The International Association of Insurance Supervisors adopted the following in late 2011 as a part of ICP 8.
Archive for May 2012
The American Banker has a new column on risk management. The first article is here. Clifford Rossi manes some good points about the JP Morgan story. But Riskviews takes issue with one point that he makes…
The paradigm of the trader and the risk manager are fundamentally at odds. The trader will believe that if they are given the funds to make one more trade, they will make up all of the past losses and post a large gain. The stories of successful traders and hedge fund managers all read the same, losses, growing losses, no one else believes in the trader. Finally, they are vindicated by a large gain that makes them the hero. When you listen to the stories from Bear Sterns and Lehman, folks who were involved all say that it was just a liquidity issue. If they just had a little more funds, they would have made the trades that would have brought the firm back.
The risk manager on the other hand believes that there must be a limit to the amount that is put at risk by the firm. Do not bet what you cannot afford to lose. The risk manager believes that even the best theory can have a run of bad luck that the firm cannot afford.
Ultimately, the risk manager is not the moral compass of the firm. The risk manager is nothing more or less than the person who is charged to make sure that the CEO and the Board understand and are fully aware and approve of all of the risk taking activities of the firm. To make that process work, the risk manager will ask the board and CEO to pre-approve some activities and to require to be notified about others.
In JP Morgan’s case, the board and CEO should have been aware of what was going on, of the size of the positions. Perhaps they did not give clear directions to the risk manager or perhaps the risk manager for some reason failed to report the risk positions.
However, it should have been a business decision made by the Board and CEO, not a decision of the trader or of the risk manager. The loss that resulted would be a decision that did not work out as intended, not even necessarily a bad decision. All decisions do not work out well. And while $3 Billion is a large amount of money, it is only a fraction of earnings for a good year for JP Morgan.
If the decision to make the trade(s) that added up to the $3 Billion loss were made by the trader and not reported to the CEO and Board, then and only then is this a risk management failure.
“Truth be told, risk management is an ever-evolving discipline. The Great Recession pointed out both the shortcomings of implementation at many companies, as well as the potential for a strong risk culture driving the risk management process. As time passes from this crisis to the next (as there is always another one around the bend), recurring trends are becoming apparent and companies across the world are getting smarter about the essential need to move risk management from the back room to a position influencing strategic decisions.”
Successful Businesses pay attention to risk.
- How much risk to take compared to their capacity to absorb risk via their level of average earnings and their capital position. They have a basket. Each basket is different. It can easily hold so much. Sometimes, you decide to put a little more in the basket, sometimes a little less. They should know when they have stacked their risk far over the top of the basket.
- What kinds of risk to take. They have a plan for how much of each major class of risk they they will pick up to use up the capacity of their basket.
- Then when the actually go to fill the basket, they need to carefully choose each and every risk that they put into the basket.
- And as long as they have those risks in the basket, they need to pay attention and make sure that none of the risks are spoiling themselves and especially that they are not spoiling the entire basket of fruit or ruining the basket itself.
But that is not what a successful business is all about. They are not in business to be careful with their basket of risks. They are in business to make sure that their basket makes a profit.
+ So how much risk to take is informed by the level of profit to be had for risk in the marketplace. Some business managers do it backwards. If they are not being paid much for risk, they fill up the basket higher and higher. That is what many did just prior to the financial crisis. In insurance terms, they grew rapidly at the peak of the soft market. Just prior to the cirsis, risk margins for most financial market risks were at cyclical lows. What makes sense for a business that wants to get the best reward for the risk taken would be to take the most risk when the reward for risk is the highest. Few do that. However, the problem faced by firms whose primary business is risk taking is that taking less risk in times of low reward for risk creates even more pressure on their income because of decreased expense coverage. This problem seems to indicate that businesses in such cyclical markets should be very careful to manage their level of fixed expenses.
+ What types of risk to take is also informed very much by the margins. But it also needs to be informed by diversification principles. Short term thinking suggests that risk taking shift all to the particular risk with the immediate best risk adjusted margin. Long term thinking suggests something very different. Long term thinking realizes that the business needs to have alternatives. For most markets, the alternatives are only maintained if a presence in multiple risks is maintained in good times and bad. Risk and reward needs to develop a balance between short term and long term. To allow for exploiting particularly rich markets while maintaining discipline in other markets.
+ Which specific risks to select needs to incorporate a clear view of actual profitability. It is very easy on a spreadsheet to take your sales projection and profit projections and multiply both numbers by two. However, it is only through careful selection of individual risks that something even remotely like that simple minded projection can be achieved. The profit opportunity from each risk for the additional sales may be at the same rate as the original margins, it may be higher (unlikely) and it may well be lower. The risk reward system needs to be sensitive to all of these three possibilities and ready to react accordingly.
From Black: Swans and Crude by Liz Ann Saunders, her tips for investing in a sideways market:
- Be diversified, especially now that asset-class correlations have begun to recede toward normal levels.
- If you like to be opportunistic, keep some powder dry in highly liquid investments for both cash needs and some flexibility to take advantage of volatility.
- Consider more frequent rebalancing if volatility reasserts itself, allowing you to sell into strength and buy into weakness.
- Focus on your long-term goals and not short-term market dips so you’re less likely to fall prey to panic selling (or buying).
- Review your portfolio and asset allocation to confirm your risk tolerance matches your financial goals.
These suggestions line up well with the Pragmatist risk attitude of Plural Rationalities. That is good because the Pragmatists expect an Uncertain Environment, which is what we hear over and over that we are experiencing.
A Pragmatist will seek to diversify. Not only will they want to diversify their risks as Ms. Saunders suggests as her very first suggestion, but they will also be diversifying their approach to risks. Pragmatists will sometimes look to limit their losses with a Conservator style risk management approach, to aggressively pursue profits with a Maximizer style approach and even sometimes to look at risk vs reward in a Manager style approach.
Notice the interesting twist in her first point “now that asset-class correlations have begun to recede”. You see that she is not a card carrying Pragmatist either. She fundamentally believes that the world should return to an orderly state where correlations and volatilities are more stable.
Mathematically, that is how you can define the uncertain market of the times – variable volatility and variable correlations, variable drift. A market model that cannot support trading.
The models for the other three environments might be:
- Boom – positive drift, low and stable volatility, low and steady correlations.
- Bust – negative drift, low volatility, high correlations.
- Moderate – near zero drift, moderate but stable volatility, moderate but stable correlations.
In her second point, she tells how to be ready for when the environment goes back to Boom or Moderate – by taking the classical Pragmatist position of under invested.
But the Pragmatist approach to risk is not really a Black Swan survival approach. If you really believe that a Black Swan event is coming, you would have the Conservator view of risk. That would lead you to move to a much lower expected upside and also a much lower likelihood of failure of your portfolio. In its purest form, the Conservator would accept almost no chance of total ruin. In actual practice, most Conservator leaning firms will accept risks that might cause a failure of the firm, but only if they have long experience with those risks and feel that they have them totally under their control.
Riskviews finds the headline Value-at-Risk model masked JP Morgan $2 bln loss to be totally appalling. JP Morgan is of course famous for having been one of the first large banks to use VaR for daily risk assessment.
During the late 1980’s, JP Morgan developed a firm-wide VaR system. This modeled several hundred risk factors. A covariance matrix was updated quarterly from historical data. Each day, trading units would report by e-mail their positions’ deltas with respect to each of the risk factors. These were aggregated to express the combined portfolio’s value as a linear polynomial of the risk factors. From this, the standard deviation of portfolio value was calculated. Various VaR metrics were employed. One of these was one-day 95% USD VaR, which was calculated using an assumption that the portfolio’s value was normally distributed.
With this VaR measure, JP Morgan replaced a cumbersome system of notional market risk limits with a simple system of VaR limits. Starting in 1990, VaR numbers were combined with P&L’s in a report for each day’s 4:15 PM Treasury meeting in New York. Those reports, with comments from the Treasury group, were forwarded to Chairman
Weatherstone. from History of Value-at-Risk:1922-1998 by Glyn Holten
JP Morgan went on to spin off a group, Riskmetrics, who sold the capability to do VaR calculations to all comers.
Riskviews had always assumed that JP Morgan had felt safe selling the VaR technology because they had moved on to something better.
But the story given about the $2 billion loss suggests that they were flubbing the measurement of their exposure because of a new risk measurement system.
Riskviews would suggest two ideas to JP Morgan:
- A firm that makes its money taking risks should never rely upon a single measure of risk. See Risk and Light and the CARE Report for further information.
- The folks responsible for risk evaluation need to apply some serious standards for their work. Take a look at the first attempt of the actuarial profession of standards for professionals performing risk evaluation in ERM programs. This proposed standard suggests many things, but the most important idea is that a professional who is evaluating risk should look at three things: the risk taking capacity of the firm, the risk environment and the risk management program of the firm.
These are fundamental principles of risk management. Not the only ones, but principles that speak to the problem that JP Morgan claims to have.
WordPress has recently added a feature that shows where the hits to the Riskviews blog are coming from. This captures views of the blog since late February, 2012 – just a few months so far. But there is a pretty wide spread of viewers.
Riskviews is totally facinated at how this media allows for such a wide audience and is humbled that people from so many places have seen fit to visit.
The top 20 countries in terms of total number of hits are the following. (WordPress does not seem to allow monitoring the much more useful “Unique Visitors”.)
Thank you everyone for your kind attention. Let it be known that Riskviews was never intended to be the platform for only one voice. Anyone who has an interest in expressing their views, once or regularly is welcome.
Just leave a comment to this post and we will discuss.
Most investors in hedge funds must be looking at them totally marginally. Certainly that is the way that hedge fund managers would suggest.
What that means is the ther investor should not look at the details of what the hedge fund is doing, it should only look at the returns. Those returns should be looked upon as a unit.
Certainly that is the only way to think of it that matches up with the compensation for hedge fund managers. They get paid their 2 and 20 based solely upon their performance.
But think for a moment about how an investor probably looks at the rest of their portfolio. They look at the portfolio as a whole, across all asset classes. The investor will often make their first investment decision regarding their asset allocation.
While hedge fund managers have argued for treating their funds as one or even several asset classes, they are almost always made up of investments, long and short, in other asset classes. So if you are an investor who already has positions in many asset classes, the hedge fund is merely a series of moves to modify the investors asset mix.
So for example, if the hedge fund is a simple leveraged stock fund, the hedge fund manager is lowering the investor’s bond holdings and increasing the stock holdings.
So if an investor with a 70% 30% Stock bond mix changes their portfolio to 65%, 25%, 10% giving 10% to a hedge fund manager who varies runs a leveraged stock fund that varies from all cash to 4/3 leveraged position in stocks, then they have totally turned their asset mix over to the hedge fund manager.
When the hedge fund is fully levered in stocks then their portfolio is 65% long Stocks, 25% long bonds, plus 40% long stocks and 30% short bonds. Their net position is 105% stocks with 5% short bonds. But that is not quite right. If you only get 80% of your performance, your position is 97% stocks and 1% bonds. That is right, it is less than 100%. Only it is really worse than that. That is the allocation when performance is good. When the stock market goes really poorly, you get the performance of the 105%/(5%) fund.
Other funds go long and short large and small stocks. The same sort of simple arithmetic applies there.
It is really hard to imagine that anyone who thinks that there is any merit whatsoever to asset allocation would participate in this game. Because they will no longer have any say in their asset allocation. What you have done is to switch to being a market timer. In the levered stock example, you now have a portfolio that is 65% long stocks and 35% market timing.
So in most cases, what is really happening is that by investing in a hedge fund, the investor is largely abandoning most basic investment principles and shifting a major part of their portfolio asset allocation to a market timer.
At a very large fee.
At his annual shareholder’s meeting Warren Buffet repeated his belief that there is no substitute for CEO attention to risk.
Anyone who has tried to do the CRO job without full unwavering support of the CEO would doubtless agree. The CRO job, just like the COO and CMO and other C suite officers job is delegated responsibility of the CEO. It is not independent of the CEO. Boards who try to set up a CRO function that reports directly to them and is intended to act as a check on the CEO are at best wasting their own and the CROs time. At worst they are creating a very unhealthy dynamic in the firm.
If a CRO is given the job of defense against killer losses and the rest of the firm is given the job of winning customers and making a profit, guess who will lose whenever there is a conflict. An adverserial risk function is not a healthy way to manage a company. By refusing to delegate the risk role, Buffet is sending a message to all of his companies that risk is important to him, the CEO of the firm that owns their company.
Now Buffet (or any other CEO that goes this route) needs to do more than refuse to appoint a CRO. A CEO who does not want any risk management to slow down his firm can quote Buffet and not appoint a CRO and then totally ignore risk.
The CEO/CRO needs to make it constantly known that they are concerned about risk by their words AND deeds. They need to talk the talk and walk the walk of risk management.
As Buffet knows, that does not necessarily mean that he needs a risk register of hundreds of risks. Berkshire Hathaway is in dozens of businesses and is actually exposed to hundreds of risks. But BH is also very large and diversified. There are actually only a few risks that need to be on Buffet’s plate as the CEO/CRO.
And what Buffet and other CEO/CROs need to do is to make sure that they are totally aware of what their firm is doing with the handful of truely killer risks. They need to make sure that:
- Everyone who could make a decision to increase the firm’s exposure to these killer risks knows that the CEO/CRO must be involved in that decision.
- The firm is being properly compensated for the killer risks that they are taking.
- The Risk Treatment programs for these risks are being properly maintained and operated.
- The firm has alternatives to the current risk treatment programs in case the existing programs become less effective or unavailable.
- The firm is carefully monitoring the risk environment that impacts those risks and any change or even strong hint of future change is brought to the attention of the CEO/CRO.
- The board is kept informed about all of the above.
Interestingly, this list does not change at all if the CEO decides to appoint a CRO. The list above can be a major part of the agenda when the CEO and CRO have their daily meetings.
In the world of Hunger Games, the odds are in your favor if your exposure to being chosen as a tribute is low.
If your exposure is high, then the odds are not in your favor.
When you are selected to be a tribute, everyone still keeps saying that phrase. It must be maddening, since the odds of winning the games is on the average 1/24. Most people would usually assume that something that has only a 1/24 chance of happening is something that they can count on NOT happening.
And you quickly learn that the tributes from some sectors HAVE done lots to make the odds go in their favor.
That is what risk management is all about – taking actions so that the odds move in your favor. Selecting the best risks at both the macro and micro level, trimming the risks with hedges and reinsurance to drastically reduce the likelihood of a killer loss and intelligently choosing the total amount of risk to take on a year are the ways the risk management makes the odds go in your favor.