I once wrote a post on university endowment investing that I thought was one of my better ones, but drew little attention.  It helped to inform another piece I wrote that was better received, The Forever Fund.  Okay, two more if you are a glutton for this kind of stuff: Liquidity Management is the First Priority of Risk Management, and The First Priority of Risk Control. (Note: university endowments had a lousy year ending in June of 2009.  Things may be looking better now, but with interest rates so low, university endowments are even more reliant on outperformance of equities and other risky assets.)

The key idea is this: understand what you are trying to fund before you begin investing.  When will the money be needed?  How much?  How realistic is the implied rate of return?  What if everyone with needs like yours tried to do this?  Would it work then?  Is the demand for investments that are optimal for entities with your liability structure greater than the available investments to be had?  Do you have some sustainable competitive advantage that few others have?

When I look at ideas like pension plans employing leverage (also here), I think they don’t know what they are doing.  Anybody remember how New Jersey decided to sell pension bonds and lever up their pension investments in risky assets?

• How the Garden State Dug a Hole
• Strapped Governments Revive Pension Bonds
• Christie Crosses N.J. Party Line Amid $34 Billion Pension Gap

That last article is timely, published today.  What began as borrowing $2.7 billion to plug a gap became a $34 billion gap.  Risky assets, particularly equities, did not perform.  Not only did they not earn enough to earn the actuarial rate needed to fund the defined benefit plan, they also had to pay interest on the pension bonds.

Trying to fill a funding gap via a more aggressive strategy is usually foolish.  If that were the best strategy, you should have been employing it already.

But consider the leverage angle more closely.  A defined benefit plan is by its nature a plan to pay out a stream of benefits over time to beneficiaries.   Typically they invest some of their assets in bonds that are shorter than the length of the stream of benefits they will have to pay.  Those bonds typically don’t earn enough to cover the actuarial funding rate, so they invest the rest in risky assets that they think that blended with the return on the bonds, will earn the actuarial funding rate or better.

There are at least three problems here:

• It would be ideal to invest entirely in super-safe debt instruments that match the expected liability cash flows, but that would require too much in taxes from the citizenry.
• But the moment that you move to funding some of the assets into stocks you open up two risks: 1) funding risk — what if the risky assets don’t perform to the degree needed? and, 2) Interest rate risk — the moment you are not matched there are risks if interest rates move against you.  This is usually a risk if rates move down.  It is rare for a defined benefit plan to buy enough long debt such that the value of bonds rises as much or more than the present value of the liabilities rise when interest rates fall.
• Pension bonds, or any sort of investments with internal leverage have the potential to increase funding risk, and they increase interest rate risk as well.  Pension bonds add another fixed claim to the existing semi-fixed claim of the benefit stream.

Are we the double-down society as far as investing goes?  It sure seems like it, and if many entities do this as a group the failure of the idea will be spectacular.  Risk premiums are not high now; take a look at Jeremy Grantham’s forecasts on page 4 of this PDF (which has many other useful bits that you can learn from).  Borrowing money to invest when risk premiums are small is playing the exact same game as we were doing with CDOs from 2005 to 2007.  If the spreads are thin, pile on more leverage!  That will get us to our earnings target.

It’s sad to see this phenomenon reappearing.  Don’t we ever learn?