Friday, June 24. 2011
The inspiration for today's blog post is a series of email and in-person conversations I've had lately with several planners, who in today's nervous economic environment have been very critical of other financial planners who have gotten defensive with their client portfolios. In the typical fashion, they have criticized that if the client has reduced equity exposure now, even if the planner is "right" and a market decline comes, the planner will lose money in the form of foregone growth for the client by being unable to time when to get back into the market. But I think we're missing the point.
Certainly, we can (almost?) all agree that if a client has a $1,000,000 life savings invested in equities, and the market crashes down 40%, taking the portfolio to $600,000, then it's probably not a great time to get out of equities. The market is likely priced for a bounce or rally, and the client may very well miss an ensuing price increase that could recover most or all of the losses. Of course, that's not guaranteed, and the markets could always go lower - as the client fears when he/she sells out "at the bottom" - but to say the least, the odds of a rally would appear to be much higher after a market crash, if only to recover the losses that just occurred.
But this is very different than the client who sells out BEFORE the crash in fear of a potential market decline. When the decline comes and the index crashes downwards, the client is already safely in cash, earning a modest cash/bond/fixed return on that $1,000,000. Which means even if the market rallies sharply after the crash... IT DOESN'T MATTER. A client who has a 40% crash followed by a 50% rally watches their $1,000,000 rubber band down to $600,000 and back up to $900,000. The client who sits in cash simply goes from $1,000,000 to $1,010,000 to $1,021,000 as they earn their meager 1%/year short-term fixed return. Which means the client who misses the crash, and the recovery, still has more money (not to mention achieving it with no downside volatility!)! In fact, if the client simply says "I won't buy back into the markets until their reach their old highs" - then nothing is lost! The hypothetical stock index goes from 1,000 to 600 and back to 1,000; the client's cash portfolio grows slightly; ironically, the client who is out actually has MORE to reinvest back into the markets when they recover to 1,000, because the client can reinvest not only 100% of his/her principal, but also a few years' worth of fixed returns, too. Or put more simply, if you miss the crash, you actually don't need to participate in the recovery, either! The recovery is only "needed" if you failed to get out of the way in the first place.
Instead, the "risk" of getting out not at the bottom when times are bad, but at a perceived/potential top when times are still good, is the risk that the market won't crash at all. If it does, the client has already won, even if the client doesn't re-enter the market until the entire recovery occurs. If there is no crash, and the market just keeps going up, the client won't participate of course. But is this always bad?
We often treat markets as if they're always the same, and the upside is always better (or at least equal) to the downside, but that doesn't seem realistic. Sometimes the risk proposition of the markets is more asymmetric, where the upside benefits are limited, but the downside risks are increased. In such environments, is it really, truly, always necessary to force clients to shoot for a tiny bit of additional upside, with a huge downside risk on the table? If we say as financial planners that we're about managing risk, in addition to trying to earn returns, why can't there be some environments where the "manage the risk" part actually trumps the "earn more returns" part?
So simply put, why is it ALWAYS so universally bad to talk about getting clients out of risky markets? Does every client really have to get every tiny bit of market upside, even when faced with the risk of a financial crisis if things go the other way? Even when it's true that if the client actually gets out BEFORE the crash, the client doesn't have to get back in at any point until the market returns to its old highs anyway?
Some of you might be saying "yes, but no one can predict these things" - yet that doesn't appear to be entirely true, either. For instance, Financial Advisor magazine did an article two years ago highlighting several planners who saw something bad coming in 2008 before it happened - including noted retirement researcher Bill Bengen. Wouldn't our time be better served by asking them how they did it, and trying to learn from them and what they saw, rather than casting stones at them for breaking with conventional wisdom?
So what do you think? Is there really a difference between the client who gets out before the crash - at the top - rather than the one who more clearly damages themselves by getting out after the crash at the market bottom? Is there EVER a time when the risk of the markets could overwhelm their upside potential? Can we learn something from planners who actually have made the calls effectively?
Psychologists Daniel Kahneman and Amos Tversky won the Nobel prize in economics for developing a theoretical framework to show how we make often irrational decisions when faced with real life economic trade-offs. As a part of their prospect theory researc
Tracked: Aug 18, 15:55
Like the old Certs ad, "stop, you're both right." No one can be certain the market will rise or fall. Taking an all or nothing approach and missing - either way - just once - hurts the client and they start looking for a smarter adviser.
How about taking small incremental bets, say not more than 2% - 5%? Those are easy to make, easy to back away from if wrong, and if you're right 51% of the time, you're a genius.
I don't think there's anything about this that has to be all-or-none.
In point of fact, most of the firms I know that are tactical in this regard have limits. A client with 60% in equities might be able to go up to 80%, or down to 40%, but no further.
I actually had a friendly debate with Harold Evensky at a conference last week where we were both speaking, trying to decide if that's a tactical strategy (60% +/- 20% in equities), or a core & satellite (40% core in equities, 20%+ that can satellite out to equities or something else).
But indeed, I see no reason why this has to be characterized as all-or-none you're 100% in equities or 0% in practice. It just makes the examples easier and less math intensive.
That aside, it's worth noting that "missing" doesn't always mean there's a smarter advisor in my opinion.
Let's say there's a 99% chance of rain. I tell you to bring an umbrella. It turns out the other 1% occurs and it doesn't rain. Does that mean I was wrong? Does that mean there's a smarter rain forecaster out there? Not necessarily. My 99% prediction could have been THE #1 smartest prediction of any weather forecaster on the planet. The mere fact that there's uncertainty means I can be the smartest and best forecaster, and not always be right.
It does, however, imply I'd be right more often than wrong. But that also means one-wrong-forecast-and-you're-out can't be an effective means to measure it.
Just food for thought.
The point is to make changes when the odds are disproportionately in your favor (i.e., when upside is very limited and downside risk is elevated).
There are measures, such as long-term valuation, that provide remarkably consistent and predictable indicators when upside is limited and risks are elevated.
Thanks for highlighting how this issue has been misframed by those who chide us so-called market-timers.
And your observation is spot-on that missing out on market gains largely becomes a moot issue if you never incurred the catastrophic losses to begin with.
Certainly, there are many advisors who are true believers in strategic asset allocation and all that entails. But the criticism of market-timers might be better redirected toward the rest of the advisor field.
Frankly, a lot of advisors are just lazy when it comes to portfolio management. It's easier to take an "ignorance is bliss" approach, strategically allocating to models/categories simply because nobody knows exactly how things will unfold.
But then when the markets have a (largely foreseeable) train wreck, those same advisors often employ the "who-could've-known" defense and hide in the financial industry herd. Safety in a large group, even though the group itself failed (and will fail again).
For those of us market-timers, it takes hard work, research, and especially independent thinking and gut-check, to manage a client's portfolio contrary to the direction of the herd.
You're exactly right...it's never a 100% in or out matter, nor is it ever a matter of being able to "precisely" time the getting in or out. Advisors who spout that are simply misframing the issue.
Investment clients are best served when advisors provide portfolio management that's at least "approximately right" most of the time while avoiding the catastrophic losses.
Thanks again for raising this issue.
I totally agree with your idea - it's about up/down side potential (or probabilities) of the market based on market valuations (P/E10).
The higher the current market valuation the lower the probability it will go up, the less it will go up if it does and, perhaps most important, the faster and farther it will fall when it corrects.
Right now the P/E10 is ~23 and Philbrick/Butler are predicting essentially 0% annualized returns from the domestic stock market over the next 10 years - they have an amazing track record of accuracy in the 10 and especially the 15 year window. Who am I to argue...the risk adjusted return from a 0% APR 10 year market is negative!
I firmly believe we OWE it to our clients to help them reduce their exposure to over valued markets - the more overvalued, the less exposure.
Much like politics, let's not make this a polarizing issue of buy-and-hold vs. market timers. It's just not that simple to generalize for everyone. Besides, I don't think we should be labeling people like my dad market timers. There is a big difference here versus very active traders. As Michael points out, my dad just brought a huge umbrella for his walk toward retirement. I prefer to feel the rain once in a while because I hate hauling that damn umbrella everywhere, but also because I have enough time to dry out. Neither is wrong because we can't accurately predict the weather.
You are asking people to guess right twice. They have to guess correctly that the market is at a high, and go to cash. Then, they must wait for the market to drop then come back, and then to guess right on getting back in after a recovery, but without it being another market high.
Bruce, sometimes "anti-timers" will hire market timers because they except that they can't know which of the 20k+ companies out there are good buys and which aren't. The average analyst on Wall Street covers 25-30 companies, and over worked analyst can cover 150 companies. so, if you don't have 133 over worked analysts, you aren't getting full market coverage. Any one analyst picking individual stocks is doing their clients a disservice.
The whole point is not how to gain additional return, it is how can we lower volatility. This can be done using a cash-hedge, proper diversification, and other strategies.
As Michael Haubrich said the other day to me on the phone, [paraphrased] If we attempt to time the market, and boost our clients expectations of returns, the best thing we can do is meet their expectation. If we help them to understand market volatility, we will typically meet expectations, and actually have a chance to beat them.
To say "a lot of advisors are just lazy about investments" undercuts many of us. Some of us are more focused on the holistic financial picture, and not just investments. That's what we call financial planning.
I find this "double-timing you have to be right twice" argument to be totally specious.
As YOU say, "the whole point is not how to gain additional return, it is how can we lower volatility."
Client A stays in the market as it crashes from S&P 1300 to 1100, and then stays in for the rebound from 1100 and back to 1300.
Client B sells out of the market at 1300 because he's terrified of a pending market catastrophe being signaled by any number of economic indicators. The market falls from 1300 to 1100 and then rallies back to 1300. Client B refuses to touch the market until it returns back to the previous 1300 high.
Who has more money? The answer is neither; they have the SAME wealth, and the SAME return. The difference is that going forward, client B owns the market at 1300 (after recovery) having had little volatility, and client A owns the market at 1300, having ridden out a miserable rollercoaster.
Thus, the irony is that client A took a radical increase in volatility, risk, stress, and sheer misery, for not a dime of additional return. No one ever said client B will earn more (although that's also a possibility); in practice, all client B did was earn the SAME return, with less risk, less volatility, less stress, and less misery. He had only ONE decision to make: to sell in times of high risk, and to simply wait until the market returned to that selling point before buying back in again (ensuring that he earns NO ADDITIONAL return, but simply the SAME return).
Why do we insist on inflicting that rollercoaster upon clients when it didn't even increase their return? And if we did deliver that, is it REALLY such a problem that once in a while we might get defensive and the bad event doesn't happen? If there's a 95% chance of rain and I bring an umbrella, was my umbrella decision wrong because it turned out to be sunny after the fact? Or did I make a correct risk management decision, simply realizing that by definition, not all risks actually materialize, but it doesn't mean managing the risk was wrong.
You seem to be separating getting out at the top of the market, and market timing (I could be misreading). You are absolutely correct that if you got out at the top, and back in after the recovery, your clients would have the same return with no volatility.
I like the graphs in this article: http://www.plcasset.com/market_timing_examined.html
If you want to say you have to guess correctly once, that is fair. You have to predict that the market is at a high, and no one knows that. Every indication for the last 2 years is to sell bonds, because they were at a high, and yet they kept going up. Markets are not rational, and therefore can not be predicted.
If you keep your clients from missing just 10 days of return out of 3,650, you have cost them average market return (that coming from the graph in the above article).
In 1975, Sharpe determined that to beat the market (and I believe this applies to lowing volatility while maintaining the same return, just as it does to trying to produce higher returns) you need to be correct 74% of the time. So my question is, can you (or anyone) tell their clients with a straight face that they can predict market highs with 3/4 accuracy (more than 50% accuracy would be impressive)?
There are numerous factors - market valuation probably being the most prominent - that clearly, blatantly, easily give you signals when the market is high and priced for sub-standard returns. Valuation easily predicted the sub-standard returns of the past decade. Its efficacy over "intermediate" time horizons (3-5 years, especially 5-10 years) is absolutely phenomenal.
So if the only problem with defending your clients in risky markets is that for a short period of time they COULD underperform an irrational market getting more irrational, and in the long run they are more successful 100% of the time, what's the problem?
And yes, show me a market with a 20+ P/E10, and I'll show you a market that has failed to achieve average returns over the subsequent decade, every time in history.
The data is there.
When would you have gotten out during the last decade? When would you have gotten back in? Chance are, you would have missed most of the tech bubble, and the crash, most of the real estate boom and the crash, and yet where would you be today?
In theory, you are correct on getting in and out. But the fact is, you would more than likely get out of the market too early, and back in too late, just like most clients.
Taking clients in an out of the market has been proven to underperform the markets. If it was that simple, every professional manager would do it. Markets are simply too unpredictable to try and forecast, and therefore it does clients a disservice to attempt it.
Again, this is all my opinion. I understand the attempt (and success at times) to find alpha for clients. I just think it plays into the emotions of the markets, and makes our clients part of the emotion trend. Our non-tactical allocation has gotten our clients 5% /year for 10 years, and 3%/year for 3 years (that was before yesterday ) so I believe what we are doing is working.
You are contrasting clients who wish to exit AFTER a crash, with the consequences of exiting BEFORE a POTENTIAL crash. It is a completely different scenario.
I could have gotten out in 1996 and been in cash for 15 years. I would have missed three different 50%+ rallies in equities. And because I would have also missed the crashes, I'D HAVE MORE MONEY BY MISSING EVERY SINGLE MAJOR RALLY FOR THE PAST 15 YEARS.
Markets are not unpredictable. SHORT TERM markets are unpredictable. If you're going to judge your client's financial success by trying to time the market a few months at a time, I fully agree it's a losing proposition.
But last time I checked, a client's financial planning time horizon is WAY longer than that. You are talking about ultra-short term market timing; I am not.
Market valuation would have told you a decade ago that your clients were going to get around 5%/year for a decade. It's not a surprise. The catch is that they could have bought a 10-year bond for more than 5% in 2000. And thereby had at least the same amount of money. With radically less volatility, uncertainty, and unpleasantness. Even if I didn't produce a dime of additional return, it was still a more stable, more certain outcome - and also one that actually does lead to higher wealth in a withdrawal scenario.
But the bottom line is, as you note: "Chance are, you would have missed most of the tech bubble, and the crash, most of the real estate boom and the crash, and yet where would you be today?" My response is "yup, and the client who avoided the tech bubble rally, the real estate bubble rally, and the post-crisis recovery rally, would STILL have comparable (or slightly more) wealth, and a less risky path with a tiny fraction of the volatility to have gotten there." Just run a cumulative return on intermediate government bonds since the late 1990s and see.
You point is correct in the sense that you are only looking at the US Equities market. You are still not taking into account being in a truly diversified portfolio, with only 1/3rd of a 70/30 being in equities, and 40% of that 1/3rd being in US equities.
Going back to 1996, our clients are closer to 7%/year, and that includes when Rick was still somewhat tactical which hurt his clients return (he will tell you he is awful at picking asset classes). And there is no way they would have been better off in the long run being in bonds/cash during that period.
I think it always comes back to client education. If they are constantly watching the markets, and get very fearful when they drop, then I wouldn't make the best investment advisor for them. If they are willing to ignore the media, not watch markets, and instead focus on what they can control in their financial plan, in the long term, that will be much better off financially.
The conversation is equally relevant whether you're talking about 100% in equities, or the 60% of your portfolio that was in equities. It was still a decision to hold 60% in equities in a grossly overvalued environment that was priced for the terrible returns we did in fact get, predictably, as we have in every instance in history.
I by no means am suggesting that anyone/everyone can do this. It does still require research and analysis. It's not a free lunch for zero effort. But there IS opportunity on the table if you take the time to look at the data.
And just because you encourage your clients not to look at the markets doesn't mean YOU shouldn't be. We've been navigating these markets effectively for a decade. In point of fact, I think our clients look at the markets even less than ever, because they actually trust that we're going to try to defend them from disaster. I find the clients who are most vigilant about the markets are the ones who actually think THEY have to do something to avoid disasters BECAUSE their advisors aren't doing it for them.
1. what specifically constitutes a "risky market"? When there is a lot of uncertainty? Because that's actually when the market is least risky. The most certain markets I can remember were in 2007 and before that 1999.
2. this all-in, all-out discussion misses the third and more pragmatic approach - modulating exposure by degrees, systematic rebalancing, averaging in over time rather than planting a flag at one specific market level.
I'm not a fan of the framework of "uncertainty" because I don't see a particularly good way to measure it. Some obvious choices might be tied to investor sentiment or something like the VIX, but at extremes the relationship tends to be more predictive as a contrarian signal, as you note.
I would see fundamental valuation factors as being the most predictive in terms of what constitutes a "risky market", especially in the context of individual investors whose risks focus more on capital impairment (especially during withdrawal phases because of the sequencing risk) and a long-term period of substandard returns, both of which bear some relationship to valuation. Individual planning clients generally don't invest with leverage (or at least, not using margin), so long-term factors with long-term implications remain relevant (even if their short-term predictive value is weak).
I'd certainly agree that it doesn't need to be (and probably shouldn't be) an all-in all-out approach. That wasn't intended to be my point (although I can see how it reads that way). Simple tilts are more than enough to be effective, and gradually implemented tilts through bad valuation environments reduce some of the challenges of trying to perfectly time execution. To me, that's actually the essence of a well-executed tactical portfolio strategy, and was what we tried to show in our Journal of Financial Planning article in 2011 (see http://www.fpanet.org/journal/currentissue/tableofcontents/improvingriskadjustedreturns/ where the optimal tactical threshold was basically +/- 20% from the benchmark).