Sharpe On Post Retirement

[mickeyd]

Can't get enough of William Sharpe. I ran across that article recently when I was researching something completely different, but could not pass up the opportunity to peek into his world one more time. It's written for financial planners, but is good reading for most of us that try to understand just a little more about retirement and what Sharpe has to do with it.

He listed a number of different uncertainties that advisors have to factor in: the risk of living longer (and outliving money), the risk of poor investment returns, the risk of poor health (and health insurance), the risk of inflation (despite the constant drumbeat of news about deflation), the risk of government programs such as Social Security and Medicare coming up short or being cut back. Sharpe adds counterparty risk, as well—the chance that those companies helping you shore up your high upside investment product, perhaps a company like Lehman Brothers, might not be there to help out if they go belly up.

“Economists were comfortable for many years with the assumption that markets have no memory,” he says. “In other words, that the probability of a higher or lower return next year is the same as the probability was last year—no matter what happened last year. Others, Jeremy Siegel [author of Stocks for the Long Run] would be sort of an extreme case, say, ‘No, no no. If the market does really badly’—he won’t say it this way, but I will—‘If the market does badly it feels quite badly for those investors and it will work to do better next time.’ If on the other hand, the market is giving you a lot of money it will say, ‘Hey, they don’t need it all the time. Maybe next time won’t be as good.’”

Sharpe On Post Retirement

 

[jazz4cash]

I'm having trouble with the statement that markets have no memory. Markets adjust gradually with little to no regard of the calendar. It NOT like a quarter that has no memory of the last 20 tosses. Yes, annual performance IS arbitrary, but it's the timeframe measurement that makes it so. If the economy is terrible in December, January's market performance is not likely to be stellar. I suspect upturns tend to be more gradual than downturns so maybe the opposite is less likely.

 

[haha]

If you believe in RTM, then you also believe that markets have memory.

Ha

 

[Nords]

I'm having trouble with the statement that markets have no memory. Markets adjust gradually with little to no regard of the calendar. It NOT like a quarter that has no memory of the last 20 tosses. Yes, annual performance IS arbitrary, but it's the timeframe measurement that makes it so. If the economy is terrible in December, January's market performance is not likely to be stellar. I suspect upturns tend to be more gradual than downturns so maybe the opposite is less likely.

Somewhere between FIRECalc "just like this" history and Monte Carlo "I can't remember yesterday" markets, there has to be a happy medium. Either that or market memory isn't significant and programmers would be recreating Bernstein's "Calculator from Hell".

Meteorologists basemark against a standard called "persistence". Persistence's forecast is that the next period (24 hours, tomorrow, five days) will be the same as the last period. Unfortunately persistence has had the best record in the business ever since it was invented... the good news is that its lead has declined rapidly as modeling has improved.

However the weather only matters if you're subject to having your crops washed away by the rain or your wedding day ruined by it. Again an 80% success rate is probably good enough.

 

[Gone4Good]

I'm having trouble with the statement that markets have no memory. Markets adjust gradually with little to no regard of the calendar. It NOT like a quarter that has no memory of the last 20 tosses. Yes, annual performance IS arbitrary, but it's the timeframe measurement that makes it so. If the economy is terrible in December, January's market performance is not likely to be stellar. I suspect upturns tend to be more gradual than downturns so maybe the opposite is less likely.

I think the idea that the "market has no memory" relates to the ability to forecast tomorrow's performance based on yesterday's. Chartists believe it can be done, but the best in their field were pound the table sellers of equities in late August . . . so there you go.

 

[haha]

I think the idea that the "market has no memory" relates to the ability to forecast tomorrow's performance based on yesterday's. Chartists believe it can be done, but the best in their field were pound the table sellers of equities in late August . . . so there you go.

That may be one idea; admittedly it is only words and not an algorithm.

But see Dimson, et al in Triumph of the Optimists, for a good argument that markets over the longer term do not follow a random walk.

Ha

 

[dex]

That may be one idea; admittedly it is only words and not an algorithm.

But see Dimson, et al in Triumph of the Optimists, for a good argument that markets over the longer term do not follow a random walk.

Ha

You could also add the business cycle, T theory, Dow theory, etc.

 

[LOL!]

Since you can't get enough of Sharpe, he made a satirical video that is making the rounds at Bogleheads and elsewhere. Here's a link: Lifetime Finance where he writes about the video in the Nov 24, 2010 posting.

 

[jazz4cash]

If you believe in RTM, then you also believe that markets have memory.

Ha

I do believe in RTM, but don't necessarily associate it with market memory. Maybe I just "don't get" the market memory concept as others do.

 

[Gone4Good]

I do believe in RTM, but don't necessarily associate it with market memory. Maybe I just "don't get" the market memory concept as others do.

If you read the article Sharpe is talking about building models, and non-correlated annual returns is just one assumption he mentions.

Recognizing that any model requires simplifying assumptions, non-correlated annual returns isn't a terrible one. Market returns may be mean reverting, but they aren't strongly so and they have zero appreciation for what day, month or year it is. In fact, mean reversion may take place on generational time scales, rather than annual ones. In which case, the annual non-correlation assumption seems pretty benign. For example, Shiller's stock valuation data shows that over the past two decades the stock market has traded above it's long-run median value 96% of the time. That persistence makes mean reversion a pretty worthless concept when planning for time frames relevant to most mortals.

 

[clifp]

Since you can't get enough of Sharpe, he made a satirical video that is making the rounds at Bogleheads and elsewhere. Here's a link: Lifetime Finance where he writes about the video in the Nov 24, 2010 posting.

I doubt Conan, Jon Stewart or Mark Steyn are worried about their jobs, but I LOLed. Especially about the part about structured products, and where to take the cruise.

 

[mickeyd]

Since you can't get enough of Sharpe, he made a satirical video that is making the rounds at Bogleheads and elsewhere. Here's a link: Lifetime Finance where he writes about the video in the Nov 24, 2010 posting.

That was one of the cleverest, most sarcastic piece of financial comedy that I have ever had the joy to read. "The Wall Street other-brothers..." Thanks s bunch LOL.

 

[LOL!]

I hope you didn't have to read it. It is also chock full of good advice in its own way (including the 4% rule of thumb).

 

[mickeyd]

I hope you didn't have to read it. It is also chock full of good advice in its own way (including the 4% rule of thumb).

I agree. Should have used viewed not read; you're right.