Why you should be greedy when others are fearful

COMMENTARY This month, the Journal of Financial Planning had an interview with William Bengen, the man whose research is behind the famous (in my profession, anyway) 4 percent withdrawal rule for retirement. While I'm certainly appreciative of his research and subscribe to his thoughts behind a prudent withdrawal rate, he missed the mark regarding how advisors should be helping their clients invest in the volatile market.

First, it's important to understand that Bengen is a very smart man. He graduated from MIT. His research on safe withdrawal rates in retirement has been the gold standard on the topic for two decades.

However, his interview was pretty revealing when it came to how advisors should be handling client assets. Bengen was asked about using dynamic withdrawal policies in the face of extreme market volatility. (Remember, we don't have to travel too far back to reach a time where swings of several hundred points were a rarity, rather than the rule.) His answer: "I think that they need to find a money manager who is willing to do something other than buy and hold. I'll be quite frank about it. Buy and hold in these environments is an invitation to disaster."

In a way, I agree with him. Buy and hold isn't a good strategy. The proper strategy is buy, hold, rebalance and tax manage.

Simply buying and holding investments means your allocations will likely drift away from your targets, giving your portfolio a much different makeup and risk profile than you originally intended. You have to rebalance back to your original allocations. You must also be vigilant about managing your portfolio for taxes. You may have losses you can book in the middle of the year that might not be there at year-end, thus losing the opportunity to have Uncle Sam share in part of the pain of losses.

However, my assumption is that he actually meant to engage in market timing, especially with his words following the above statement: "You need a money manager who is willing to withdraw your funds from the market when there is high risk present -- and I believe the risk is very high now -- and be willing to then further invest you when values improve and the risk is reduced."

This actually flies in the face of when you would want to be in the market. While it's difficult to stay in the market when risks are high, that's when expected returns are high. It's easy to stay in the market when risks seem low, but that means you buy when expected returns are low and sell when expected returns are high, which doesn't make sense.

My profession owes a lot to Bengen for his work on retirement withdrawal rates. Honestly, all prudent investors do as well. But I'd still prefer the advice of Warren Buffett, who said investors "should try to be fearful when others are greedy and greedy only when others are fearful."

[LarryswedroeCBS]

John
Yes you do have to make an assumption about future growth rate of the economy and IMO there is no reason to believe it will be faster or slower than it has been for the last 100 years. That at least IMO is a reasonable assumption.
Best
Larry

[johndc43]

Now it seems we are in complete agreement in concept and only disagree about the future growth rate of the economy. IMO a slower growth rate is the better assumption, but I hope I am wrong.

Thanks Larry

[LarryswedroeCBS]

JohnDc
Okay, now clear, However, we have to clear up one issue here.

Yes you have to make a forecast of future returns and SD and correlations. But the forecast does not and should not be based on past returns, but should instead be based on current valuations as valuations matter. High valuations mean low expected returns and vice versa.

Each year we change our inputs into the Monte Carlo Simulations we use to help determine a prudent withdrawal rate. But the forecasts/assumptions are based on the Gordon Model, not our opinions. And that would not be market timing at all, but altering your plan because the assumptions you put into the plan have changed and that impacts the need to take risk.

I hope that is helpful and I believe puts us in agreement.

Best wishes
Larry

[johndc43]

I agree that valuations matter as you suggest, but there still is a simple yet major area of disagreement. Valuations aside, if one chooses historical precedent to make forecasts, he is expressing his opinion that history is the best determinant of the long-term prospects for the market. On the other hand, some believe that the long-term prospects for the economy and markets is worse than historical. In either case it is only an opinion, but I happen to side with the worse-than-historical group. Note that I am referring to long-term prospects and how this affects one's investment plan, I am not suggesting that the market got it wrong in the short term.

[LarryswedroeCBS]

John
Unfortunately the evidence shows that those efforts to time the market are highly unlikely to prove productive
In fact studies have found them to be highly counterproductive

One study on Tactical Asset Allocation strategies found that not one single pension plan that had engaged in the strategy benefited. Not one.

Even Peter Lynch, perhaps the greatest fund manager of all time said he was always fully invested

Even Vanguard just shuttered its tactical asset allocation fund-VAAPX

And believe Don Philips of Morningstar recently stated that he has never seen a mutual fund benefit over the long term from market timing strategies.

IMO the right strategy is to anticipate that bear markets will occur and build that into your plan. And then rebalance. That allows you to do what all investors endeavor to do, buy low and sell high.

Best wishes
Larry

[johndc43]

Larry

I agree that market timing strategies don't work, but that was not my point. We must all base our long term investment plan upon a certain set of assumptions. We can assume a pattern of boom or bust based upon historical precedent and arrive at, for example, a 4% SWR. If we accept those historical assumptions, then I fully agree with the buy-hold-rebalance strategy that you suggest.

The thing is, I don't buy into those historical assumptions any longer, and I think Bengen and Bogle probably agree. Of course, I have no idea what the future economic and investment picture will be, but I do believe that it will be something worse than the the historical precedent that you, and most others, use as a base case.

My point is quite simple. It is best to accept a new base case that is worse than historical; adjust your investment plan accordingly, considering individual circumstances; and then stick with your new plan (buy-hold-rebalance if you like) until fundamental circumstances change for better or worse (if ever).

I don't consider this a market timing strategy. It is probably better described as quasi-permanent adjustment to a new paradigm.

John

[johndc43]

Bengen is a very smart man. So are Buffett and Bogle. There is a reason these people sometimes veer from the absolute course of buy-hold-rebalance. These people never recommend market timing in the popular sense of jumping in and out. What they do recommend is reevaluating the big picture of investments resulting from major shifts in the economic climate. In other words, they suggest not continuing to base an investment plan on historical patterns when these major shifts occur. Kudos to them IMO.

[johndc43]

Bengen is a very smart man. So are Buffett and Bogle. There is a reason these people sometimes veer from the absolute course of buy-hold-rebalance. These people never recommend market timing in the popular sense of jumping in and out. What they do recommend is reevaluating the big picture of investments resulting from major shifts in the economic climate. In other words, they suggest not continuing to base an investment plan on historical patterns when these major shifts occur. Kudos to them IMO.

[MIO42]

What a load of kr...........p

[LarryswedroeCBS]

MIO42
Thanks for your insightful comment. Perhaps we can all benefit from your wisdom. Can you share with us exactly what you disagree with and why you think it is incorrect?
Best wishes
Larry