A quote from Taleb’s book, beautifully sums up how we view and manage risk.
“The existence of a risk manager has less to do with actual risk reduction than it has to do with the impression of risk reduction. . .. By “watching” your risks, are you effectively reducing them or are you giving yourself the feeling that you are doing your duty? (41-42 FBR)”
In the article “Can we keep our promise” Mr. Arnott say we cannot measure returns sensibly without effectively measuring risk. He also talks about how we get sidetracked into primarily measuring peer risk versus measuring it as a whole or as multidimensional risk. Since financial managers are paid for peer risk management that becomes the main focus
To understand the multidimensional aspect of risk, let’s look at the three major category of risk
- Peer Risk – in other words the fear of being wrong and alone – this risk gets the most attention and it measure success based on similar portfolios managed by others and peer benchmarking. Our well-being depends on how others perceive us and how we rank compare to them.
- Asset Risk – this risk refers to an asset whose value may fluctuate due to changes in interest rates. We try to measure the asset’s default potential or market value fluctuation. Few example of asset would be equities, high-yield bonds and currency
- ALM Risk – asset liability management risk is the practice of managing risks that arise due to mismatches between the assets and liabilities. One example of this risk is pension plans. It arises mainly due to the following
- Interest risk – The risk of losses resulting from changes in interest rates and their impact on future cash-flows.
- Inflation risk – The uncertainty over the future real value (after inflation) of the investment.
- Longevity Risk – The risk to which a pension fund could be exposed as a result of higher-than-expected payout ratios. Higher than what a company originally accounts for as people live longer.
- Adverse selection risk – is the risk that sick pensioner’s tends to withdraw amount as lumsum whereas healthy pensioner’s preference a monthly payout. Thereby increasing the payout ratio.
Lastly there is always a career risk which exists as a person might lose job if he/she adopts an unconventional way of viewing and managing risk.
In his article, Mr. Arnott say that that these risks are interdependent and by focusing on any one risk investors leave themselves exposed to the other two risks. Suppose the company focus on reducing peer risk, by doing so they take large absolute risk and larger risk related to liabilities of the fund. All three risk are important individually too, all three matters, managing Peer risk is important as returns relative to peers outline the competitive positioning of an enterprise. Asset Risk is important as we may lose excess money due to market value fluctuation and ALM is important as they have the highest impact on our liabilities even with minor fluctuation on interest rate and future payouts. Despite the importance of all three dimensions of risk the primary objective for consultants and financial advisors has been to focus on Peer risk.
At the time this article was written, GAAP accounting allowed much of this risk to be smoothed out over several years, so the impact on reported earnings was about 80% (based on 5 yr smoothing) lower than the actual impact. This took away focus from the ALM risk and downplayed the importance of managing the same. Arnott’s approaches to sizing up the Asset-Liability Risk is to convert the % data(stand. Dev.) into a financial metric such as dollar at risk per share of common stock. This helps in putting things in perspective and helps in identifying the quantum of impact. As we see from table below the average fund has much more exposure to Asset Risk and ALM Risk than to Peer Risk. The ALM has the highest impact at about 60% more than the other two.
Table 1. Potential Losses for 19 Large Pension Plans for 2003 Due to:
Peer Risk Asset Risk ALM Risk
St. Dev. % 2.5% 12.4% 15.0%
Impact on EPS -$1.28 -$2.61 -5.12
St. Dev %: this is the standard deviation (volatility) for each risk around its expected value (its mean)
EPS: Earnings per Share (for 2003, the 19 firms with these pension plans had an average EPS of $2.24)
Mr. Arnott say that long term pension plan (> 30 yr) usually comprise of 1/3 of the total responsibility but it’s extremely sensitive to interest rate. In short its small liability but has big volatility. This small liability is five times more sensitive to interest rate than a pension liability which is due next year.
Based on the table/figure 2 – we see that when interest rate drops from 7% to 5% the long-term liability increased 76% as against short-term liability increasing by only 6%. Also the volatility of the longer term pension liability is asymmetric. If interest rates were raised by a certain % say the liability would fall by x% and if they were decreased by the same % then instead of a X% rise we actually notice a (X+Y) %rise. The volatility is skewed, it is not symmetric in positive and negative direction. Which makes it extremely important to be managed in the right way.
To balance the three risk , Mr Arnott suggest to invest such as to bring each of these three area to similar level of risk and then targeting to beat any two benchmark out of three. This sift in benchmarking would give the consultant or investing officer incentive to focus on the performance relative to the liabilities and offer more “tolerance towards peer risk”. This mean that as long as they are not being punished for falling short of peer risk, they can focus on risk that have greater impact.
The current most common allocation for pension plan is too little on short term bond and the primary reason for this is stated as long term bonds yields too little. Arnott recommend that most pension plan should have a modest level of investment on long term bonds as such to substantially reduce the risk on pension plan. He says that merely by investing 10% of fund’s asset in ultra-long strip, we can eliminate ~30% of interest rate sensitivity and ~40% -50% of asset liability mismatch.
Ref: Can we keep our promise by Arnott