Institutional memory

Ninety-year-old TIAA-Cref is one of the US’s oldest providers of retirement services and came into the financial crisis with experience of past crises galore. But did history help it when the crunch bit, and how will its business model learn from this experience? Laurie Carver reports

After a net loss of  $4.1 billion (£2.4 billion) over the 18 months to June 30, the Teachers Insurance and Annuities Association of America’s (TIAA) life insurance subsidiary, TIAA-Cref, saw its revered AAA debt rating from Moody’s stripped away at the beginning of October, while Standard & Poor’s put the company on watch for a downgrade. Moody’s saw the losses – much of them as a result of structured credit and real estate instruments that have dogged the financial sector over the crisis – as reducing “the cushion available to protect TIAA’s non-policyholder senior creditors”.

But the investment structure that has endured and evolved since its founding in 1918 was sufficient for Moody’s to keep its investment rating in the top bracket, reflecting confidence in its ability to honour its commitments to its policyholders. Its customers, primarily higher education teachers, have seen their account balances drop, but on scales much smaller than holders of the tax-advantaged 401(k) programmes popular in the US, and the rate paid out has not dropped below its guaranteed minimum 3% since 1948. Today the rate is around 3.65%.

TIAA-Cref holds its assets under management in an allocation close to the market portfolio considered the optimum allocation in economic theory. This entails holding quite a lot in alternatives – the $177 billion general account backing the TIAA life products has $69 billion in structured finance, $18 billion in commercial mortgages, $6 billion in real estate and around $7 billion in other investments such as private equity, agriculture and timberland. A $4.1 billion net loss in the 18 months to June 30 might be a disaster for another firm, but under the circumstances – and given the $374 billion assets the organisation has under management – it can seem like getting off lightly. How has the company managed to outrun its losses?

Source of liquidity
Towards the end of 2006, Brett Hammond, managing director and chief investment strategist at TIAA-Cref, read a story in the Wall Street Journal. An investor had bought a public toilet in London for £160,000, with the aim of converting it into apartments. Shortly after, the economists working on asset-backed securities (ABSs) in the company’s credit research team told him the residential mortgage markets were in the midst of an unsustainable bubble, and that the ABSs the company held were susceptible to a downturn.

They were particularly uncomfortable about the rapid expansion of US subprime loans, which federal mandates at the publicly backed Fannie Mae and Freddie Mac institutions had caused to proliferate ahead of affordability.

Hammond and his team made the decision to exit the market. “There were signs in the subprime world that the sector was untenable in the long term,” says Sanjeev Handa, head of global markets at TIAA-Cref. “We looked at some of the players in the game and saw that some of the products were dependent on a particular style of finance, and would not survive a liquidity crunch. So we redeployed our portfolio. We moved higher in the capital structure and bought stronger tranches.”

This process culminated in late 2006, before the summer 2007 collapse. It is now notorious that the products’ tranches were far from uniform. What was necessary was rooting out the bad loans from the good, the so-called granularity problem.

“We combed through our ABS portfolio to evaluate subprime exposure,” says Hammond. “As a result we had remarkably few writedowns, because we didn’t have that much to begin with and sold a lot of it. We concentrated on moving the portfolio up in credit quality to investments we thought could better weather the storm.”

It is believed TIAA-Cref suffered multi-billion dollar writedowns for the first half of 2009, but Hammond would not confirm this, or how much of the ABS portfolio was sold off. “Throughout the crisis we have been able to focus on yield-to-price ratios, not saying ‘we have to sell’.”

Whereas before the company had been buying at the lower end of the credit spectrum in the hunt for extra yield, it moved from an AA-B portfolio to an AAA. Given the battering of the credit rating agencies’ reputations since the crisis this would mean very little, were it not for the crucial point that these are not external ratings. The research team is, among other things, effectively an in-house rating agency – and its role was critical in the formation of the allocation strategy that has seen the company emerge relatively unscathed from the crisis.

The firm’s perceived financial strength has meant it has not suffered the customer exodus of some of its rivals. “We actually had positive inflows throughout the crisis. Our guaranteed product acted as a port in the storm, delivering a steady positive return and mitigating volatility. We didn’t need to have extra liquidity there – we are a provider rather than a demander of liquidity,” says Hammond.

The good, the bad and the ugly
The genesis of the research team lay, in fact, not in structured credit but in private debt issuance. For more than 90 years since it was founded by legendary New York tycoon Andrew Carnegie TIAA has been a preferred institutional investor for many corporates.

The need to estimate default and loss rates on privately issued bonds inspired the company to set up a credit research team, and by the mid-1990s this expanded into the 40-person team it is today.

“We had been a big player for decades in non-public issuance, which required us to do our own credit analysis, and that developed into the team today. We use rating agencies as one tool to evaluate credit securities, as one input into our own research process, and we assign our own rating,” says Handa.

He sees the department’s independence as key. “We don’t outsource our credit work to the outside world. We believe we can add value by doing things our own way.

“An important point is that it is separate from the portfolio management. This gives it independence, and encourages vigorous debate and discussion, which we believe leads to the best investment ideas.”

The work of the research team is the prism through which the portfolio managers can formulate the policy for the trading team. Research handles all aspects of credit risk, from default probability to loss distribution; understanding the mechanics of how a particular product works, whether it is private or public issuance; the underlying fundamental strength of the company; and more esoteric aspects such as the existence and make-up of any liquidity premium. The complex nature of this last point is one reason Hammond believes a top research team is essential.

“That is why you need credit research. You can take an index, but that contains bluechips like General Electric,” as well as more risky companies. “So it doesn’t provide an opportunity to do fundamental analysis to find the best opportunities.”

He sees a naive blanketing of backed-out index default and liquidity premiums from credit spreads as unhelpful in asset allocation. “Saying ‘x% default premium, y% liquidity premium’ is irrelevant at the index level, because we have so much variation between the good, the bad, and the ugly.

It’s relevant at the individual level.” The research team assigns default and liquidity premiums to individual companies, not to broad swaths.

But the main focus of the portfolio managers is more prosaic. “What we care about most are yield and default. What is something yielding and is it going to default? Because we haven’t had to seek out liquidity, through this crisis we have been able to focus on what something is yielding, and if it is satisfactory we can hold onto it.”

Hammond is guarded about the future of ABS, but doesn’t rule out a move back into the asset class. “We’re not going to buy into any old thing. But if there’s something at the higher end of the credit spectrum that gives us a good yield, we’ll look into it. It depends on where. Our research may tell us one thing when we’re looking at public bonds, and lead us in another direction when considering structured securities. The trade-off between risk and yield has remained clearer in the public bond market than the structured securities market.”

To rise out of the carnage wrought in the financial services industry, businesses will have to adapt their strategies and evolve their business models. Luckily for Hammond and his team, TIAA-Cref has some experience in this area.

Never let a good crisis go to waste
Shortly after President Barack Obama was elected last November, his chief of staff in waiting, former congressman Rahm Emanuel, famously said: “You don’t ever want a serious crisis to go to waste,” and it’s a maxim that TIAA-Cref has benefited from in the past, as the pioneer of the variable annuity (VA).

The product that today – thanks to spectacular implosions at companies such as Connecticut-based The Hartford – is a byword for incautious product design began as a response to an earlier financial crisis, the high inflation suffered by the US and others in the aftermath of WWII. In the late 1940s, as the money supply expanded to provide credit and benefits to returning veterans, inflation in the US spiked. The urban consumer price index hit 19.7% in 1947, and after a brief deflationary contraction rebounded back to 9.3% in 1952.

With many savings invested in the massive issuance of pre- and intra-war US treasury bonds, US retirement plans were in danger of a spiralling devaluation. The Treasury would not issue its first inflation-linked security until 1997, so the opportunities for explicit hedging were limited. The result was an exodus towards equity markets, and the stock market duly boomed. But how were providers to structure the benefits of retirement products in such a changed investment landscape?

The solution launched in 1952 was the first variable annuity, the College Retirement Equity Fund (the ‘Cref’ of the firm’s moniker), which coupled an equity-heavy account in the accumulation phase with an annuity in retirement whose rate fluctuated to reflect the fund performance and costs due to mortality rates, above a guaranteed floor rate. This basic, even quaint by today’s standards, design is still the norm at the company. “We focus on creating a built-in way for people to create a stream of income in retirement that they cannot outlive,” says Hammond.

Nowadays one thinks of VAs as synonymous with a far more complex guaranteed instrument, with guaranteed withdrawal benefits, ratchet-style return floors and so on. The complexity of such products, and their option-like risks, became a massive problem in the crisis to providers who neglected hedging programmes, such as Canadian insurer Manulife. The emphasis on simplicity shielded TIAA-Cref from this exposure.

“Our VAs are pure mark-to-market on the investment side and are really modelled after that original pioneering effort. We don’t have equity-linked performance guarantees, so we didn’t have the problem of unfunded guarantees being deep in-the-money that some in the industry did.”

The positive inflows into their products over the crisis could be seen as testimony to the endurance of the uncomplicated design, and Hammond is positive about its merits in comparison to its tax-advantaged competitors. “We believe there’s more to defined contribution schemes than 401(k)s, because they have no provision for retirement income.”

The simplicity of design also helps in another crucial area where VA providers have had to backtrack – cost (see Life & Pensions, April 2009, p. 20). “We don’t have sales commissions, and the annual combined administration fee is in the 50–60 basis points range, which is nearly half the industry average. We’re committed to annuities because people are getting their money’s worth in the technical sense of paying the present value of what you expect to get out.”

Adapt or die
So how will business at TIAA-Cref adapt following the most recent turmoil? Hammond sees a need for the industry to overhaul much of its practice, and rattles off a familiar list of intellectual casualties from economic theory to have come under attack: Markowitz portfolio theory; normally distributed returns; the diversification benefit. In particular, the classical view that an optimal allocation proportions assets to mimic their market share is much disputed.

“Markowitz-Sharpe tells us the optimum allocation is the market portfolio, but what is this? In the old days it would have been an equity index – you could invest in private equity 30 years ago, it’s just that nobody did.” As betas rose and correlations aligned in the market, investing in the market was outperformed by more traditional asset allocations.

“In this event, where the diversification benefits seem to have disappeared – traditional 60/40 equity/bond split portfolios outperformed the market.”

But TIAA-Cref has persisted with its diversified allocation, and Hammond says it may be even closer to the market portfolio than previously. “The reason to have a diversification effect is not necessarily to take care of short-term portfolio events, but rather for the long run. You need to be able to stomach an event like this every now and then.”

The crisis has brought into focus the importance of asset allocation appropriate to an investor’s needs, rather than elaborate hedging programmes that can be expensive. While TIAA-Cref does take both hedging and speculative derivatives positions (mainly in futures), the principal risk management methods focus on generating returns with a certain risk profile at a certain time.
“What we have been telling people is we do not even need to look at your asset allocation to tell you it’s changed. What people need to look into is rebalancing, which means buying into things that have just gone down. Much of the time that meant buying into equities,” which have, of course, boomed since March.

He sees the main lessons to be not just in including extreme scenarios – although that is a major point – but in setting asset allocation according to the investment and liquidity time horizons.

“You have to anticipate there are going to be events like this. If you are truly a long-term investor and can be sure you can stomach the interim risk, then getting close to the market is a great idea. If you can’t, you may want to stay closer to a traditional approach,” says Hammond.

“But it’s not just can you stomach risk or not, it’s what your time horizon is. If you are going to invest in the market portfolio, you may want to have a dedicated liquidity portfolio to match your short term liabilities, and absorb the interim risk before returns are realised. One of the big problems we’ve had over the crisis is that institutions were having to sell whatever they could sell, but what they would have preferred to sell had become illiquid. As equities remained liquid throughout the crisis, this exacerbated their fall, and led to even more volatility.”

He continues: “We’ve been in this business – taking care of people in higher education and non-profit organisations  since 1918. We’ve been through the roaring twenties and the depression. Post-WWI and post-WWII hyperinflation. The oil shocks and the Carter malaise in the seventies. The 1987 crash. We’ve been through the dotcom boom and the Asian crisis and we’ll get through this. We also know that these things are going to happen, and it’s our job to protect the investments of 3.5 million people.”