Problems with Constant Compound Interest (2)
I had many good comments on part 1 of this series. One was particularly good that focused on the economy as a whole, and how the promises made by the government were unlikely to be fulfilled with Social Security and Medicare. Our government attempts to finance programs on the cheap by relying on future growth, which does not always happen. They move up spending rapidly if growth is large, and small if growth is negative. In either case, the government grows as a fraction of the economy.
But let me consider asset allocation projections. It is really difficult to consider average projections of asset returns, whether in real or nominal terms. Asset returns vary considerably, and the number of years from which average returns are calculated are few relative to the volatility of returns.
We have created an industry of asset allocators, many of which have allocated off of foolish long-term historical assumptions, as if the past were prologue. Even if they use stochastic models, the central tendency is critical. What do they assume they can earn over long-dated investment grade debt? The higher that margin is, the more they lead people astray. Stocks win in the long run, but maybe by 1-2%, not 4-6%.
Consider defined benefit pension funds — after all, it is the same problem. What is the right long term rate to assume for asset performance? Alas there is no good answer, and with private DB plans continuing to terminate because of underfunding, the answer is less than clear.
Scoundrels use the mechanism of discounting to their own ends — they make future obligations seem smaller than they should be, magnifying profits, and minimizing capital needs. Cash flows are inexorable, though. There are few ways to avoid the promises from pensions.
Investors, be aware. Realize that long term investment assumptions are probably liberal. Also realize that long term assets and liabilities that rely on those assumptions have liberal valuations as well. After all, who wants to under-report income when the accounting is squishy?
Now, for my readers, what have I missed?
June 11th, 20094:36 pm at
Not only is the past not prologue, but there strong demographic reasons to expect lower returns in the future, possibly permanently. In fifty years world population will turn flat to down leading to a world of low to negative returns. Some innovation still, but possibly not even a world worth investing in. Always a chance of the miraculous, but not one to wager heavily on. Yet not necessarily an unsatisfactory existence where capital can be consumed without being replaced.
June 12th, 20093:23 am at
I like your post and want to comment on a couple of items. You point to the peak of the 1980’s inflation rates and the associated interest rates.
Robert Samuelson wrote a book called The Great Inflation and it’s Aftermath. http://tiny.cc/z9H9V
Basically you can explain a great deal the US stock market history of the 40 years by the spike in interest/inflation until the mid 80’s and the subsequent decline. Since you need an interest rate to value any cash flow, the decline in interest rates made all cash flows more valuable.
The thing that is odd and sort of ties this together is the last year. After interest rates crossed the 4% level things started blowing up. The amount of debt that can be financed at 3% to 4% is enormous. That is, as everyone knows, on of the root causes of the housing bubble. Anyway, starting last year, treasury interest rates continued to decline and all other rates went through the roof.
I was looking at this chart yesterday. _ http://tiny.cc/eCZzF The interesting thing to me was that when the system blew up, treasury rates continued to decline and all non guaranteed debt rates went through the roof.
Most of this is obvious and everyone knows the reasons. The one thing that seems novel is thinking of this as the continuation of a very long secular trend — or secular cycle. I don’t want to get overly political, but the decrease in inflation/interest in the 90’s to the present was a function of productivity/technology and Foreign/Chinese imports. Anyway, one effect of these policies was a huge rise in asset values, especially in the FIRE (finance, insurance, real estate) sector of the economy at the expense of our industrial and manufacturing sectors. This was also a redistribution of wealth from the rust belt to the coasts.
It is much more complicated then the hand full of influences I mentioned, but the one thing i haven’t seen discussed a lot is the connection of the current catastrophe to the long term decline in inflation/interest rates since the mid/late 1980’s. If you think about it, declining interest rates increase the value of financial assets and are an enormous tailwind for finance. I suppose if you had just looked at the curve, it would have been obvious that the trend couldn’t continue. Prior to the blowup, there were lots of people financing long term assets with short term, low interest rate liabilities. That was a big part of the basic playbook for structured finance, hedge funds, etc.
The reason that the yield spread exploded is well known. Here is a snippet from Irving Fisher. http://capitalvandalism.blogspot.com/2009/01/deflationary-spirals.html
June 12th, 20099:37 am at
David, one other tangent in terms of future return is technological innovation and destruction. Whale oil to kerosene lamps to electricity. Who has a secretary to type up memos for them? IBM barely saved itself from obsolesence (how many competitiors did?)
So, that A rated bond or ‘classic’ growth stock may not grow to the sky in a straight line.
This is one of the problems I have with projections out 20-30 years. The path of growth is not linear- as you point out. Regards.
June 12th, 20099:54 am at
Totally agree, David. For most institutional plan sponsors and consultants, asset allocation is the kingpin of everything that is done. But when those assumptions are off just slightly, it wreaks havoc on the entire plan. A couple things come to mind:
1. Since it is so incredibly difficult to make predictions and properly discount future assumptions, it seems to me one key is flexibility. People talk about the inherent advantage of the institutional investor, be it a pension fund, mutual fund, insurance general account, etc., because they have access to types of vehicles or securities that Average Joe investors do not. However, what is rarely discussed is the complete lack of flexibility that institutional investors face. The cost to transition hundreds of millions of dollars between asset classes is not insignificant in most cases, so flexibility becomes something that may be ideal but is absolutely not feasible.
2. Because institutional investors are lumbering (see #1 above) and generally have medium-to-long term time horizons, agency theory comes into play here as well. Investment management shops obviously want to cater to the huge institutional dollars, and what types of strategies tend to play the biggest role in any asset allocation? Large Cap US Equities. How can a manager make the most money for himself? By managing Large Cap US equities and employing a low turnover approach, thereby incurring fewer transaction costs and thereby having higher capacity to manage even greater and greater sums of money. But in order to manage that type of strategy, you inevitably end up making fairly explicit forecasts about what the future holds and then discounting back to present. While that strategy has worked reasonably well for a few asset managers, it tends to be more of an on-again, off-again type of approach that has little predictability and repeatability in terms of performance. BUT it makes the asset managers more wealthy than if they employed a more flexible approach that had more limited capacity…and therein lies the agency rub.
I’ll stop my rambling now.