Exactly ten years ago today, on a cold, grey, autumnal day much like today, a small team of financial engineers pulled off an audacious transaction that changed the art of what, up until then, had been considered possible in the world of risk management for pension plans. I was one of those involved, and this is the story of how it happened.
By 2003, pension funds had begun to feel the effects of an esoteric accounting rule change. Introduced in 2001, FRS17 obliged corporations to recognise on their balance sheets, the impact of wildly volatile deficits in their defined benefit pension plans. This was a new development, brutal and unprecedented. Healthy, vibrant corporations were suddenly exposed to meaningful debt obligations they hadn’t signed up to. After all, how could they have known that running a defined benefit pension plan would result in the accumulation of obligations that would threaten the stability of the company itself?
By and large, ten years ago, the FRS17 problem was regarded as a temporary irritant. Growing pension deficits were caused largely by swollen liabilities which, in turn, were due to low interest rates and higher than anticipated inflation expectations. The liabilities were all inflation linked; higher inflation meant higher liabilities. There was, in those days, universal certainty that interest rates would rise and that inflation would fall. It was just a matter of time. Given enough bed rest, the patient would recover.
And yet. Our calculations demonstrated that there was a material risk that interest rates could fall a lot further and inflation could rise much more; and, if that happened, there would be a burgeoning of all pension plan liabilities, the like of which would be the stuff of nightmares.
So we girded our loins and headed out to spread the word. It was a three part sermon that went like this:
- It is a mistake to believe that this is as bad as it can get. Falling interest rates and rising inflation makes for a toxic combination for pension plans;
- You can do something about it. By insuring (also known as “hedging”) against the scenario outlined in part 1, you can protect the pension plan;
- If you don’t hedge, you should prepare for an extraordinarily volatile and difficult next ten years.
Some Finance Directors and some Boards of Trustees were polite enough to give us a hearing but, in 2003, only one actually took action: Friends Provident Pension Scheme.
In mid-2003 I had a meeting with Paul Cooper, visionary actuary at Friends Provident and one of my closest clients. For the first time that year, after 180 presentations, I had found someone who was genuinely intrigued by the notion that a pension scheme could reduce its exposure to dangerous market movements. What followed was an extraordinary adventure as actuary (Paul) and investment banker (me) embarked on a journey across uncharted territory.
The first time I presented to Paul Cooper I laid out a proposal to de-risk the entire £600 million pension scheme; there were two simple numbers we discussed for over an hour. Just two numbers. The first was the amount by which Friends Provident Pension Scheme’s liabilities would increase given a micro (0.01%) fall in interest rates. That number was just over £1 million. Imagine that for a second. Interest rates move around in the traded capital markets all the time; and the tiniest of tiny moves jacked up the liabilities by a million quid! What’s more, this implied that a one percent fall in interest rates would ratchet up the pensions bill by more than £100 million. No-one was predicting a one percent decline in long term interest rates, but that wasn’t the point. If it happened, the pension liabilities would rise by an insanely large amount.
The second number we discussed in meticulous detail was the amount by which the pension liabilities would increase for every micro (0.01%) rise in inflation expectations. This number was also circa £1 million. Which meant that a one percent rise in inflation expectations would drive up the pensions bill by just under £100 million. The nightmare scenario in my presentation was a world in which interest rates FELL by 1% AND inflation ROSE by 1%. The pension plan would take a combined hit of £200m!
But the chance of interest rates declining with inflation simultaneously rising was non-existent. Colonel Gaddafi had more chance of becoming the Pope. At least, that’s what Conventional Wisdom said. Conventional Wisdom back in the day reasoned that if inflation rose, the Bank of England would inevitably raise interest rates to bring it back down. It would be automatic, a sure thing. Guaranteed. Newtonian physics. A world in which interest rates could fall by 1% and inflation could simultaneously rise by 1% was the stuff of pure fantasy. This was one of the first things one learned at the LSE. To suggest otherwise, just showed you knew very little about Economics.
An hour into our discussion, Paul and I both agreed that if interest rates fell and inflation rose it would be an unmitigated disaster for the pension scheme. There were simply no equity-like assets that could offset that kind of damage. What we just could not agree on was how likely that scenario was to materialise. Colonel Gaddafi and the Vatican featured heavily in our discussion.
So we went back to the evidence.
We were on the sixth floor of the Merrill Lynch offices in King Edward Street overlooking Paternoster Square. The table was strewn with empty coffee cups. All the water bottles were empty; fizzy and still. “Persuade me”, Paul said. “You have my undivided attention.” (I had written an article for Finance Magazine saying that unhedged liabilities were the elephant in the room but, in order to be convinced, Paul demanded more evidence than I had yet presented).
I made three arguments:
Argument One: Over the previous 100 years, the real yield had been down to zero on several occasions. Indeed, the “normal” levels of 3 or 4% (and the floor of 2%) so readily quoted by Conventional Wisdom were the outliers, the exception, NOT the norm. We had experienced zero percent real yields in the past and we would likely see them again. I had 100 years of data and the 100 year graph.
Argument Two: The risk of materially lower real yields was a risk the pension scheme just could not afford to take. If Conventional Wisdom was wrong, the entire pension scheme would be in jeopardy. And, crucially, in 2003, taking out insurance (hedging) was relatively affordable. Why wouldn’t you??
Argument Three: A new era of transparency (post Enron / Tyco / Worldcom) had ushered in sweeping changes to global regulation and accounting rules. Pension schemes were suddenly forced to mark their liabilities to market and report them as a debt, squatting like a giant toad on the sponsoring corporate’s balance sheet. No exceptions.
It seemed reasonable to infer that this development would eventually and inevitably spur all companies to compel, at all costs, their defined benefit pension schemes to manage the volatility of the pension deficit. Which, in turn, would mean that demand for gilts and swaps (as hedging instruments) would soar. In the absence of massive increased gilt supply, the price of gilts would rise in the coming years, and would keep rising. And if (as seemed plausible, even probable) the government decided to take the same uncompromising “mark to market” approach to its vast unfunded public sector pension liabilities, that would have an incredible impact on the price of gilts (upwards) and their yield (downwards). That, in turn, would cause all defined benefit pension liabilities to sky rocket to infinity and beyond, since pension liabilities increase at warp speed as yields fall. If the Land of Mordor (aka Negative Real Yields) did materialise, then failure to hedge, I argued, would prove catastrophic. Maybe not for several years, but eventually it would prove fatal.
We went over it and over it. Again and again. Paul was wrestling with the fact that no-one else shared this view, no other pension scheme had done such a transaction and last, but not least, he would need to persuade an entire board of trustees, and the board of Friends Provident, of some seriously unconventional wisdom. It was a huge ask. We had been talking for the entire afternoon. Finally, he leant back in his chair. “OK. I get it, I buy it, I’m going to make it happen.”
And he did.
Paul Cooper is an intriguing guy. He’s softly spoken, completely bald, highly intelligent and he has an acute, dry sense of humour and a gift for getting to the point.
He doesn’t tire of interrogating the evidence and calmly persisting until, Columbo-like, he gets a satisfactory answer. He is also an actuary with years of experience dealing with banks, asset managers, advisers and trustees. Back in 2003, he was almost uniquely qualified to guide the pension scheme through the labyrinth of capital markets, derivatives and pension maths necessary to “make it happen”.
And so it was that at 11:00 on a freezing Tuesday morning – 2 December 2003 – Friends Provident Pension Scheme took decisive action. The scheme implemented a landmark hedging transaction fully protecting itself against falling interest rates and rising inflation. They were the first pension scheme to do so, and they set in train the multi-billion hedging transactions that have since become commonplace and are known generically as Liability Driven Investing or LDI. It was the culmination of months of discussion and planning. Even so, the fledgling inflation swaps market was extremely thin coming up to Christmas and this was a very long dated and large derivatives transaction. In fact, back then, it was the largest of its type to date.
I was working alongside my favourite trader, Jonathan Mitchell (Merrill Lynch’s Head of Inflation), Rob Gardner (now Co-CEO at Redington) and legendary ALM guru, Philip Rose. For us, it was like landing a 747 on the Hudson. For Friends Provident Pension Scheme, there was no precedent for a pensions-hedging transaction of this size, maturity, or type and, at the time, it flew directly in the face of conventional wisdom.
But, looking back, exactly ten years later to the day, the transaction proved far more effective than any of us involved with it would have believed possible. The 30 year real yield on 2 December 2003 was 2.13%; these days it is around 0%.
That utter collapse of the real yield (i.e. a sharp fall in interest rates AND an equally sharp rise in inflation expectations) should have cost the pension scheme around £213 million in increased liabilities; but it hasn’t. Why not? Because against all the odds, it hedged that very risk. Many much larger pension schemes still haven’t hedged. In some cases that failure to take decisive action over the last ten years has added billions to the value of the liabilities. Astonishingly, many pension plans still haven’t implemented a coherent risk management strategy. In the public sector, with one or two exceptions, the science of pensions risk management is still in its infancy.
So, ten years on, this is my tribute to Paul Cooper. These days, he works in the finance industry in Luxembourg. Make no mistake; his grasp of the situation ten years ago, and his skill in persuading all stakeholders to take action, is the reason one pension plan was fully insured against what turned out to be the worst storm conditions anyone could have imagined.