Wednesday 26 September 2007

Raising standards at no cost

Standards are useful, but when they cost money then many, who would benefit from understanding them and using them, just don't bother.

If you have business in China you might like to know that just a few days ago we arranged for the Chinese Risk Management Standard which was in the more traditional script used in Hong Kong and Taiwan to be translated into simplified Chinese which is more easily accessible to those who live in mainland China. This translation was carried out in less than 2 days by the staff of Ximco Corporation in Shanghai and we are very grateful for their efforts.

The original Risk Management Standard was developed in 2002 by the Institute of Risk Management, the Association of Insurance and Risk Managers and the National Association for Risk Management in the Public Sector. It has been downloaded over 100,000 times in English and is available in French, German, Italian, Spanish, Portuguese, Polish, Dutch, Danish,Swedish, Russian, Arabic, Japanese and now in two forms of Chinese. Here is the link to the new simplified Chinese Risk Management Standard. It is available free of charge, as it should be.

Download the Simlified Chinese Risk Management Standard...

Saturday 22 September 2007

Unintended Consequences - Bah, Humbug!

The comment from Mervyn King that he was debarred from covert operations to rescue Northern Rock by EU rules, a comment which some European sources have said is a complete misreading of the EU legislations' intent, reminds me of what AIRMIC's Counsel said about the Treasury's implementation of the Insurance Mediation Directive - see my previous blog.

With the Insurance Mediation Directive the lawyers in the Treasury, aided by those in the FSA, were so keen to stick to their principles of regulation (sound familiar?) and the purity of their operation that they disregarded the spirit of the EU Directive and gold plated the regulation so that some groups, such as buyers of insurance, would have to be unnecessarily regulated.

Perhaps they did this because such an approach gives them more power ( this would be the way Japanese bureacrats would work), perhaps because then they can't be blamed as they've covered every eventuality, but perhaps most of all in the "nanny knows best" attitude that pervades the British Civil Service they did it because they are micromanagers and they fear that events will get out their control. Regardless of the FSA's commitment to risk based principles too many of their lawyers are promoting the opposite of a risk taking approach.

Of course in the case of Northern Rock events did get of control. There were "unintended consequence of recent legislation" said the Governor of the BoE which prevented him from acting as he would have liked to restore stability. This is a miserable excuse. The triumvirate of the Treasury, the BoE and the FSA should have been regularly conducting crisis management testing to ensure that any legal obstacles to their actions were identified and action taken before the event to remove them.

Any dependency model which starts with the objective of preventing a run on the banking system ( surely one of the top risks for the triumvirate) would have then searched out the obstacles and assigned weightings to them. The causes of the run are immaterial to the exercise when it comes to legislative limitations on actions. What I would like to know is " Were such models developed, were they tested and why weren't the obstacles which the Governor spoke about identified?".

It seems that the regulators who have preached business continuity planning and stress testing to their markets have been remiss when doing the same for their own operations.

Wednesday 19 September 2007

How to panic the Treasury

It is no surprise that any decision involving three parties in likely to be a dog's breakfast. And so it is in the case of Northern Rock where the triumvirate is the Treasury, the Bank of England and the Financial Services Authority.

My experience of dealing with the Treasury is that they can only be panicked into action if their preferred modus operandi of doing nothing is going to result in ignominy or ridicule.

Just after 9/11, when it was clear that the insurance market for aviation risks was going to be withdrawn at the end of seven days, the way we (AIRMIC) got the Treasury to come to the party was by leaking a letter to Brown from concerned airlines to the Financial Times who obliged by running it as their lead story on the Thursday. By the Friday evening Brown had announced that the Treasury would act as insurer of last resort, this gave the rest of the world the weekend to figure out what they were going to do before the Monday deadline for removal of coverage.

When the Insurance Mediation Directive from the EU was passed into law by the Treasury they did not devote enough legal resource to defining who they were anticipating the directive to cover, which meant that after they and the FSA had had a go, it turned out that risk managers who purchased insurance on behalf of their subsidiary companies would be considered intermediaries. When asked whether there was any public interest at stake or whether they had intended to regulate risk managers the answer from both bodies was "no". The only way out they advised was to provide a legal opinion as to why risk managers should not be regulated. After considerable expense it was only when AIRMIC got an opinion from counsel which castigated Treasury for their shoddy work of drafting the EU regulations into UK law that suddenly the FSA found an answer and saved everyone a lot of effort.

When I asked the Cabinet Office for a refund of the legal fees they declined.

Tuesday 18 September 2007

Unintended Consequences of names

Back in the early 1990s the Austrians were found to be adulterating their wines with diethylene glycol (anti-freeze) in order to make them sweeter. When this became public knowledge in Japan the consumers not only stopped buying Austrian wines ,they also gave Australian wines a miss too. If you cannot differentiate your brand sufficiently in the consumers' mind then you may lose out through no fault of your own.

There is a limit to this same name concern and somehow I don't think that Northern Foods or Hard Rock Cafes need to be as agitated at present as the Alliance and Leicester.

However a consequence of the Northern Rock bailout means that the chances of Brown calling an October General Election have almost dropped to zero and Alistair Darling may yet join Norman Lamont and James Callaghan as a Chancellor for whom events in the markets exposed the limitations of their power.

Thursday 13 September 2007

Man's laughter Manslaughter

Attending the DVS Symposion in Munich today and reading the comments in this morning's Financial Times that"when bombarded by complex information it is human nature to look for a short cut" I am reminded of Ezra Pound's observation that the German word for "to condense" is "Dichten" and for "poetry" it is "Dichtung".

Perhaps risk managers should flex their poetic imaginations and as an example of how to condense the risks associated with a sense of humour consider the upside and downside, the respect that is due to the power of an apostrophe as well as the complexities of English pronunciation in the title of this blog.

Tuesday 11 September 2007

We must do better than this

A recent study by PCW as reported by the Financial Times in their September 10th special report on Risk Management stated: ( my comments in italics)

"400 senior financial services executives of which almost a quarter (so in the range 95-98) increased their annual spending year on year on risk management by more than 25% year on year for 3 years ( which means they doubled it).

Many companies (not quantified, why not, when everything else is?) do not appear to think they are getting best value for money - (the insinuation appears to be that those who doubled their spending were not getting good value- but there is no data in the article to back this up).

Only half of the risk managers ( how many risk managers were there in the 400?) questioned said the function contributed substantially ( no quantification) more value than it did 3 years ago - and just 23 percent of executives in other disciplines ([400-RM]*23%) believed the same (perhaps these were the ones who doubled their spending on risk management).

Fewer than one in 10 ( no information as to which group, the 400, the non-risk managers or the risk managers is being recorded here) believed that risk management was very effective at enabling managers to make better decisions (does this include financial risk management, before or after the sub prime developments?)."

My gripe is that the writer didn't allow us to see a much wider and linked section of the data so we could make up our own minds as to their importance. Instead confusion was created by selecting from different groups, some with numbers and percentages and some not.

There is no denying though that if only 10% of any of the groups believe that risk management is very effective at enabling managers make better decisions then there is a need for clear and measured examples of how risk management has been effective. Paul Hopkin at AIRMIC is working with DNV on just such a project to quantify the benefits of enterprise risk management in companies with embedded ERM and it will be ready in the first quarter of 2008.

In the meantime you must make do with garbled statistics.

Monday 10 September 2007

Gordon Brown and the election risks

The risks facing a new Prime Minister as complex, none more so on when to take the risk of an election. The following article has been submitted from a long time observer of the political process in the UK:


In the last two months there has been speculation that Gordon Brown might call an early election. What are the calculations of the risks involved that might determine his choice? The problem for all Prime Ministers is that they alone ultimately make the decision to ask the Queen for a dissolution of Parliament. The convention is that the Queen will not refuse a dissolution, if the Prime Minister has a majority in the House of Commons. This makes the burden of deciding whether or not to call an election an extremely anxious one for Prime Ministers, since a mistake can terminate their political career, not to mention those of many of their associates. Being able to decide the date of the general election (rather than having to serve a fixed term as under most other democratic constitutions) is a supreme advantage but also a supreme curse.

Some of the factors that the Prime Minister must consider are as follows:

1) The state of the opinion polls. The national polls moved strongly in Labour’s favour in July and August, opening up at one stage an eight point lead over the Conservatives. However in the last two weeks the gap has narrowed again.. Most Prime Ministers require a long period ahead in the polls before they will call an election. Margaret Thatcher even before 1983 when she had an overwhelming lead was very reluctant to call an election. James Callaghan famously refused to call an election in October 1978, when Labour had recovered in the polls to be level-pegging in the polls, and suffered a major defeat six months later. The most cautionary tale for Prime Ministers, however, is still Harold Wilson who had a ten point lead in the polls in the spring of 1970, called an early election for June, which everyone expected him to win, but he lost.

Other factors are at work too. The national opinion polls are supplemented by private polls, focus groups, and constituency polls. Unless all these are pointing in the same direction, they are likely to make a Prime Minister more risk averse. There is considerable evidence that Labour does consistently better in national polls than in actual polls, and part of this is because in many of the target seats which the Conservatives have to win to deprive Labour of its majority, local polling show that they are ahead or the contest is very even. Since there are a large number of target seats in London and the South East, this increases the risk of calling a general election on the basis of national polls.



2) A second issue is finance. A British general election campaign is short (typically just over three weeks) but very expensive, and one of the concerns of the Prime Minister is that he has enough funds to fight it, otherwise he could hand a significant advantage to his opponents. Labour’s finances are improving, but have some way to go. This is another negative factor.

3) A third issue is the rationale for calling an early election. This can add to the risk if voters and the media judge that the reason given for calling an early election is insufficient or bogus. This can then rebound on the Prime Minister and his party, as it did with Edward Heath in 1974. Gordon Brown’s reason for calling an early election looks superficially a good one, namely that he would be seeking his own mandate as Prime Minister. But the fact that he has now been two months as Prime Minister and has not announced that he needs to seek a mandate might undermine him, because voters might think the only reason he was calling an early election was because he was being opportunistic (seeking to exploit the temporary disarray of the Conservatives) rather than principled.

4) A fourth issue is opportunity. Some commentators have suggested that irrespective of any of the above, Autumn 2007 represents the best chance for Brown to win an election at any stage in the next three years (an election has to be called by June 2010). This is based on the disarray of the Conservatives in July and August, the Brown bounce in the opinion polls, which stemmed from the perception that this was a new government with new policies. Everything that happens from now on it is suggested will be less favourable. The economy might well go into recession, and house prices fall. Real wages may stagnate, credit may become tighter, and there may have to be sharp squeeze on public spending. All these would damage the standing of the Government and its popularity, since economic competence remains one of the key determinants of voting choices. In addition Brown may not be able to deliver changes to the NHS and other public services which the public see any differently from the changes of the last ten years. The danger that Brown may soon be seen as running not a new government but a continuation of the Blair Government is very real, and that too could mean that October 2007 is as good as it gets for Brown and Labour. The problem for Gordon Brown is deciding whether these risks outweigh the other risks that the poll evidence is not giving an accurate picture of Labour strength in key marginals. If he could be sure that Labour would be defeated if it went to the polls now, he would of course not risk it. He does not want to be the shortest serving Prime Minister since George Canning. But he cannot be sure. It is a balance of risks, and if as most people expect he does not call an election this autumn, he might come to regret it. But most Prime Ministers usually calculate that it is better to be in office than out of it.

Monday 3 September 2007

"Over time markets will do extraordinary,even bizarre things. A single big mistake could wipe out a long string of successes. We therefore need someone genetically programmed to recognise and avoid serious risks, including those never before encountered. Certain perils that lurk in investment strategies cannot be spotted by use of the models commonly employed by today's financial institutions."

Warren Buffett at his last shareholders' meeting as reported inThe Business - London's first global business magazine.

I like the line " including those never before encountered.

Sunday 2 September 2007

Super Crunchers- How anything can be predicted

Weekend Financial Times Magazine carries an extract from a book by Ian Ayers, entitled "Super Crunchers: How anything can be predicted", published this month by John Murray.

Risk Managers should understand the developments in database decision-making which is what this book is about. One of my contacts in Seattle, a man who headed up Digital Equipment Coorporation's operations in Japan for many years, has sent me a presentation that estimates that by 2020 computers with the capacity of the human brain will cost less than $1000, so the computing power necessary for such a world of database decision making will be available to almost everyone on a median developed world income.

Good news is that the machines,,for the moment, will still need us. Ayers says "Humans are crucial not only in deciding what to test, but also in collecting and, at times, creating the data."