Adrian's Rule Revisited

[rwwoods]

Adrian’s Rule states that you set your equity percentage of your portfolio to twice the maximum bear market loss that you are willing to accept. This rule applies very well to an investor in the accumulation phase, but significantly understates the drawdown risks in the withdrawal phase. I will illustrate with an example.

I make the following assumptions. One investor is in the accumulation phase and one is in the withdrawal phase. Each have the same risk tolerance and can accept a 30% drawdown in a bear market, so they set their equity to 60% and bonds to 40%. The withdrawal rate is 4% per year of the initial balance with no inflation over the period of the analysis, and the withdrawals are taken at the start of the year. Both investors start with the same $1000 balance and no funds are added over the period of the analysis.
A bear market lasts two years and results in a net 30% drawdown in the portfolios. The following three years in a bull market results in gains of 20%, 10% and 9% per year.

The results show that the investor in the accumulation phase has regained his losses. However, the investor in the withdrawal phase is still down 23%. Obviously, the bear market risk is much higher for the investor in the withdrawal phase. Not only is the maximum drawdown 36%, the time to recover the loss is very long.

So, for this example, one could modify Adrian’s Rule to state that, for an investor in the withdrawal phase, set the equity percentage to 1.67 times the maximum acceptable loss. For this example, the equity would be lowered from 60% to 50%. Of course the results will vary depending upon the length of the bear market and the speed of recovery. YMMV.

Code: Select all

Date	Gain/Loss	Accum	Withdraw
31-Dec				1000	1000
1-Jan				1000	960
31-Dec	-17%		830	797
1-Jan				830	757
31-Dec	-16%		697	636
1-Jan				697	596
31-Dec	20%		837	715
1-Jan				837	675
31-Dec	10%		920	742
1-Jan				920	702
31-Dec	9%		1003	766

[Adrian Nenu]

Obviously they do not have the same risk tolerance because all other things being equal, one is in the accumulation stage while the other is in the distribution phase with significantly larger portfolio, requires income and can't handle the same percentage loss as the investor in the accumulation phase. It's up to investors to carefully consider their personal circumstances while determining what kind of losses they can handle or can't, then plug the numbers into the formula.

Adrian
anenu@tampabay.rr.com

[Midwest_Investor]

I still like the simplicity of Adrian's original rule, so I'll stick with it.

his original rule has been a good rough guideline for me.

[detifoss]

I'm not a huge fan of the rule (though I can't argue with its simplicity), but this example doesn't have anything to do with the rule.

The maximum tolerable loss for an accumulator is much higher than it is for a retiree.

I'm the former and my max 'tolerable loss' is somewhere around 90% (so I guess I should be 180% equities... - i think there is a nice thread on how that works out!), but when I retire 30+ years from now I would think my maximum tolerable loss is going to be less than 20%, perhaps even less than 10%, thus my equity % will be around 20-50%...

[Adrian Nenu]

The risk of loss asset allocation rule or formula is not for everybody. There are exceptions just like with everything else in investing. But it is helpful to most investors because portfolio risk of loss and ability to handle loss must be determined before suitability can be determined. It forces investors to THINK HARD about risk and its consequences on their financial goals.

Adrian
anenu@tampabay.rr.com

[jeffyscott]

Adrian’s Rule states that you set your equity percentage of your portfolio to twice the maximum bear market loss that you are willing to accept.

I make the following assumptions. One investor is in the accumulation phase and one is in the withdrawal phase. Each have the same risk tolerance and can accept a 30% drawdown in a bear market, so they set their equity to 60%...

I don't think the rule is what "drawdown" is acceptable, it is what loss is acceptable. Spending is not a loss. In addition, he indicates that 50% is the absolute maximum equity, regardless of the tolerable loss part.

Finally, a tolerable loss means, I think:
What permanent loss are you willing to accept?
not:
What temporary decline, that you will surely recover in a year or two (because we secretly don't really believe stocks are risky) would you be willing to see?

[cinghiale]

Borrowing from a discussion on another thread:

On Adrian's "Nifty 50/50" thread, jeffyscott wrote:

Our maximum tolerable loss is based more on $ than %.

I think this perspective and distinction is important, and should be brought into this conversation. When applying Adrian's rule, one should ask the personal question of maximum allowable loss in both real and percentage terms, and then, in my opinion, opt for the lower, more conservative amount. Determining the maximum must feel right and be cognitively satisfying both ways.

[Adrian Nenu]

I think this perspective and distinction is important, and should be brought into this conversation. When applying Adrian's rule, one should ask the personal question of maximum allowable loss in both real and percentage terms, and then, in my opinion, opt for the lower, more conservative amount. Determining the maximum must feel right and be cognitively satisfying both ways.

It doesn't matter how you put it - expect 40% - 60% bear market loss on one standard deviation bear market. Could be more. You have to account for that happening several times during your investing lifetime. Sometimes twice in one decade. Returns may be or may not be the historical average - nobody knows. But you have to account for the possibility that they will be much less than historical average. That is equity risk. Investing is simple, isn't it?!

Adrian
anenu@tampabay.rr.com

[jeffyscott]

Yes, but there is a difference to most people between, for example, the risk of losing 50% of $50,000 or 50% of $1,000,000.

[Adrian Nenu]

Yes, but there is a difference to most people between, for example, the risk of losing 50% of $50,000 or 50% of $1,000,000.

Exactly, that's why personal risk of loss should be periodically evaluated and equity exposure should be reduced as distribution phase nears.

Adrian
anenu@tampabay.rr.com

[dbr]

Yes, but there is a difference to most people between, for example, the risk of losing 50% of $50,000 or 50% of $1,000,000.

Exactly, that's why personal risk of loss should be periodically evaluated and equity exposure should be reduced as distribution phase nears.

Adrian
anenu@tampabay.rr.com

Equity exposure could be increased depending on the circumstances of the individual.

[BigD53]

Obviously they do not have the same risk tolerance because all other things being equal, one is in the accumulation stage while the other is in the distribution phase with significantly larger portfolio, requires income and can't handle the same percentage loss as the investor in the accumulation phase. It's up to investors to carefully consider their personal circumstances while determining what kind of losses they can handle or can't, then plug the numbers into the formula.

Adrian
anenu@tampabay.rr.com

I like Adrian's formula. But agree that it may need to be/should be revisited and adjusted during retirement years.

And of course, for most rookie investors, the best laid plans and formulas go out the window during a good bear market. :lol: Seeing the actual losses on their monthly account statement is the best "teacher", and should definitely open their eyes to the risk involved.

Everybody is wide-eyed and eager during the bulls.....

[BigD53]

It's unfortunate that more financial planners and advisers don't emphasize the risk aspect more often. Instead, they will fill your head with "Look how much money you can make in stocks! Look at these historical numbers. Look how your money can grow. The market always goes up over time."

When I started investing, the only mention of "risk" and possible loss of money was a small sentence buried way at the back of the prospectus.

The FIRST thing out of an adviser's mouth should be: "Sit down and shut up. Are you willing to lose 50% of your portfolio? Can you afford to lose 75% of your life savings? Maybe even 100%??? And you may never recover from that loss? That's a worse-case scenario of course, but let's discuss the real risk involved with stock market investing....." :lol:

That guy probably wouldn't have any clients, but at least he's being practical and honest.

[rwwoods]

I was suprised as to how much larger than forcast my drawdown was in this bear market. Once I analysed it, it was apparent that the common rules for loss, such as Adrian's rule and Swedroe's rule, understates the risks for those in the withdrawal phase.

Withdrawals obviously cause a drawdown in equity. Those drawdowns along with those caused by bear markets creates a risk that one may outlive their equity, especially if the withdrawals are too large or if a severe bear market hits immediately after retirement. However the drawdown came to be, the end result is the same - less equity and increased risk of ruin.

The question remains, how should the common rules for acceptable loss be modified to properly forcast loss for those in retirement? Obviously any rule is a guess, but it should be our best guess. Unfortunately, I suspect that the variables are too great to derive a meaningful rule other than to say that one's loss may increase by up to 5%.

[Gekko]

nobody knows what their true risk tolerance or acceptable loss is until it happens. everyone is brave in a bull market and fearful in a bear market. compare posts here on this very forum from October 2007 vs. March 2009. i bet most people THOUGHT they had a high tolerance when you asked them in October 2007. and in March 2009 their tolerance suddenly changed. market conditions and emotions will continuously cloud our judgement. hence - i like an age based allocation strategy which is immune from emotion and arbitrary floating variable nebulous feelings of what's "acceptable" -

Stocks % = (100 - Age)

[Chuck T]

Personally, I think Adrian's rule or formula for choosing an AA is good solid guidance. It like most rules/formulas is a starting point. It is intended to make investors aware of the potential significant risk of investing in equity funds. That is even more true if you are near, or in retirement where there may be little time left to recover from major market losses. As has been said, everyone's situation is unique and must be evaluated.

Following the rule/formula may help new investors avoid the crushing blow of a 2008 market downturn. I know it helped me and I am glad Adrian posted it.

[TwoCyclists]

"The question remains, how should the common rules for acceptable loss be modified to properly forecast loss for those in retirement?"

We can't forecast loss, but we have seen a 90% loss from equities in the Great Depression and losses close to 50% have occurred three times in the past 10 Bear Markets (starting in 1957).

Here's what I have done, FWIW. Calculate the minimum annual income you need to live on in addition to your retirement income from pension/Social Security, etc. Plan to withdraw that amount of annual income at approximately 2.5% from a portfolio of highest-quality fixed income assets (e.g. Treasuries, TIPS, CDs, MM).
That is your fixed-income allocation. Above that, invest in a portfolio of diversified, low-cost, deeply liquid equity asset classes (e.g. domestic, foreign developed and emerging market).

I've been pretty happy with that so far....

[Adrian Nenu]

It's unfortunate that more financial planners and advisers don't emphasize the risk aspect more often.

Because it would hurt sales. You don't see it in many books either. Remember, these are mostly salespeople who earn the largest commissions from equity securities, not from FDIC insured CDs. Secondly, most of them have no idea how to measure risk because it is not required by designations or training, except CFA. If they don't know the risk of a proposed portfolio, they cannot determine suitability.

At the minimum should be able to show clients the risk of loss of a proposed portfolio. Then figure out if the potential loss is something the client can handle and compensate for or not. Also the psychological impact the loss would have on the client, family, etc. Make sure diversification is adequate. Then adjust the asset allocation to match the clients risk tolerance and other needs.

Adrian
anenu@tampabay.rr.com

[Rager1]

The question remains, how should the common rules for acceptable loss be modified to properly forcast loss for those in retirement?

The method I use (that Michael Lebouf once suggested) is to establish a maximum dollar amount for equities. It's easy to see, in dollars, what a 50% drop would do to your portfolio. During periods when the market is trending up, it forces you to sell equities when you exceed your dollar limit.

Ed

[investnoob]

This a great discussion. The one thing I like about Adrian's rule is that it is possible to change according to your specific circumstance. But there are kinks that each individual investor would have to work out. For example...


Tolerable Loss x 2 = Equity Allocation ≤ 50%

@ 30 yrs from retirement, Tolerable Loss = 50%. So Equity Allocation equals 50%.

@20 yrs from retirement, Tolerable Loss = 30%. So, here, an investor would have a decision to make. Should he keep his Equity Allocation at 50%, or lower it due to his lower tolerable loss? I would lower. His tolerable loss has dropped by 40%. So, he could lower his equity allocation by 40%.

or,

@$10,000, Tolerable Loss = 50%.

@100,000, Tolerable Loss = 30%. And so on.

After being a member of this board for over a year or so, I've realised that the second set of examples are most relevant to me.

I cannot imagine amassing a half million dollars only to risk 30 to 60% of it in a traditional 60/40 stock/bond split. I could not afford losses like that, regardless of what age I was - I think. So, at that point my equity allocation would be seriously reduced. My investment goal would be changed from accumulation, or looking for returns, to protection.

But, had I not been reading this board, I doubt I would have arrived at this conclusion on my own. This formula really helps me to think it through.

[CaliJim]

bump it up

[alpenglow]

Bumping up a blast from the past.

It is always interesting to browse the post titles on the front page. How quickly we gone from "I'm 100% stock" to Taleb portfolios and max acceptable loss.

[jeffyscott]

Because in my view, this should be considered as I described above:

Finally, a tolerable loss means, I think:
What permanent loss are you willing to accept?
not:
What temporary decline, that you will surely recover in a year or two (because we secretly don't really believe stocks are risky) would you be willing to see?

This becomes a rebalancing limit for me. For example, now in retirement, I can do this by choosing to not continue to rebalance once fixed income is down to $X.

[cinghiale]

Like jeffyscott, I chimed in on this back in 2010, before Adrian took his ball and went home (never quite understood the reason why).

Along with Adrian’s rule, I would mix the following three ingredients into the recipe for asset allocation and dealing with risk:
— Mr. Bogle’s “Age in Bonds” suggestion.
— Larry Swedroe’s admonition to gauge one’s “ability, willingness, and need” to take risk on the equity side.
— The sleep at night factor. (This one seems to be making a comeback on threads discussing the last few weeks of equity market downturns.)

I think that wisdom can be found in the interstices of these four gauges.

[GrowthSeeker]

Adrian’s Rule states that you set your equity percentage of your portfolio to twice the maximum bear market loss that you are willing to accept.

Another way to look at the withdrawal phase would be to keep Adrian's Rule unchanged, but to multiply your inherent risk tolerance by a fudge factor.

So, if you're in the accumulation phase, your permissible maximum bear market loss is X%.
Then in the withdrawal phase your permissible maximum bear market loss is .8 * X%. (or whatever fudge factor you like).

Either change the Adrian multiplier or insert a new multiplier to your "risk tolerance".

For example, when I was deciding on a pf AA, I used several heuristics to come up with a number. One of them was to say: "I'm a 50:50 stock:bond 'kind of guy', but since I'm newly retired and feel especially vulnerable to sequence of return risk, I'm going to multiply that 50 number by 80%: therefore AA = 40/60.