The current economic recovery looks long and difficult, and makes
providing for an ageing population an even greater challenge. Will
economies rebound to provide for the health care needs of retirees? At
a conference organised in November 2008 by the Sim Kee Boon Institute’s
Centre for Silver Security at the Singapore Management University, Todd
Groome, adviser in the monetary and capital markets department of the
International Monetary Fund (IMF), spoke on how the ongoing financial
crisis could shape retirement risk .
According to Groome, the current economic crisis has brought home a
number of lessons. One is that new products can hold unknown risks.
Another is that risk management may not have been up to the task since
many of the standard quantitative models and users of these models
underestimated the systematic nature of risks.
Among lessons learned are to “nip it in the bud”. Early detection
and cure will reduce spill-over effects. Also, diversify to avoid
counter-party risk. Everyone using the same risk management techniques
amplifies systematic risk.
Recent Analysis on Financial Crisis
There has been a substantial amount of analysis of the recent
economic crisis. Some examples of recent work include: Enhancing market
and institutional resilience (Financial Stability Forum); Credit risk
transfer (Working Group on Risk Assessment and Capital); Observations
on risk management practices during the recent market turbulence
(Senior Supervisors Group); Supervisory lessons from the sub-prime
mortgage crisis (Basel Committee on Bank Supervision); Study of market
best practices (International Institute of Finance), and; Risk
management practices including the identification of risk management
challenges and failures, lessons learned and policy considerations
(International Monetary Financial Committee (IMFC).
From this work have come a number of broad findings. “The epicentre
of the market crisis was sub-prime mortgages and structured credit
products,” stated Groome. With them came innovative financing, such as
asset backed security CDOs (collateralised deposit obligations). From
these arose more potent variations such as CDO-squares (baskets of
CDOs), and synthetic CDOs (CDOs combined with credit default swaps).
Second, risks were often under-estimated due in part to product
complexity and over-reliance on quantitative analysis, including by
rating agencies. Investors learned too late that many risk evaluations
were wrong. Data shows that CDOs were indeed different. Default rates
on Moody’s Baa bonds have been 2.2% since 1983. Default rates (since
1993) on Moody’s Baa CDOs have been 24%.
Third, mark-to-market accounting amplified price gapping
and volatility among these typically illiquid securities including, in
some cases, market losses which might not ultimately materialise as
credit losses. As Groome pointed out, “Taking write-downs in illiquid
markets will amplify the loss.” To some extent, this may have been
avoided in Asian markets which permit write-downs directly against
owners’ equity rather than income if management (and auditors) feel the
loss is not permanent. Even then, however, it is still a heavy weight
hanging over earnings.
Fourth, there has also been a substantial wealth effect with the
decline in home prices followed by the sagging stock market. The
mechanism by which wealth links to consumer spending is uncertain but
the direction is known: It is negative.
Fifth, it wasn’t true in every case that poor risk management was to
blame. In many cases, banks and other companies did their best based on
available data within sample periods. Often, however, economies
experienced an out-of-sample event when the economy moved into
uncharted territory. “It makes the risk manager’s job very difficult,”
Sixth, bad risk management played a role. According to Groome,
“There may have been over-reliance on models and, in many cases, too
similar risk management and mitigation strategies.” While the models
may have been accurate, the fact that everyone used them meant all
users came to the same conclusion at the same time, thereby increasing
systematic risk. It is similar to technical market analysis. If
everyone agrees that three ‘up’ days conveys a buy signal, all will
rush to purchase at the same time, creating a self-fulfilling prophecy.
Risk Management Failures
According to Groome, “One failure was properly recognising
counter-party risk. An example is the monoline insurers.” He noted
that, “The likelihood of a systematic credit event -- or more likely
the volatility of the exposure being insured -- overwhelming the
resources of the monoline insurers was not anticipated.”
Second, many investors may have misunderstood the unique nature of
CDO ratings and how they differ from ratings of other debt. This may
have caused them to underestimate CDO risks. Some, but not all, of the
fault lies with the rating agencies. After all, S&P, Moody’s and
Fitch did not issue a warning, for example, that their CDO ratings were
not comparable to identical ratings for other bonds.
Third, there was also some misunderstanding about the nature of
credit ratings. Groome explained that, “While a security may be
AAA-rated, it does not imply anything about the stability of its market
pricing or liquidity. The failure to understand this resulted in
medium-term illiquid assets being funded, in some cases, with
short-term debt such as Asset-Backed Commercial Paper and Medium Term
Fourth, risk models are only as good as their inputs and
assumptions, and a model’s output is rarely “the answer”.
Unfortunately, underlying assumptions usually went unchallenged.
Sometimes, the extent to which the model depended on crucial
assumptions was not fully thought through. Similarly, regulators or
supervisors did not adequately challenge assumptions. Few asked
hard-hitting, informative questions. Those who did were seen as
“worrying about things that would probably never happen”, and were too
Lessons Learnt and Relearnt...So Far
As Groome observed, “The financial crisis offers lessons about how
to avoid contagion -- second and third round effects.” Among the
lessons he mentioned was that, “High credit ratings do not translate
into stable liquidity. Also, concentrated positions, such as with a
single insurer, should be a concern.”
Funding should also reflect basic banking and funding principles
such as better matching of asset and liability term structures,
particularly illiquid positions. In a downturn, problems arise when
reasonably long-term and illiquid assets are funded with short-term
debt which must be refinanced.
Looking forward, the loss in stock and real estate values caused a
large decline in the value of many pension funds. About half of the
losses in pension funds were in household plans. “In the future, the
greatest challenge may be longevity risk,” said Groome. All populations
are growing older, and many national pension plans are under-funded.
Several solutions may be pursued, he said, including drawing down
pensions at a later age or indexing pensions to longevity, saving more
and/or working longer. Investing in higher risk/return assets is a
debatable alternative since there is no assurance that, at retirement,
the market will be at the ‘up’ and not the ‘down’ part of the cycle.
Health care is probably the biggest risk management challenge since
it has the greatest volatility of outcomes and cost estimates. For many
countries, the solution has been to transfer the risk to government
which picks up much of the health care bill. “These costs are rising
with ageing populations, increased longevity and technology
improvements in the field of medical care, “Groome stressed. “They are
unlikely to be sustainable.”
The central question, as articulated by Federal Reserve Chairman Ben
Bernanke in a speech at Jackson Hole in August 2008, is “How to
strengthen our financial system … to reduce the frequency and severity
of bouts of financial instability in the future.” Bernanke gave an
interesting overview and possible roadmap to the “regulatory response”
and the roles of the central bank going forward.
Key Questions Remain
Groome highlighted several key questions and issues still pending.
With regard to Central Banks: They are a liquidity provider of last
resort, but liquidity for whom? It used to be for banks only, but has
permanently expanded to other financial institutions such as
broker-dealers or insurance companies. This may be further expanded to
include automobile companies and other manufacturers.
With regard to regulatory responses: Will the world build a new
infrastructure with a greater ability for firms to fail without
contagion? This may be the contribution of a central clearing house for
With regard to regulation vs market discipline: Will the future
include more rules, or more disclosure and transparency to support
better market discipline?
With regard to the influence of non-banks: It is increasing. It
includes insurance companies, pension funds, hedge funds, private
equity and sovereign wealth funds.
With regard to retirement savings: It has been a trend to push more
of the responsibility away from corporations and government and toward
individuals. The problem of government fiscal and financial stability
is seen in a study compiled by S&P in 2006. It shows sovereign
ratings could decline sharply due to ageing populations.
According to the S&P study, by 2030, sovereign ratings could
drop to non-investment grade for France, Italy, Japan, Korea and the
US. For others ratings do not fall as far, but still drop. For the US,
sovereign ratings could drop from AAA in 2005 to BBB in 2020, and
non-investment grade by 2030. By 2040, the sovereign ratings of all
countries, except Canada, drop to non-investment grade. (For Canada, it
drops only to AA from AAA.)
It dramatically shows the potential seriousness of the problem. There
does not yet appear to be action directed at an obvious solution.