Numbers can lie Good today, but how about tomorrow?

By Andreas Von Bubnoff
September 17, 2007 in print edition F-1

Sagittarians are 38% more likely to break a leg than people of other star signs – and Leos are 15% more likely to suffer from internal bleeding. So says a 2006 Canadian study that looked at the reasons residents of Ontario province had unplanned stays in the hospital.

Leos, Sagittarians: There’s no need to worry. Even the study’s authors don’t believe their results.

They’re illustrating a point – that a scientific approach used in many human studies often leads to findings that are flat-out wrong.

Such studies make headlines every day, and often, as the public knows too well, they contradict each other. One week we may hear that pets are good for your health, the next week that they aren’t. One month, cellphone use causes brain cancer; the next month, it doesn’t.

“It’s the cure of the week or the killer of the week, the danger of the week,” says Dr. Barry Kramer, associate director for disease prevention at the National Institutes of Health in Bethesda, Md. It’s like treating people to an endless regimen of whiplash, he says.

Take the case of just one item: coffee. Drinking two or three cups per day can triple the risk of pancreatic cancer, according to a 1981 study. Not so, concluded a larger follow-up study published in 2001.

Coffee reduces the risk of colorectal cancer, found a 1998 study. Not so, according to one published later, in 2005.

“I’ve seen so many contradictory studies with coffee that I’ve come to ignore them all,” says Donald Berry, chair of the department of biostatistics at the University of Texas MDAnderson Cancer Center in Houston.

“What about the man on the street?” asks Stan Young, a statistician at the National Institute of Statistical Sciences in Research Triangle Park, N.C. “He reads about coffee causing and not causing cancer – so many contradictory findings he begins to think, ‘I don’t trust anything these scientists are saying.’ ”

These critics say the reason this keeps happening is simple: Far too many of these epidemiological studies – in which the habits and other factors of large populations of people are tracked, sometimes for years – are wrong and should be ignored.

In fact, some of these critics say, more than half of all epidemiological studies are incorrect.

The studies can be influential. Often, in response to them, members of the public will go out and dose themselves with this vitamin or that foodstuff.

And the studies also influence medical practice – doctors, the critics note, encouraged women to take hormones after menopause long before their effects were tested in randomized clinical trials, the gold standard of medical research.

Some of epidemiology’s critics are calling for stricter standards before such studies get reported in medical journals or in the popular press.

Young, one of the foremost critics, argues that epidemiological studies are so often wrong that they are coming close to being worthless. “We spend a lot of money and we could make claims just as valid as a random number generator,” he says.

Epidemiology’s defenders say such criticisms are hugely overblown.

They are “quite simplistic and exaggerated,” says Dr. Meir Stampfer, a professor of epidemiology and nutrition at the Harvard School of Public Health and a professor of medicine at Harvard Medical School.

What’s more, some things simply cannot be tested in randomized clinical trials. In certain cases, to do so would be unethical. (Care to assign half the people in a trial to smoke cigarettes?)

In other cases, a trial of adequate size and duration – say, to test whether coffee drinking raises or lowers the risk of Parkinson’s disease – would have to control the habits of huge numbers of people for decades. That would not only be hugely expensive but also virtually impossible.

Stampfer cites examples of findings of epidemiology that, he says, have stood the test of time: smoking’s link to lung cancer, to name the most notable.

–

Watching for clues

In epidemiological studies (also called observational studies), scientists observe what’s going on – they don’t try to change it. From what they observe, they reach conclusions – for example, about the risk of developing a certain disease from being exposed to something in the environment, lifestyle or a health intervention.

There are different ways to do this. Cohort studies follow a healthy group of people (with different intakes of, say, coffee) over time and look at who gets a disease. They’re considered the strongest type of epidemiological study.

Case-control or retrospective studies examine people with and without a certain disease and compare their prior life – for how much coffee they drank, for example – and see if people who got the disease drank more coffee in their past than those who didn’t.

Cross-sectional studies compare people’s present lifestyle (how much coffee they drink now) with their present health status.

Epidemiological studies have several advantages: They are relatively inexpensive, and they can ethically be done for exposures to factors such as alcohol that are considered harmful, because the people under study chose their exposure themselves.

But epidemiological studies have their minuses too, some of which are very well known. Suppose a study finds that coffee drinkers are more likely to get a certain disease. That doesn’t mean coffee caused the disease. Other, perhaps unknown, factors (called “confounders” in the trade) that are unrelated to the coffee may cause it – and if coffee drinkers are more likely to do this other thing, coffee may appear, incorrectly, to be the smoking gun.

A much clearer picture of the role of coffee on disease could be found, in theory, via a randomized clinical trial. One would divide a population into two, put one group on coffee and the other not, then follow both groups for years or decades to see which group got certain diseases and which didn’t.

The problem, however, is that such a study is very expensive and takes a long time, and it can be difficult to control people’s lives for that length of time.

Despite their shortcomings, epidemiological studies are often taken seriously, so much so that they can change medical practice. Such was the case after dozens of epidemiological studies, including one large, frequently cited one that came out of Harvard in 1991, had shown that taking estrogen after menopause reduces the risk of women getting cardiovascular disease.

“There was such a belief,” even with the medical community, that hormone replacement became part of standard medical practice, says Dr. Lisa Schwartz, associate professor of medicine at Dartmouth Medical School in Hanover, N.H., even in the face of an increased potential risk of breast cancer. In fact, some scientists and doctors said it would be unethical to do a randomized clinical trial to check if the hormone effect was real.

But in the hormone epidemiological studies, women choosing to take hormones may well have been healthier in other ways, Kramer says. And that fact – that they were healthier – could explain the lower risk of heart disease, not the hormones.

“To get hormone therapy, you have to go to a doctor and have to have insurance,” Kramer says. “That means you are in the upper strata of society.”

Eventually, a randomized clinical trial was conducted, as part of the so-called Women’s Health Initiative. Findings published in 2002 not only found no protection to the heart but actually reported some harm.

Epidemiology’s detractors say they have no trouble finding other cases than hormones where frequently cited and sometimes influential epidemiology studies have later turned out to be wrong or exaggerated.

In 1993, Harvard University scientists published two cohort studies reporting that vitamin E protected people from coronary heart disease. One, the Nurses Health Study, followed over 87,000 middle-aged female nurses without heart disease for up to eight years. It found that the 20% of nurses with the highest vitamin E intake had a 34% lower risk of major coronary disease than those with the lowest fifth of intake.

The other study followed almost 40,000 male health professionals without heart disease for four years – and found a 36% lower risk of coronary disease in those men taking more than 60 IU of vitamin E per day compared with those consuming less than 7.5 IU.

In the three years after these studies appeared, each was cited by other research papers more than 400 times, according to John Ioannidis, an epidemiologist at the University of Ioannina School of Medicine in Ioannina, Greece. Vitamin E therapy for heart patients became widespread – a 1997 survey published in the American Journal of Cardiology reported that 44% of U.S. cardiologists reported routine use of antioxidants, primarily vitamin E.

The therapy was finally put to the test in a Canadian randomized clinical trial of about 2,500 women and 7,000 men aged 55 years or older who were at high risk for cardiovascular events.

The findings – reported in 2000 – showed that an average daily dose of 400 IU vitamin E from natural sources for about 4 1/2 years had no effect on cardiovascular disease.

Yet, Schwartz says, seven years after that finding, her patients continue to take vitamin E in the belief that it will protect their hearts. “I am still taking people off of vitamin E,” she says of her patients, some of whom have heart disease.

–

Study of studies

In a provocative 2005 paper, Ioannidis examined the six most frequently cited epidemiological studies published from three major clinical journals between 1990 and 2003. He found that four of the six findings were later overturned by clinical trials.

Vitamin E didn’t protect the heart in men or women. Hormone therapy didn’t protect the heart in women. Nitric oxide inhalation didn’t help patients with respiratory distress syndrome.

Another finding turned out later to be exaggerated: Taking flavonoids reduces coronary artery disease risk only by 20%, not by 68% as originally reported.

The only finding of the six that stood the test of time was a small study that reported that a chemical called all-trans retinoic acid was effective in treating acute promyelocytic leukemia.

The studies that overturned each of these epidemiological findings, Ioannidis says, “caused major waves of surprise when they first appeared, because everybody had believed the observational studies. And then the randomized trials found something completely different.”

To be fair, Ioannidis also tested whether the most frequently cited randomized studies held up. He found that these had a much better track record. Only nine of 39 oft-cited ones were later contradicted or turned out to be exaggerated when other randomized studies were done.

True, Ioannidis looked at only six studies. But Young says he sees the same trend in his own informal counts of epidemiological claims. “When, in multiple papers, 15 out of 16 claims don’t replicate, there is a problem,” he says.

Belief can be costly, Young adds. For example, one part of the large, randomized Women’s Heath Initiative study tested the widely held belief – based in large part on epidemiological studies – that a low-fat diet decreases the risk of colorectal cancer, heart disease, or stroke.

The findings suggested that there was no effect. “$415 million later, none of the claims were supported,” Young says.

Other scientists, while more cautious than epidemiology’s most outspoken detractors, agree that there are many flawed studies. When Kramer first saw Ioannidis’ number, “I said to myself, ‘It can’t be that bad,’ ” he says. “But I can’t prove that it isn’t. I know there are a lot of bad studies out there.”

Ioannidis, Kramer says, is voicing what many know to be true.

–

Method in doubt

Why does this happen?

Young believes there’s something fundamentally wrong with the method of observational studies – something that goes way beyond that thorny little issue of confounding factors. It’s about another habit of epidemiology some call data-mining.

Most epidemiological studies, according to Young, don’t account for the fact that they often check many different things in one study. “They think it is fine to ask many questions of the same data set,” Young says. And the more things you check, the more likely it becomes that you’ll find something that’s statistically significant – just by chance, luck, nothing more.

It’s like rolling a pair of dice many times without anyone looking until you get snake eyes and then claiming you’d only rolled it once. Often, epidemiological researchers ask dozens, maybe hundreds of questions in the questionnaires they send to the people they study. They ask so many questions that something, eventually, is bound to come out positive.

That’s where the Canadian star sign study comes into play, Young says. It was only because the authors deliberately asked a lot of questions – to prove a point – that it was able to come up with significant results for something that couldn’t be true.

The study’s lead author, statistician Peter Austin of the Institute for Clinical Evaluative Sciences in Toronto, says that once he cleaned up his methodology (by adding a statistical correction for the large number of questions he asked) the association between Leos and internal bleeding and Sagittarians and leg-breaks disappeared.

–

On the defensive

Many epidemiologists do not agree with the critics’ assertion that most epidemiological studies are wrong and that randomized studies are more reliable.

“Randomized studies often contradict one another, as do observational studies,” says Harvard’s Stampfer, who is an author on both the frequently cited vitamin E and hormone replacement studies that Ioannidis says were later refuted.

Instead, Stampfer says, the two types of studies often test different things. “It’s not an issue here that observational studies got it wrong and randomized trials got it right,” he says, referring to the hormone replacement studies. “My view is that [both] were right and they were addressing different questions.”

For one thing, the randomized studies on hormone replacement and vitamin E that Ioannidis cited in his 2005 paper looked at different populations than the observational studies they refuted, says Stampfer, who takes vitamin E himself.

In the hormone replacement case, the observational studies looked at women around the age of menopause. The randomized trial looked at women who were mostly well past that age.

In fact, Stampfer says, a recent reanalysis of the Women’s Health initiative data suggested a trend that hormone therapy may be less risky for younger than older women. The effect was not statistically significant, but, Stampfer says, it’s further support for the idea that hormones have different effects depending on when women start taking them.

And the vitamin E studies? The 1993 observational studies followed people who didn’t have heart disease. The randomized study looked at people with known heart disease who were on many other medications. All those meds could easily override the effect of vitamin E, says Dr. Walter Willett, a professor of epidemiology and nutrition at the Harvard School of Public Health, who was a coauthor on the hormone and the vitamin E epidemiological studies.

And finally, the low-fat trial from the Women’s Health Initiative. It’s not surprising, Willett and Stampfer say, that this gold-standard trial failed to find what epidemiology had – that low-fat diets ward off heart disease, colorectal cancer and stroke. The women in these trials didn’t stick to their diets.

“The compliance with the low-fat diet was definitely far lower than anticipated,” Willett says, “and probably far worse than even acknowledged in the papers.”

Such arguments do not sway epidemiology’s detractors.

Each time a study doesn’t replicate, “they make a specific argument why the studies are different,” Young says. He concedes that epidemiology did uncover the truth about the risks of smoking – but only because the effects are so strong.”Even a blind hog occasionally finds an acorn,” he says.

Yet epidemiologists warn that discarding results because of a correction for multiple testing may risk missing true and important effects – especially in cases where there’s a good biological reason suggesting an effect, such as in studies of drugs that have been shown to work in animal experiments. And setting the bar too high might sometimes be dangerous, says Sander Greenland, a professor of epidemiology and statistics at UCLA. “Do you want to screen for medical side-effects with the attitude, ‘So what if we miss side effects?’ ” he asks. “That’s deadly. That’s ridiculous!”

The debate is unlikely to be resolved any time soon. “If you put five epidemiologists and five statisticians in a room and have this debate,” Young says, “and try to get each one to convince the other side, at the end of the day it will still be five to five.”

–

(BEGIN TEXT OF INFOBOX)

Study guide for research

It can be hard to make sense of the blizzard of studies on vitamins, diet, lifestyle and health risks that roll off the presses almost daily. It can be an even trickier call to decide whether to change one’s habits as a result of the latest findings. Here are a few tips from experts to help you assess the research.

* Replication. Don’t change your lifestyle just because of one study. The next one might show the exact opposite. You want to see studies replicated.

* Size of the effect. If a study reports that eating 10 rutabagas daily lowers your risk of ingrown toenails by 0.05%, that might not be a reason to go hog wild on rutabagas.

In the case of epidemiology studies, researchers suggest that you look for at least a doubling or a halving of an effect.

But also keep in mind that the absolute risk is important – doubling the very rare risk of being struck by lightning, for example, is not very significant.

* Give randomized, controlled trials more weight. Many experts consider these more reliable than observational studies.

* Statistical significance. The finding should be statistically significant for you to pay attention to it. A “trend,” although interesting, isn’t enough.

* Size of the study. Bigger is better. A study that only looks at 20 people is likely less reliable than one that includes 20,000 people.

* Length of follow-up. Generally, the longer time a study tracks people, the more reliable the results will be, and the more likely it will be to detect an effect.

* Consistent findings. The more precise the results within a study, the better. Take, for example, a study that finds that a treatment extends life 50 days. If the study’s 95% confidence interval (a statistical measure of precision) is a tight 45 to 55 days, it should be taken a lot more seriously than if the confidence interval is a loose zero to 100 days. In the latter case, the actual life extension could easily be zero days.

* Where was it published? Some researchers say that top journals are more likely to reject most unreliable studies. But beware: Such journals also tend to publish “surprising” studies that show something for the first time.

So, once again: Wait for other studies that show the same thing.

–Andreas von Bubnoff

By Jennifer Hughes in London and Nikki Tait in Brussels

Financial Times – 17/10/2008 – Asia Ed1 – 17

Growing pressure on the International Accounting Standards Board could lead Europe into developing its own regional rulemaker, in a move that would kill off the long-running effort to develop a single set of global accounting rules.

The IASB this week gave into pressure from the -European Union and eased its rules on fair value accounting, but an EU-wide meeting next week could see it presented with a “wish list” of further demands to which it could not accede.

Some fear that any -resulting stand-off between the two bodies could allow groups hostile to the IASB’s accounting rules to call for a European standard setter instead of the IASB.

The IASB sets accounting rules for more than 100 countries, including those under the EU umbrella.

There is thought to be -support among the French and Italian financial -communities for such a move, but it is opposed by some of the world’s biggest banks, including HSBC.

“It is absolutely essential that there is a single body of international accounting standard setters. It would be a considerable step backwards and most regrettable if we went to regional rules just as convergence on a single global set of standards is looking like a reality,” said Douglas Flint, chief financial officer of HSBC.

The wish list compiled at the EU meeting is likely to cover a series of issues linked to fair value accounting, the controversial process whereby financial institutions have to mark most of their assets at market prices. These have plunged as a result of the credit crunch, forcing many banks into losses and undermining the capital reported on their balance sheets.

The danger for the IASB is that the list will contain requests, such as rule changes, that it has already dismissed and cannot re-visit without damaging its credibility.

Banks are seeking further clarification over this week’s rule changes, which will allow some to shift holdings from fair value to amortised cost, a treatment that would smooth out market volatility. They are also likely to ask to be able similarly to shift assets containing embedded derivatives.

“Without trying to cause a problem the IASB may not be able to deliver what is asked because it would be severely criticised for both giving in and producing bad accounting standards,” warned Peter Wyman, a partner at PwC.

The European Banking Federation was emphatic that it would like further changes to be initiated by the international standard-setter. However, pressure could still be applied by the threat of an EU carve-out. However, the Commission can only remove parts of the rules, not add new clauses.

“We could then have a situation in Europe, in this most critical year-end period, where we would have in effect only a limited standard on financial instruments and an accounting free-for-all,” said Mr Wyman, who warned a carve out could make accounts far harder to compare by limiting the guidance available.

Hughes, Jennifer

16 October 2008

Financial Times – Asia Ed1 – 17

It will ease the strain on banks’ balance sheets, won’t cost the taxpayer a penny and is quick and relatively painless, apart from a handful of upset accountants.

The answer: relax fair value accounting, where companies have to mark most of their financial holdings to market prices.

Throughout the credit crunch this debate has grown, spilling over from academia into the banking world. In the current convulsions it has stepped up a gear and gone political.

Roughly, the debate lines up banks alongside politicians, particularly French and Italian, calling for some easing of fair value. On the other side are most regulators, accountants and investors. Gordon Brown, the UK prime minister, is also in this category.

Yesterday, the reformers got at least some of what they wanted, with the European Union poised to ratify changes made by the International Accounting Standards Board designed to even up its standards with US GAAP (generally accepted accounting principles) and in effect constituting an easing of its rules. Those who want the changes claim that fair value has undermined banks’ balance sheets by forcing them to report assets at current weak market prices even though the banks have no intention of selling them and in fact expect values to recover. Supporters of fair value claim that a balance sheet should simply reflect current economic reality, however nasty that is.

This week’s alterations will allow some more reclassification of assets from fair value to “held to maturity”, a category where gains are spread steadily over the lifetime of the asset and values are only written down if they are permanently impaired. But the banking lobby does not yet seem satisfied. Groups in both Europe and the US have smelled blood and are pushing for even more.

European industry lobbying for the changes made this week was fierce, with hints that politicians should end the IASB’s power in favour of some regional body.

In the US, the American Bankers Association called on the Securities and Exchange Commission to overrule the US accounting standards setter, claiming that the group “still refuses to recognise the realities of the current situation” and describing the current markets as “distressed”.

What we have here is a number of banks protesting because they do not like the answer. Both US and international accounting standards already include leeway for illiquid markets and allow for the use of other factors beside current market price.

This very issue was addressed by the IASB last month, when an expert advisory panel produced guidance on valuations in illiquid markets with reference to the credit crunch. However, the group, which included big banks and insurers alongside auditors and regulators, produced a definition of “distressed” or “firesale” which is so narrow as to almost never apply.

So, for “distressed” in banking lobby parlance read “depressed” markets and be wary of their misuse of technical terms. No-one can disagree the latter is true, but it doesn’t have the same get-out-of-jail-free status with the auditors as “distressed”.

The banking lobby is also confusing the role of accounts. These should simply be a true and fair record of management’s stewardship of the business. How the owners, regulators and tax authorities that read accounts choose to interpret them is their choice. Complaining about what the accounts show, when we’re talking about a system supported by such users of accounts as investors and regulators, is akin to blaming a torch for shining a light on the mess in your cupboard.

By Nikki Tait in Brussels and Jennifer Hughes in London

16 October 2008

Financial Times – Asia Ed1 – 15

Accounting rules blamed by some banks for exacerbating the financial turmoil look set to be eased across the European Union, bringing the 27-country bloc in line with changes agreed by international accounting rulemakers.

An EU regulators’ committee in Brussels voted unanimously in favour of accepting the emergency changes made by the International Accounting Standards Board on Monday. These will give banks more leeway in how they value certain assets whose prices have plunged.

Lawmakers in the European parliament then quickly endorsed the vote, while member states also gave their unanimous support. This means the changes, which are optional, can now apply to calculations of banks’ third-quarter results if they wish.

Under the rule change, financial institutions would be able to “reclassify” certain instruments. This means they can move them from their trading books, where they must be marked at “fair”, or current, market values, to their banking books. Here, they can be reported at amortised cost – so any further falls in market prices would not have to be reported, and any gains would be spread over the lifetime of the assets.

The IASB changes followed heavy pressure from European banks and politicians. European companies had complained that US rules gave their American rivals greater flexibility.

The issue was picked up by EU finance ministers last week, who urged international accounting standard boards to work together and “welcomed the readiness” of Brussels to take appropriate action “as soon as possible”. They demanded that the reclassification issue be solved by the end of the month.

However, there were fears discrepancies could arise between the approach taken by the London-based IASB and EU rulemakers, but that worry appeared to recede yesterday. The financial community, however, remains divided about the wisdom of easing up on fair value accounting.

Those in favour include some banks and insurers who believe their balance sheets are being weakened unnecessarily. Those against include many regulators, auditors and some investors, who think that using market prices reflects current economic reality, however harsh that may be.

Mark-to-market accounting exacerbates the crisis

Don Stammer

The Australian 15/10/2008 – 5 – Wealth – 1

THE enduring crisis in financial markets has brought to everyday use words such as “sub-prime”, “toxic assets”, “securitisation” and “wholesale funding”.

“Mark-to-market accounting” will soon join this list. Mark-to-market accounting did not cause the credit crisis — which reflects excessive lending and the earlier neglect of risk — but it has exacerbated it. And changes to this accounting standard are likely to play a small role in easing the grip from tight credit.

In mark-to-market accounting, a wide range of assets (and some liabilities) are valued at their current market prices.

Where the market prices of loans or securities decline, the value of these assets in the accounts of financial institutions, managed funds and hedge funds has to be written down — reducing profits and shrinking their capital bases.

In normal times, mark-to-market accounting operates reasonably well. During a financial crisis, however, the market prices for some loans and securities — even those with little risk on their interest payments or repayments of principal — can plunge because of a few transactions in an illiquid market.

In turn, the value of quality securities, including corporate debt or mortgage securities that carry AAA-ratings, has to be written down in the balance sheets of the financial institutions and funds that hold them — even though there is little risk of loss when the loans or securities mature. The Bank for International Settlements — often called the central bank for central banks — has suggested that the existing accounting standard could be overstating the expected losses on AAA-rated sub-prime mortgage securities by as more than 50 per cent.

The Bridgewater group, a US fund manager, warns that “ending the credit crisis will be highly unlikely without some type of accounting accommodation”.

“Because mark-to-market accounting on existing assets threatens bank capital today, it increases solvency concerns today, which raises funding costs and accelerates the need to sell assets today, this depresses the prices of those assets, which threatens capital and raises funding costs.”

As part of its recent package of measures to address the credit crisis, the US Congress has insisted on an early review of mark-to-market accounting.

Of course, the main action to alleviate the credit crisis will still have to come from the huge rescue package announced recently in the US, from further moves by European authorities to support strained banks, and from other central banks making bold cuts to their interest rates — as the Reserve Bank did last week.

But a change in accounting standards that avoided the shortcomings of mark-to-market but that still preserved the integrity of financial accounts would present a less magnified dimension to the woes facing investment markets — and make some contribution towards reviving credit flows, which are disastrously constrained.

An understanding of how the accounting standards require loans and securities to be valued in the balance sheets of banks and other financiers also helps in any assessment of how much, in the end, the recently announced bailout of US banks and other lenders will cost the US taxpayer.

The US Treasury, which can borrow at interest rates of 2 per cent on medium-dated securities, is likely to be able to buy distressed assets at prices above the levels at which they are currently recorded on the books of banks and other financial institutions and earn a reasonable return on them — and even show a small profit — as the loans mature.

Nearer to home, there’s an effect from mark-to-market accounting that applies to many Australian investors. A number of managed funds that invest in corporate bonds and other interest-bearing securities, and which are soundly managed, have had to reduce their unit prices for two reasons.

One is that the market prices have dropped for some of their quality, but illiquid, investments, particularly when the funds were selling investments to finance withdrawals.

Also, market prices declined on their holdings of quality non-government securities as interest rates on these securities rose relative to those paid on government securities.

Bonds and mortgages on which there is no default must eventually rise back to their pre-determined values.

As Warren Bird of Colonial First State Global Asset Management likes to put it, a bond that falls in value from say 100 cents in the dollar to 95c hasn’t permanently lost 5c; anyone who holds, or invests in, that bond after it has fallen in price to 95c will eventually see that 5c gain as part of their total return. (In technical language, they will receive the coupon interest payments as scheduled plus the amortisation of the bond to its termination value.)

Investors can expect a recovery in the return on managed funds that hold quality corporate bonds and mortgage-backed securities, as these assets close in on their maturity dates. Mark-to-market accounting might at times be viewed as arcane, but it’s important in several ways that matter.

Dr Stammer chairs Praemium Limited and the Investment Committee of INGIM’s Portfolio Solution Group. The views are his alone.

Entrada do governo acaba com era do deslumbramento em Wall Street

Valor Econômico – 15/10/2008

O histórico investimento do governo americano em Wall Street é um sinal do declínio das finanças modernas, e promete menos risco, contracheques mais magros e o enterro da audaz auto-imagem de Wall Street. “É uma mudança revolucionária – participação do governo, supervisão estatal. Eu não acho que as pessoas realmente compreendem a magnitude disso”, disse Peter J. Solomon, ex-banqueiro que agora dirige seu próprio banco de investimentos. O plano do Departamento do Tesouro dos Estados Unidos para adquirir fatias dos principais bancos e garantir suas dívidas provavelmente significará “mais restrições sobre o risco e mais reserva de capital, e, portanto, menor rendimento no longo prazo”, disse num relatório a clientes o analista David Trone, do banco de investimentos Fox-Pitt, Kelton. Num universo cheio de motoristas particulares, entregas de sushi no meio da noite e ternos cortados sob medida, existe a sensação de que os excessos logo se tornarão coisa do passado.

E que isso já deve começar no fim do ano, quando a maioria dos bancos deve cortar pessoal. As mudanças são graves por causa da natureza desta crise. Em outras anteriores, como o colapso da bolha da internet depois de 2000, Wall Street teve poucas perdas, em parte porque as firmas não contavam com bilhões em ações de tecnologia em seus próprios balanços. A última debacle do mercado que ameaçou várias empresas importantes ocorreu em 1990, quando o colapso do mercado de “junk bonds”, ou títulos de alto risco, derrubou o banco de investimento Drexel Burnham Lambert Inc., que acabou precisando ser socorrido pelo First Boston Corp, hoje parte do Credit Suisse. Mike Holland, administrador de fortunas da Holland & Co., firma de investimentos de Nova York, compara o cenário atual com os anos 70, quando inflação crescente e juros altos sucederam o otimismo agressivo e a prosperidade dos anos 60. O rendimento das bolsas acabou indo por água abaixo. Na próxima década, disse Holand, veremos “menos casas nos Hamptons (balneário chique de Nova York), iates e festas na Sardenha”, reduzindo o “fascínio por Wall Street e pelo mundo dos bancos de investimentos”. Solomon, de 70 anos, começou a carreira em Wall Street na Lehman Brothers em 1960, quando a firma ocupava um acanhado edifício de 12 andares e o departamento inteiro de banco de investimentos ficava num único andar. “Comprávamos nossos próprios almoços e andávamos de metrô”, diz ele. Mas tudo mudou nas décadas seguintes, quando o setor financeiro conquistou uma fatia maior da economia americana. O crescimento estava ligado, em parte, às montanhas de capital emprestado que Wall Street podia usar em suas operações e para oferecer crédito. Os lucros explodiram, assim como os salários e a importância cultural. Em 2006, 37% dos formandos da Harvard Business School rumaram para Wall Street, ante 26% em 2004. À medida que a economia desmorona e a economia desacelera, o sentimento muda. O canal de TV a cabo CNN tem anunciado um programa no horário nobre chamado “Dez Mais Procurados: Culpados pelo Colapso”, acusando ex-executivos do alto escalão da Lehman Brothers Holdings Inc. e da American International Group Inc., assim como Christopher Cox, o presidente da SEC, a comissão americana de valores mobiliários, e até o ex-senador republicano Phil Gramm, este último por conta de seu papel na liberalização de Wall Street, em décadas passadas. Solomon traça a gênese da crise, comparada por ele ao estouro de um rebanho de búfalos rumo a um despenhadeiro, à abertura de capital das maiores firmas de Wall Street. A Merrill Lynch abriu o capital em 1971, o Morgan Stanley em 1986 e a Goldman Sachs em 1999. “A partir daí é que se uniu responsabilidade limitada com acesso ilimitado ao capital, a partir daquele momento é que Wall Street começou a correr até cair de cara no chão”, disse Solomon. Agora os contribuintes, como investidores no negócio, não tolerarão mais tanto risco, diz ele. “A compra e venda de ações, que corresponde a uma boa fatia de nosso negócio, não vai mais se recuperar e isso prejudicará imensamente as margens”, disse um banqueiro do alto escalão de uma das firmas que receberá investimento do governo. E, quando o potencial de lucro se esvanece, os bancos são forçados a entrar em acordos de fusão, dizem analistas, reduzindo mais ainda o número de vagas e os salários.

15/10/2008

Valor Econômico

O novo plano de socorro a bancos apresentado ontem pelos EUA transformará o Tesouro americano em importante acionista de algumas das maiores instituições financeiras do país, um movimento agressivo cujo objetivo é restaurar a confiança de investidores e reativar o mercado global de crédito. O Tesouro investirá US$ 250 bilhões para injetar capital diretamente nos bancos em troca de participações acionárias. Metade irá para oito grandes bancos que aceitaram ser os primeiros a entrar no programa. O restante irá para os que se candidatarem nas próximas semanas. Milhares de bancos menores, cooperativas de crédito e outras instituições financeiras poderão participar. Quatro bancos receberão US$ 25 bilhões cada: Citigroup, JPMorgan Chase, Bank of America e Wells Fargo. Dois bancos de investimento que estão se convertendo ao varejo, Goldman Sachs e Morgan Stanley, terão US$ 10 bilhões cada. O Bank of New York Mellon terá US$ 3 bilhões e o State Street, US$ 2 bilhões.

É possível que o plano transforme o Tesouro no maior acionista individual dessas instituições, dependendo das condições em que os aumentos de capital forem feitos e da eventual participação dos atuais acionistas e de outros investidores nas operações. Ontem, US$ 25 bilhões representavam um quarto do valor de mercado do Citigroup. O plano reproduz vários elementos dos pacotes de salvamento lançados nos últimos dias na Europa e que os EUA relutaram em seguir. “Lamentamos ter que tomar essas ações”, afirmou o secretário do Tesouro, Henry Paulson, ao anunciar o plano. “Mas é o que precisamos fazer para restaurar a confiança no nosso sistema financeiro.” Os US$ 250 bilhões equivalem a mais da metade de tudo o que bancos americanos e europeus conseguiram levantar de capital novo desde o começo de 2007, quando começaram a reconhecer os enormes prejuízos gerados pelo estouro da bolha do mercado imobiliário americano. Mesmo assim, o plano oferece apenas uma fração do volume de recursos que muitos analistas acreditam que ainda será necessário para reforçar as reservas de capital dos bancos e compensar as perdas ainda não contabilizadas. O FMI estima que os bancos vão precisar de US$ 675 bilhões nos próximos anos. Os US$ 250 bilhões fazem parte do pacote de US$ 700 bilhões que o Tesouro apresentou em setembro para salvar o sistema financeiro. A intenção original era usar o dinheiro para comprar ativos ruins que comprometeram a saúde dos bancos, ajudando-os a limpar os balanços. O Tesouro ainda não sabe como organizar os leilões para adquirir esses ativos. Levará semanas para acertar os detalhes. Nos últimos dias, o aprofundamento da crise e as medidas tomadas na Europa convenceram o governo americano a aceitar a idéia de que injetar capital nos bancos poderia ser uma resposta mais simples e eficaz. Teoricamente, os US$ 250 bilhões permitiriam alavancar US$ 2,5 trilhões em empréstimos em condições normais, num cálculo conservador. Mas ainda é cedo para saber o impacto que o apoio do governo terá sobre a oferta de crédito nas condições atuais, com os EUA e outras economias avançadas em recessão ou muito perto dela. Outra medida anunciada ontem pelos EUA ataca esse problema por outra frente. A Corporação Federal de Seguro de Depósitos (FDIC), agência governamental que protege os correntistas contra o risco de falência dos bancos, vai garantir também, temporariamente, dívidas dos bancos e depósitos mantidos pelas empresas em contas usadas para pagar salários e fornecedores. Essas garantias incluirão papéis que os bancos poderão emitir a partir desta semana, e até junho de 2009, para buscar financiamento no mercado. O objetivo principal do governo é fazer os bancos voltarem a emprestar entre si, o que deixou de ocorrer nas últimas semanas. A medida também ajuda o governo a ganhar tempo, evitando uma corrida aos bancos mais frágeis. Em junho, havia 117 bancos na lista de observação da FDIC, que garante depósitos de 8,4 mil instituições. Os bancos que aceitarem ter o governo como sócio dinheiro também terão de aceitar limites para a remuneração de executivos, uma exigência que o Congresso americano fez para tornar politicamente viável a aprovação do socorro aos bancos. Funcionários do Tesouro sugeriram aos bancos que o efeito prático desses limites será pouco significativo.

Drive for clearer accounting continues

Jennifer Hughes

13 October 2008

Financial Times

Surveys FNM1

19

Since pensions were first dragged onto company balance sheets in the UK almost a decade ago with large parts of the industry kicking and screaming, the hullaballoo has rarely died down.

That move, by the UK Accounting Standards Board caused a furore in part because of its timing. Although many did not in principle like the light shed on scheme funding, the most immediate problem was that it came into force as the dotcom bubble burst, meaning that tumbling stock markets savaged the value of pension holdings, making deficits look even larger.

Accounting rulemakers believe that by making the funding, and how it is managed, clearly visible on the balance sheet, investors and other company stakeholders will have a clearer picture of the risks and demands of pension schemes.

However, many in the industry have linked the accounting changes with the significant rise in the closure of defined benefit schemes to new entrants in recent years as companies feel they can no longer face the risk of these volatile liabilities skewing their balance sheets. In the UK, less than a sixth of DB schemes are now open to new entrants, down from half in 2003, shortly after the ASB’s rules were introduced.

The more recent introduction of similar rules in the US by the Financial Accounting Standards Board in 2006 has attracted less controversy, perhaps because the switch from DB to defined contribution plans was well under way. But talk of convergence between US and international accounting standards may heat up the debate in the US as tighter rules are put on the table for discussion.

This year, the UK’s ASB came back with another suggestion that could have a dramatic effect on reported funding levels; in a paper produced in January, the board suggested changing the rate at which future liabilities are discounted – a move that will make the liabilities look bigger.

The good news for scheme sponsors is that the ASB no longer sets their accounting rules; that is now down to the International Accounting Standards Board.

The bad news is that Sir David Tweedie, the man blamed by newspapers for “destroying” pensions when he led the ASB in introducing the previous changes, has moved from the ASB to head its international counterpart, which is considering altering its rules on the topic.

That pensions pose a volatility risk to balance sheets is not in doubt. Recent market gyrations are a case in point. According to Watson Wyatt, the actuarial consultants, the combined schemes of FTSE 100 companies had a deficit of £12bn (€15bn, $21bn) at the end of August, but by the middle of last month, that had swung to a £7bn surplus.

The reason was a surge in corporate bond yields – which links directly to the ASB’s latest proposals. The future liabilities of a scheme are currently discounted to their present value using a blend of AA-corporate bond yields, which are designed to reflect the returns expected on fund assets.

However, in the current crisis corporate borrowing costs have spiked sharply higher as investors have demanded higher returns for the greater risk they perceive. In spite of that rising risk, the bigger number has served to reduce scheme deficits, meaning it has worked in the opposite way to that intended, since expected fund returns have not changed and if anything given the current market, are under some threat.

What the ASB has proposed is using the much lower “risk-free” rate, usually the yield on long-term government bonds. The suggestions triggered a wave of protest at the extra burden this would place on schemes as their reported deficits leapt overnight.

The National Association of Pension Funds said the idea could double the liabilities reported by “young” pension schemes while Pension Capital Strategies, a consultancy, calculated the “risk-free” rate would have raised the combined deficit of FTSE 100 schemes from £8bn to £100bn.

Although the ASB no longer has the power to change the rules directly, it does have a high profile in the pensions accounting world and strong links with the IASB. The IASB’s own current review has focused on removing “corridor” accounting, which allows some smoothing of the effect of market moves. But in a second phase, it is likely to address the thorny issue of discount rates.

“Given the revolution that has already been launched by pensions and accounting standard setters in their drive to reflect economic reality, it is easy enough to see more changes coming,” says Dawid Konotey-Ahulu of Redington Partners, a pensions consultancy, who warns that if anything, the credit crunch will speed up the ongoing trend towards making full market volatility transparent for investors.

IASB watchers say there are criticisms of its current proposals, which include new methods for classifying schemes, but that it would be unfair to say the pensions world was speaking with one voice.

“There are grumblings about how accounting has forced the closure of DB schemes, and I’m sure there will be more, but the experts don’t actually speak with one voice on this, you get some widely divergent views,” says one accounting rulemaker, who staunchly defended the central premise of transparency in what must be a warning to anyone hoping to change accounting’s current path.

He adds: “It’s the job of accountancy to portray as fairly as possible the true economics of what is going on. DB pensions have always been expensive to provide and they’ve only become more so given longevity and other factors. Accounting has just made that cost more transparent.”

The Guardian

The world’s largest insurance company, AIG, spent $440,000 on a lavish corporate retreat at one of California’s top beachside resorts a few days after accepting an $85bn emergency loan from the US government to stave off bankruptcy.

Details of the week-long getaway enraged legislators at a congressional hearing yesterday where AIG’s former bosses were accused of spending taxpayers’ money on pedicures, golf games and cocktails.

Crippled by losses on financial insurance companies, AIG was bailed out by US taxpayers on September 17 to avert a collapse which risked causing further failures.

The House oversight committee, which is investigating the company’s problems, confronted AIG executives with an invoice from the St Regis resort in Monarch Beach, south of Los Angeles, detailing an eight-day company event which began five days after the rescue.

“Average Americans are suffering economically,” said Henry Waxman, chairman of the committee. “They are losing their jobs, their homes and their health insurance. Yet less than one week after the taxpayers rescued AIG, company executives could be found wining and dining at one of the most exclusive resorts in the nation.”

The bill shows that AIG spent $139,375 on rooms, $147,301 on “banquets”, $23,380 on spa treatments and $6,939 on golf at an eight-day company event which began on September 22.

“US taxpayers will be, in effect, paying for this,” said Elijah Cummings, another Democrat, who demanded to know who was responsible for the outlay. “I think that person ought to be fired.”

Rates for the 325 rooms at the resort are typically upwards of $500 a night and the travel guide Fodor’s gives the place a rave review, saying: “Exclusivity and indulgence carry the day here; you can even have someone unpack for you.”

AIG has defended the event, saying it was to entertain freelance insurance salesmen who sold life, health and accident policies for the group’s US subsidiary.

But former chief executive Robert Willumstad, who stood down as a condition of the government’s bail-out, conceded that the getaway “seems very inappropriate”.

“If I had been aware of it, I would have prevented it from happening,” he added.

Defending their conduct in AIG’s final days, Willumstad and his British-born predecessor, Martin Sullivan, blamed the company’s problems on accounting rules which required it to write off billions of dollars on mortgage-related securities.

Sullivan said these rules had “unintended consequences” in making AIG’s books look worse than they actually were against the backdrop of a “financial tsunami”.

Carolyn Maloney, a Democrat from New York, accused the two executives of blaming accountants for AIG’s difficulties when in fact they had been “wrecking a great company” by gambling on obscure derivatives. “I think you should apologise to the American people for your mismanagement,” she said.

“Looking back at my time as CEO, I don’t believe AIG could have done anything differently,” said Willumstad. “The market seizure was an unprecedented global catastrophe.”