the 1-5 had a forty basis point advantage in funding over the 3.

So, I called up a new client in Boston that I had been cultivating, and offered him a “special.”  A 1-5 GIC at 40 bps over the 3-yr GIC rate.

“That’s one special rate.”

“It is a special rate.”

“How much can you do?”

Thinking of the amount of debt I was in, I said “Six Million.”

“Sweeten it by another basis point, and we are done.”

I replied, “Done.”

My present problem solved, as the investment actuary, I looked at the situation, and with the mortgage market falling apart, and rates being forced up, I drew the conclusion that we were in a self-reinforcing situation where interest rates were going to rise.  I began to sell GICs with abandon, telling the investment department not to hedge my sales, and not to stretch for yield.  I sold my entire year’s quota in the next six months, and the investment department upgraded the credit quality of the portfolio as rates rose.  By the time Confederation announced its insolvency, I had almost finished my year.  Good thing, given that the Confederation insolvency would kill of my line of business two ways: 1) charges from the guaranty funds.  2) Provident Mutual’s credit rating was now too low to sell GICs.  I ended up closing the line of business two years later.

Now, as a note to risk managers, it is probably a bad idea to give control of hedging policy over to the line of business actuary, even though it worked out for the pension division of Provident Mutual.  We were special, not because of me, but because we stayed in close touch with the investment department, and hired actuaries that understood investments, which was rare at that time.

So, out of the horrible year 1994, we came out of it better off, while many were licking their wounds, and three of our competitors had blown up.  That wouldn’t keep the business from being eliminated two years later, but it did protect the interests of our dividend receiving policyholders.

The First Priority of Risk Control

This brings me to the main idea of this piece.  What is the first priority of risk control?  It is to make sure that the company/individual in question is never forced to take action at an adverse moment.  Consider all of the financial companies that are being forced to dilute common sharholders in order to survive.  Consider the universities that entered into illiquid investment programs with the promise of earning higher returns.  Many of them are choking on a lack of liquidity at present, as the fall in the markets has driven:

  • the endowment down
  • giving is down
  • willingness of parents to send children to expensive universities is down
  • what’s worse, illiquid investments have suffered worse than liquid investments.  Very tough to sell in the secondary markets — akin to consorting with guys who wear “panky rangs,” as they say down south.  (Loan sharks.)

The first priority of the risk manager is to make sure that there is never a call on cash (or any other resource) that he can’t meet on favorable terms.  It means running at a lower ROE, and having more surplus assets to cushion the company.  It also means doing stress-testing beyond what is imaginable to most of your peers.  My interest rate scenarios for cash flow testing went up 9%, and went all the way down to zero.  We made sure that we could survive.  If you can’t survive, you can’t play for the next round, no matter how good the last few rounds were.

David Swensen is a bright guy, as are many of his peers in similarly tough situations.  Yes, they enjoyed the boom as the leverage built up, but now what do you do when faced with demands for liquidity, and precious little liquid assets to deliver?  You estimate the duration of the crisis, and if it is longer than your liquid assets can finance, you sell some of your best illiquid assets now, and play for time.  If the duration is shorter, sell liquid assets.

Most life insurance companies that died in the 90s died from a liquidity mismatch.  Liquid liabilities, and illiquid assets.  Works great in a bull cycle, and lousy in a bear cycle.  That is true for any institution, though, so if you manage risk, be careful of the illiquidity of your assets.  Even owning a home that requires two incomes to pay the mortgage is not a risk worth taking.

Ask yourself, for your firm, and for your household, are there any conceivable situations where you will not be able to meet your obligations on favorable terms?  If not, you are placing those that you care for at risk.  Take action now to reduce that risk while you still can.

1 Comment

  • Steven Milos says:

    David,

    Talk about a backhanded compliment! I thought this was pretty funny: “one of the few people at Provident Mutual who could have worked at AIG and thrived.”

    Great piece,

    Steve