Investment Risk and the Growth of Wealth: The Importance of Course Corrections
I have a problem. In fact, I think we all have a problem:
We have been way too focused on returns, resulting in the utter destruction of our wealth.
The investment industry has been built using tools that might be appropriate for understanding investments, but are totally worthless for investors. In real life, real people, using real money do not care about returns. We care about our wealth, and the two are not the same. They are not the same!
Risk and investment returns
Nowhere is this paradox more evident than the core tenet that risk goes down over time. When talking about investing, everyone says that risk goes down over time. Every investment book, every investment advisor, and for sure any class you took in college, all shout this idea from the rooftops, an idea illustrated in the chart below.
It’s time to back up and understand what this chart is really telling us.
It measures the best and worst periods in terms of annualized returns from 1926 through March 2009 for the S&P 500, with the average return for the entire period in grey.
For example, while that average return for the entire period is 11.28%, you had a one year period that was positive 162% or negative 67%, a three year period that ranged from up 43% per year to down to 42% per year and so on. As you measure longer and longer periods, the best and the worst start to narrow around this long-term average of 11%.
Just to make sure we are clear before we move on:
- This is a chart of the range between best and worst annualized rates of return for different periods of time.
- It clearly shows that the range of potential outcomes (in % terms) narrows over time.
Somewhere along the line, someone locked in a back room at a university decided that the range of returns (measured in percentages) was a good proxy for risk. Somehow it is was useful to measure risk in terms of how much something goes up and down in percentage terms.
This is where things get sticky.
Risk and the growth of wealth
It might be true that “risk” goes down over time, if you define risk this way. If you are an investment, this would be a good way to measure risk. But you are not an investment — you are an investor. Real people do not care about their return. We care about our money. We don’t eat percentage pie. You can’t pay your kids’ tuition with percentages; you pay with dollars. Everyone I have ever met thinks of risk as the possibility that they will run out of money.
So the question is not about the accuracy of this chart, but about the conclusion that we draw from it. When you are a real person, measuring risk in terms of returns is like trying to measure temperature with a ruler. Fine tool — wrong application!
Investors measure risk in terms of dollars; investments measure risk in terms of percentages.
When we measure risk in terms of wealth, risk goes way up over time!
The chart looks like this:
This chart shows what happens to a dollar if we invested during those best and worst time periods from chart one. Depending on the sheer luck of timing, $1 could have ended up growing to $62.52 — or just $9.51 thirty years later. That is a huge range of potential outcomes. If you add in withdrawals, things get even more “risky”.
This should make sense. It’s basic math, compounding over time. When you leave on a trip from Los Angeles to Miami, if you are two inches off course, you most likely would not even notice when you are over Nevada, but without any course correction you will end up in Maine! Maine is not a bad place — unless you planned to be in Florida.
The issue is that most of us are on auto-pilot, thinking that if we just “set the course and forget it” that our risk will go down over time.
That is simply not that case.
So what?
I’m still working through all the implications of this, but they at least include the following:
- The timing of major financial decisions has huge impact on our success. If you happened to retire with a lump sum at the beginning of one of the best 30-year periods, your experience would be dramatically better than if you retired just before one of the worst. The problem is that we really don’t know which one we are in until after the fact.
- Average is not normal. If you set a course based on some version of a long-term average return, then you are likely to end up somewhere else. Most of the time, the surprise destination appears to be good, but sometimes it’s bad. In fact, it’s bad often enough to make me question some of my long-held beliefs.
- Making course corrections is at least as important as setting the course!
- Rates of return are only one factor in meeting your long-term goals. There are other levers to pull when we need to adjust, other levers that we have much more control over, including retirement date, savings goals, withdrawals, etc.
- We really can’t rely on the idea that if you buy and hold, your risk goes down. While that may still be the best approach to investing, in terms of meeting your financial goals, there are a lot of other things that you need to think about as you go forward.
The point of this article is really just to raise awareness. I have been thinking about this for a long time. I have been in the financial-planning industry for over 10 years. I really can’t think of one concept that is more important to understand right now. Here’s a video that summarizes my thoughts on this subject:
I want to leave this open-ended. This is the beginning of a discussion, not the end. However, it’s important to understand that the process of planning for your future remains the same:
- Determine where you are today.
- Define the destination.
- Chart a course to get there.
But now that you know your risk of not getting there is actually growing over time, make sure you take the time to do regular course corrections. Are you on course to meet your goals? Have you made corrections in the past — or do you expect to do so in the future? How do you feel about the relationship between wealth and risk?
Become A Money Boss And Join 15,000 Others
Subscribe to the GRS Insider (FREE) and we’ll give you a copy of the Money Boss Manifesto (also FREE)
There are 63 comments to "Investment Risk and the Growth of Wealth: The Importance of Course Corrections".
“We don’t eat percentage pie.” Indeed!
I created a ppt video about this concept a few months back–showing that the possible range of ending values widens as time horizon increases. Somebody even commented to “correct” me, mentioning that returns become more predictable as time goes by. Apparently it’s quite drilled into our heads.
While I’m obviously not a fan of checking one’s portfolio too frequently, this (along with rebalancing) is a reason why we need to do it at least periodically.
I’m a bit confused by this post. Whilst the recommendations it makes are good ones, it’s not like people don’t already know this. It seems odd to shout from the rooftops about the dangers of ‘autopilot’ as if it’s some newly-discovered threat, when every PF blog and book is hoarse from repeating the mantras of asset allocation, regular rebalancing, and dollar-cost averaging. Is it not true that those three concepts address all 5 of this post’s conclusions?
I think the key to this is not so much course correction as it is dollar cost averaging.
Most of us don’t approach investment only with a single lump sum. Instead, we have some savings to begin with (this is the part that might have earned as little as $9.51), and some monthly and yearly amounts that we are able to budget and add to our savings. By virtue of dollar cost averaging, these monthly and yearly amounts will often do much better for us than the dollar that only made $9.51. I think this is a much stronger answer to this supposed problem than course correction is.
Where course correction comes in is via rebalancing. That is important, too, but I don’t think it is as much the key to these problems as dollar cost averaging is.
Also, keep in mind that $1 turning into $9.51 over 30 years is not too shabby to begin with. That’s an average 7.8% return per year.
In general I think the tone of this post isn’t warranted when you take a careful look at the supposed problems it claims.
i think the average is misleading, because it is assuming something about the distribution of 30-year trajectories. i don’t know how the average here was calculated, but i assume it summed all N 30 year returns and divided by N. this implicitly assumes that all N trajectories are equally probable.
i’d guess that is unlikely. if you happened to end your 30 year run at the height of the stock bubble in 2000, you cashed out with a big chunk of money. but that is very unlikely to happen, so it shouldn’t be influencing your average upwards.
Good morning, JD. This is a topic which I’m interested in, I’m glad you’re starting to cover investment topics, and I made my attempt to work through the article. However it is very disorganized, and very confusing, difficult to follow, too technical, and boring. The gap in quality in between the guest posters and your own posts is stunning. Maybe you could rewrite this post in your own words and tone, and repost it, so the rest of us can follow it and enjoy. I don’t think most people who visit your blog come because of the depth information presented, but instead because of your engaging tone, and well-thought out posts, and almost all of the guest posts fail to live up to the standard I expect from your blog. I would much rather see you cover these topics because you are the blogger who I can trust to make things interesting and transparent.
This is an interesting post. However, I am not sure that I agree with your definition of ‘risk.’ I have two major objections, actually.
1) Based on the second chart, it seems that you define risk as “not obtaining the maximum increase in value” or perhaps “not obtaining the projected average increase in value”? I think many people define risk as “losing money” and in this context, risk *does* decrease with time invested. You can see in your second figure that the worst case scenario results in negative values initially, but eventually gives positive final values (ie. no loss in money). Of course, this is what would be expected from the traditional definition of risk (ie. fluctuation in return rates), which brings me to my second point.
2) Both graphs show the same thing — the exact same amount of risk. They just have two different ways of displaying it. You know this is the case, because in order to generate the second graph you used entirely only data from the first graph. So, you actually cannot decrease or increase the associated risk and/or spread. What you have (mathematically) done is to effect a change in variables that appears to spread things out more, but in reality, since it is a simple transform from one (rate of return) to the other (final value) you have not fundamentally changed anything.
Having said all of this, it think that thinking in terms of total value is a good way to go, since (you are right) this is what people really care about. It is nice way to emphasize that you probably will not get the exact average return on a fund (unless you wait infinite years — but who has that much free time). In the end, however, I feel like the main point should be that given a positive average rate of return, if you wait long enough, you will not have lost money (though you may not have made much). I think this is a better definition of risk (based on actual, realized, net losses).
For what it’s worth, I quite appreciate a more technical post (guest or otherwise) from time to time. And I wouldn’t exactly call this article “very confusing.”
I think the 11% average is skewed as well. While the average return might be 11%, it does not quite translate to dollars.
Example: I have 100 dollars
First Year: + 20%
Second Year: – 15%
Third Year: + 10%
Fourth Year: – 20%
Fifth Year: + 15%
Average Percent Return = (20 – 15 + 10 – 20 + 15)/5
= 2% Average return
Now lets put the original 100 dollars in play and see what we end up with
First year (100*1.2) = $120
Second Year (120*.85) = $102
Third Year (102*1.1) = $112.2
Fourth Year (112.2*.8) = $89.76
Fifth Year (89.76*1.15) = $103.22
After 5 years, 103.22 is only 3.2% more than the original value.
Most places just assume its compounding 2% every year for 5 years and you would come up with this: $110.40
Real value $103.22
Assumed average return value $110.40
Point is, even though it averages at 11%, you are more likely not going to make near that amount in the end. Diversify investments, don’t put it all into the stock market!
I think this is mostly just common sense. But it also forgets about something: goals change.
Sure, some long-term goals stay the same but those are usually universal. For instance, we all want to be able retire some day. While the details differ slightly from person to person it basically means we need to save for retirement.
Now other goals change widely. Someone may have gone to school to become a teacher and then decide they hate kids. Or someone might have a baby and decide to quit their career to be a stay at home parent.
While course corrections are a good idea overall, they tend to come more when you change your goals and less as you are trying to accomplish the same goal.
Risk really is mitigated simply through diversification. Sure, you need to rebalance every so often, but if you are truly diversified, a buy and hold strategy across multiple avenues does give minimal risk. The other thing to help minimize risk in a down economy is to plan for the worst. If you plan your retirement for a down economy, anything better is just extra icing on the cake.
I’m no expert, but I think you’re wrong.
Yes, over the long term we see a greater spread between the best and worst possible outcomes. But even the worst possible scenario ends up giving them 9.51:1 (not a great rate of return). But consider the worst possible rate of return over one year -67%.
It’s the opposite side of the coin of whether to invest in one company or 40. You are hedging your bets in both instances–in effect reducing your risk by playing the odds.
Or to look at it another way, consider the thought experiment of the gambler’s ruin. That game suggests that no matter how much you are up, if you’re playing against house odds, in the long run you will end up with $0. It’s the opposite with investments (admittedly, not all but that’s what funds and indexes are for).
Over the long term, odds are in the investor’s favor. Perhaps they won’t make 163% every year, but as you’ve shown us, they should see an average return of 11.28%.
Interesting post. I know from my experience that someone just starting out in investing doesn’t have much risk because they don’t have much invested. You are far better off having a bad year (-40%) if you young because that 40% might only be $4k or $40k. Later on that 40% might be $400k if you are 100% equities.
I think asset allocation has to be part of the discussion as well – I’m not a believer in being too conservative but if you get a point in life where you have a lot of money invested and a big loss will set you back – then lower your equities allocation. You can manage your own risk
@Scott – Dollar cost averaging becomes a lot less relevant as your portfolio grows. If you annual contributions are only 5% of the portfolio then they just aren’t as significant.
@Frugal Bachelor – If you could harness all the energy you expend complaining about posts on this blog and in the blogosphere in general and write something useful, then maybe you will be part of the solution to the problem you continually complain about.
One other thing to add – the idea that the risk goes down over long periods of time is true. But for equities – that “lower risk” area is always in the future. Someone who starts investing in equities when they are 25 might have a 40 year time horizon until retirement. Any money that they invest when they are 25 has a reasonable chance of getting a decent return over the next 40 years.
But what about when that person is 55? At that point it’s only 10 years to retirement and who knows what the 10 year return will be. The fact that they have been investing for 30 years is irrelevant at this point. I think people who got “burned” in last year’s equity meltdown either couldn’t handle the volatility of their portfolio or they were at a point in life (ie 55 with a big portfolio) where they should have had a lower equity allocation.
I found the article interesting since it delved into a topic not generally covered in this blog. I do think that the author is a little too alarmist about his conclusions, which have issues that other posters already pointed out. I do agree with a major underlying point that Carl has however, the average return of the stock market (or bonds or whatever) will probably NOT be achieved by any of us and thus shouldn’t be the main factor in our decision-making. The reasons are many: past performance does not guarantee future results, dollar cost averaging, time of entrance in the market, time of exit from the market, asset allocation, rebalancing, the list goes on and on.
The main point is that stocks, due to their inherently more volatile and risky nature, provide a better long-term return than other asset classes because they HAVE TO – if they didn’t no one would deal with their risk. Everyone would be investing in bonds, CDs or a simple savings account instead. So, stocks should be the main engine of growth for most people, but they need to be tempered with bonds, cash and real estate. The appropriate percentages of each are up to the average investor, his/her goals and his/her risk tolerance.
It seems to me that several of the commenters are still thinking in terms of investment risk when what the article is getting at is that there are two different but equally important types of risk, investment risk and financal plan risk. It is true that the rate of return will likely start to approach the average after X years. But in order to reach your goals, as you get closer to the end of your timeframe, you can’t have a bad year or bad 10 years. If you do it’s much harder to reach your goals. Say your one year away from your goal of retirement. First of all you won’t cash out everything because you want it to continue to grow somewhat during retirement more than likely. Second, you are back to the one year side of that first graph where you could have 160%+ return or -67% return. That is a big risk. Looking in a larger time span, each year has the risk of being great or horrible and that is the risk that grows. Over 10 years it is unlikely that each on will be horrible but you could still end down over 10 years.
The point being that the longer out you try to forcast, the less clear the picture is. Think weather. I know that January in Williamsburg is usually cold I can’t tell you what the temp will be in any given day of January 2010 much less 2050. I know what the agerage is from however long they have kept those records here so I could probably get reasonably close but I don’t know.
@Scott: ABCs is probably right (It makes sense but I’m just beginning my investing so I have no clue). Also 7.8% is great but if your goal requires that you make the average of 11.whatever%, then 7.8 % sucks. Which brings WiseMoney’s comment into perspective. Goals do change which is part of a course correction, at least from what I understood from the article and video.
To break it down again (and I know i’m sort of rambling): For your portfolio today, there is a big chance that it’ll either tank or skyrocket. For your portfolio 30 years from now, there rate of return will probably be around 8-10%-ish. For the goal of retiring in say 30 years, the chances of you makeing it are slim as of today. In 29.9 years, if you’ve kept on top of your goal to retire in 30 years, you’ll likely know exactly what day you can retire. As time increase your average rate of return will approach the average. As time increases the likelyhood that you’ll make your goals decreases.
Carl,
This is one of the clearest, most direct explanations of what the myth/fallacy of time diversification. Great job!
But, it’s your same point that many investment advisers actually have been trying to shout from the rooftops. I even detect some of the louder ones’ influence in your post, like Dave Loeper.
For those that don’t see the risk that Carl is referring to, I don’t think the point is that it’s still a positive return or that it’s not the maximum. The risk is the very real chance of not having enough of the dollars you’ll need in the future or having too many (which means you sacrificed more than you had to along the way, and that’s a risk too). To see the risk you must think in terms of dollars, not just what you have, but also how you got it and what you need.
A few points of clarification:
[1] RISK is not the range of one year returns. IT IS the possibility that you will not have enough money to meet your goals.
[2] Making course corrections to your financial plan is not the that same thing as asset allocation & rebalancing. THAT WILL NOT SOLVE the problem you have if your equity investments earn 2% per year for 20 years when you had planned on them earning close to the long term average of 11%. At that point your plan will be grossly underfunded and you WILL have to make some painful choices EVEN if you had rebalanced perfectly (whatever the means).
To sit around a assume that if you just hold on, just rebalance, you will end up meeting your goals is CRAZY. If you are 300 miles off course, you can’t just continue to drive the same direction. You have to make changes. Not to your investments, but the other “levers” of your plan (save more, delay your goal, adjust your goal, etc…)
[3] While I can appreciate that this is too complex for some. I would argue those that say it is too simple are missing the point. YES the math is simple but the implications are huge and largely ignored.
[4] If I am driving 60 miles an hour down a road, and 60 miles ahead there is a bridge out there is a lot I can do with the next hour. I can change direction, slow down, stop for lunch while they fix the bridge. If I am driving 60 miles an hour and find out about the bridge with a 1/2 mile to go: I AM IN TROUBLE.
There are some comments that suggest that the guest poster is at fault for presenting an old topic in a new way. Isn’t that what most of the information found in PF blogs, articles and news does? While I appreciate constructive and appropriate criticism of information, (I think that is why the information here is generally so good, people who share here know they are subject to scrutiny) I think some people only take in information as critics, not as someone truly willing to learn from others. If people are checking in each day waiting to see something completely new and revolutionary, I’m afraid you’ll be disappointed.
When I come to this site I don’t expect to find something so profound that I am taken back and awed by the information. I come here to be reminded of the values I have in my own finances and look for different ways to apply the information shared.
Thank you Carl for the food for thought.
An excellent, thought provoking article Carl.
This helps to show that we are the biggest investment mistake in our wealth equation.
I have written extensively about this on my blog using calculators at moneychimp.com In November of 2008 Warren Buffett wrote that he was “buying American” stocks. The impressive thing was not that he was buying stocks, but the more subtle aspect of the article, Warren Buffett’s time horizon is infinite! The only way that risk diminishes over time vis a vis stocks is through an infinite time horizon. Because individual investors ALWAYS have a time horizon, market timing remains an aspect of investing and risk, therefore cannot diminish over time.
@Chett @17.
I agree with you. If I hear info again, since I – and my thoughts and my situation – change, the info may strike me a different (and perhaps for the first time, useful) way. Repeated – if well articulated – info welcome here. 🙂
You can think of risk so many ways. You may set off on foot for a restaurant you can see several blocks away (retirement), but if there’s a hole in your pocket (inflation) or someone robs you (Madoff) you may not be able to pay the menu prices. But if you put some of your money in each of your pockets and socks (diversification) and maybe have some silver in a hidden pocket (commodities), you can reduce the risk. Silly, but just another way to think of risk.
So, if you save twice as much, have you reduced your risk by half? By saving twice as much, you can assume half the interest rate (3.9% vs. 7.8%), but that seems to me as having the same risk – but twice the money at risk. Thoughts?
While Carl does have some argument to stand on, He’s very preachy and seems to miss the point. He throws the baby out with the bathwater when inventing his own definition of risk. He’s very quick to toss out decades of study when it dosen’t align with his current argument.
Risk *is* variability of returns. What Carl is talking about is how to choose how much risk to take on.
@Brian,
He’s not redefining risk, nor is this his invention, nor is he tossing out decades of research.
He’s suggesting that it may be more appropriate to consider a different definition of risk.
Risk is the very real chance that thinks won’t go the way you intended. That could be volatility. It could also be bad timing.
It’s not about “how much” risk to take on, it’s about knowing what you are risking.
@Brian: Thanks for the feedback. It is all very valuable.
This is not MY definition of risk, nor am I talking about “how much risk to take on”.
What I am saying that measuring FINANCIAL PLANNING risk using the tool that was designed for measuring INVESTMENT RISK is like measuring temperature with a ruler. It just doesn’t work.
Try telling a 60 year that was hoping to retire this year that risk goes down over time. They have been saving for 35 years and just happen, out of pure luck (bad), to have the last decade of below average returns come at the wrong time.
The problem is not in understanding this concept, it is in the implications of it. It means that we need to understand that there are 2 very different types of risk that lead to two very different types of behavior.
@Carl
You seem to be getting pretty defensive, without correcting the flaws that have been pointed out. Your example of a 60-year-old retiring this year most certainly is addressed with asset-allocation – someone on the verge of retirement should be mostly in bonds rather than equities, and would therefore be relatively unaffected by below average returns.
And ranting about “the problem you have if your equity investments earn 2% per year for 20 years when you had planned on them earning close to the long term average of 11%” is based on a false premise. Unless you have a time machine you can’t predict the return on ANY investment over that period of time. Even a 20 year bond is unpredictable because you don’t know what the inflation rate will be. Risk is inherent and you can’t avoid it. “If I am driving 60 miles an hour down a road, and 60 miles ahead there is a bridge out there is a lot I can do with the next hour.” is NOT a valid metaphor, because when investing you cannot possibly know whether or not there is a bridge out in 60 miles, so you cannot possibly make any concrete decision about what to do. All you can do is try to cover all bases – i.e. diversify, allocate assets differently as you age, and rebalance.
Carl – I am no investment expert (believe me, I’m barely a beginner), but isn’t it true that a 60-year-old should probably be holding more stable assets if they are planning on retiring soon?
I was under the impression that the closer you get to retirement, the less money you would have invested in the market. This is what most experts recommend, if I am not mistaken. If this were true and a 60-year-old investor had their assets tied up in bonds and CDs, etc., then their retirement may only be mildly affected by the current economy. I don’t know of anyone who recommends that someone aiming to retire soon should throw all their funds into stocks.
Please correct me if I’m wrong. I would really like to know if this is the case. Thanks!
Wouldn’t the absolute worst case be someone who was excessively into stocks up to 2008, and now is excessively into long-term bonds just in time to get creamed by rising interest rates (if you have to sell a bond, capital losses) and inflation (value of interest payments worth less in future)? First, you lose 30-50% in stocks, then much the same happens to you in bonds. To avoid this, you have to accept low interest on shorter term bonds now. So now we have to save not twice as much, but four times as much because short term interest rates are lower than 3.9% on US Treasuries.
Ok, everyone ramp up their saving to 40% of income. Everyone good?
I’m watching the back an forth between Carl and some of the commenters, and it is not unlike one person saying, “that sports car is red,” and then the response is, “no, that car is fast.”
The points about using asset allocation are fine points, but even if you are 50%/50% stocks and bonds, 25%/75%, or any other allocation, the range of potential outcomes increases with time. Adding less volatile asset classes may slow the widening of the range, but there is still an increasing risk as time passes that ought to be monitored. So the car can be both, fast and red.
I think Carl chose the S&P 500 because it is an easy subject for contrasting the different risk measures.
@Russ, the “bridge out in 60 miles” is not about what the investments will return for you. You’re absolutely right that you cannot foresee that. But you can see whether the likelihood of being able to afford to fund a future goal is fading. You are not avoiding risk; you are measuring it, just not in terms of investment volatility or average return-to-date.
@Russ: We’re on the same page.
@Dylan: 60 miles down the road, the bridge may be out, but you can’t know that until you get there and taking another route when this route is shortest is foolish. That bridge being out is unlucky, but that doesn’t mean your strategy is wrong.
Similarly, the 60 year old who has gotten a 2%ROR over the last 20 years has gotten a crap string of hands, but that doesn’t mean he’s played the game wrong. And of course this is why orthodox financial planners will transition a person’s assets into bonds as they approach retirement.
@Dylan
I was specifically addressing the part of the metaphor that said “there is a lot I can do with the next hour”. I dispute that. If I look at my investments and figure I’m not hitting my 8% a year target, what can I do in ‘the next hour’? Switch to bonds? Cash? Gold? (Alright, that last one’s a joke.) None of these work, because I have no idea how any of those assets are going to perform *either*.
In actual fact, since I can’t *know* that the bridge is out, my best course of action would be to continue since statistically speaking the chances are that the bridge isn’t out at all.
My best course of action is, as always – diversify, evolve my asset allocation, and rebalance. The risk analysis in this article, absent a time machine, gets me nothing that isn’t covered by those three pillars.
Dylan, I think you summed it up pretty well there. It appears to be two entirely separate conversations.
Carl’s post did not argue against intelligent asset allocation, diversification, or rebalancing.
The point of the post is simply that when we say that “stock returns become more predictable over time,” it’s a bit misleading because we’re only looking at percentage returns. When we look at total return as measured in dollars, the range of possible values increases over time.
As far as I can tell, all Carl was saying was that we need to recognize that fact and plan accordingly. Seems like a good point to me…
The stock market is a Ponzi scheme. The CEO’s are at the top and investors are at the bottom. Sorry to be so cynical but sometimes it’s necessary.
Carl –
This is your best summary of this information to date. Keep working through this paradigm because it is important. Risk is officially defined as “deviation from an expected result.” Somehow the general understanding of this has morphed into the garbage that people now hold dearly as gospel truth.
Average is NOT normal – so why do studies insist on focusing on average returns. While I don’t believe in Monte Carlos (mostly because they use bell-shaped return assumptions rather than fat-tails), a Monte Carlo clearly shows that risk (deviation from expected results) increases over time. A fat-tail model would only deliver even wilder results.
The ultimate distinction – Wealth Building results from good financial planning – not just investment planning. That is not a reality that financial salespeople in the investment houses want their clients to hear.
It’s hard to validate charging a fee for advice when that advice is limited to only investments – which have such wild results in both the short and long terms. Since there are plenty of more powerful ways to build wealth than just investments, good financial planning advice CAN take risk off the table, though – and THAT validates a fee being charged.
Ditto to @ObliviousInvestor
@Russ – I think you have a better understanding of how investment risk can alter the timing and achievement of your goals than most.
I think that is Carl’s point. A lot of people don’t see the need to either alter their goals or alter their investment style to fit those new goals. I take it that in the course of diversifying, evolving your asset allocation, and rebalancing, you are doing so with respect to your goals and objectives.
I don’t think that Carl’s article is a knock against any of those principles, just a reminder that those principles need to be applied with your ever-changing goals in mind.
Great discussion! This issue is really important to understand so having such thoughtful comments and questions is helpful.
@Elizabeth [25] Thanks for the question. Most financial planners would agree that the as you get older [or closer to a goal] that you will reduce your exposure to equities, BUT you will still have something in equities the swings will still effect you. Which means that you need to be paying attention and making course corrections [save more, spend less, retire later, etc]. The sooner you are aware that you are off course and headed for a cliff the more effective you financial planning course corrections will be.
@Dylan [27] Thanks
@ObliviousInvestor: Thanks for the added clarification. It really is a simple point, but it seems to get really confusing because we are not used to separating “investment risk” from “planning risk”. Or % risk from $ risk.
@Ryan & Russ: I think that you are looking at this from the investment perspective. You are correct that you have no way of knowing if the bridge is out in terms of the future investment returns.
BUT we can know if we are off track to hit our goals in terms of dollars. If we built a plan that counted on a 8% return and we got a 1% return for 3 years it will be clear that we are off course. We will have less $ then we projected at that point.
This does not mean we did something wrong. It just means that the % return we had planned on did not happen. We should now going into this that we are not going to get the exact return in % terms as we plan on. Returns are variable.
BUT now we are slightly off course. We need to make some changes. If we don’t this little detour could derail the entire plan. This does not mean we change our investments. Like Paul [26] and Russ [29] said changing the investments is NOT the right thing to do. But we have to make changes to other levers that go into the plan. We could save more, delay the goal, lower the spending goal, etc…
The point is, if we get a string of “bad returns” we need to be paying attention. If we look at our plan often we will know when we are off course and headed for a bridge that is out. If we catch it early we can make changes that we have control over and hopefully avoid the cliff in the future.
Hope that helps.
Dylan & Obilvious are right about the point Carl is trying to get across. Russ et al are correct as well that the three basic pillars help mitigate the risk of rate of return being below average.
Maybe metaphors should come with a key, a legend just like we used to put on the maps and graphs we made in elementary school.
@Russ and Ryan,
I think the disconnect here is that the target is not meant to be a rate of return (another investment measure, not an investor measure. The correct measure should be a probability of a least meeting a future cash flow need.
Let’s look at the guy with the crappy returns and say he want’s to retire a 60 and spend $50,000 per year before inflation. He’s 40 now, and as each year of crappy returns occur, he gets further from that goal (the bridge is going out).
We can measure this, because at age 45 (for example) we can see if his current assets and expected savings are reasonable to achieve his goals in 15 year. If poor returns have him off course and it’s no longer reasonable to plan to achieve the same goal, knowing this, he has time to make changes like save more, delay retirement, lower his spending goal. This is the stuff you can do in that “hour” (in some cases it may even make sense to change the asset allocation).
He has this time because he is not focusing on a rate of return but rater a likelihood that his current financial situation (a result, in part, of past investment returns) is one that can achieve his goal with a reasonable chance of success.
If you accept the idea that this guy’s risk of missing his goal goes down over time, the solution would be to hang in there and do nothing and hope returns are that much better to make up the difference (or won’t be crappy again in the years right before he turns 60). But, if you recognize that his risk (not volatility) of missing his goal increases with time, you can plan for the possibility of bad luck, so you don’t have to make those major course corrections along the way.
I’m not saying you have to plan for a worst case scenario, but planning for average means coin flip odds there will be bridges out and course corrections along the way.
Financial planners transition assets to more conservative allocations over time when the more aggressive allocations no longer would result in a reasonable probability of success. This is often a natural result of planning for increased risk with increased time.
Wow, some of the comments make me wonder if I’m reading the same post that you are 🙂
I would like to let some more comments trickle in before sharing my thoughts, but bottom line is that I think Carl raises an important point and does a good job of explaining why it’s important.
Also, if you’d like to hear Carl talk about this issue further in a conversation with Bill Schuletheis, check out their most recent episode of BehaviorGap Radio:
http://www.behaviorgapradio.com/
@ObliviousInvestor
“the range of possible values increases over time”
I agree that this is true, however, it does ignore the fact that the probablity of getting a total return close to the projected average *increases* with time.
Think of flipping a coin. Each time you flip it, there is a 50% chance of heads and a 50% chance of tails.
If I flip the coin once, then there is a narrow distribution of possible outcomes, either 1 head or 1 tail.
If I flip the coin twice, then the possible outcomes have widened, now it is either 2 heads, 2 tails, or 1 head and 1 tail. So there is three possible final outcomes, with the last one (one head, on tail) being the most probable (in a 2:1:1 ratio).
If I flip the coin 100 times, then there are LOTS of possible outcomes (anywhere from all heads to all tails), BUT it is most *probable* that I will arrive at an outcome close to 50 heads and 50 tails. If I were to do this experiemnt (100 coin flips) 1000 times, then we would see a clustering of outcomes near this 50/50 mix.
The message is clear; as I increase the number of times that I flip the coin, then I will be widening the possible outcomes, BUT narrowing the probablility of achieving close to the ‘average’ (read: most probable) outcome.
Likewise, in the post, Carl argues that the possible distributions of final value widens as time goes on — which is true. However, he fails to account for the fact that the distributions of final values will be weighted such that (as time goes on) one is more likely to acheive a value near the average. This is the main problem (I feel) with his post. It makes it look like there is more ‘risk’ than there actually is.
Again, I think that it seems like a useful way of thinking about finances (as it will induce caution), but I am not sure that it is technically correct in the manner that he thinks it is. Of course, I suppose that I could be wrong too.
Anyway, i just thought that I would use your post as a jumping off point, since I always seem to like your comments. I hope you do not think that I am attacking you in any way.
Deviation in returns is pretty interesting stuff, actually, and I will be taking a look at it over at my blog (starting today) over the comming weeks.
So, it seems general opinion is that Person A needs to sit on some chunk of risky (good-risky, not bad-risky) stock to get their happy % they need to retire. BUT if they hit their retirement date still holding too large a % of this risky stock at the same time the curve is way down then they are screwed. Hmm. So Person A needs to try to avoid this so they put less in the risky stock 5 years before they retire, but now their happy % isn’t quite so happy so they need to work another year or two? Confusing! This whole thing seems awfully like gambling, and far too complex and messy for people who are just trying to save money and eventually retire.
Wow, excellent post Carl. Several folks seem to really get what you are talking about.
Put simply, a real financial plan starts with where you are today, establishes an endpoint [$50K/year for 25 years for example] and then fills in the process of getting there. If you find that you need to get a better return in order to get to your goal then you need to adjust your investments. Or maybe you need to save/invest more [assuming that is possible]. Bottom line, is that is what a financial plan is all about, making changes as the reality of your situation exposes itself!
Carl’s always worth reading and the graph juxtaposition is cute, but as others have said this post isn’t actually very relevant in the way it claims to be to real-life investing.
In reality, as a private investor you’d be successively investing into different 30 year periods.
(Or more precisely one 30 year period, one 29 year period, one 28 year period, etc, assuming you were aiming for a retirement date and the total conversion of your investments after 30 years into some form of fixed income).
Dollars invested in one year might be going into a good 30-year period, and dollars invested in another would be going into a different multi-year period.
Over time you’d expect an ‘average’ result (without quibbling over what exactly average means here).
Perhaps Carl’s issue is more commonly encountered by finance professionals – I believe he is one – who are faced from time to time with investing big lump sums from individuals who have suddenly come into wealth or decided to do something with that bulge under the mattress.
But for most of us, this particular issue isn’t an issue.
Monevator said: “Perhaps Carl’s issue is more commonly encountered by finance professionals – I believe he is one – who are faced from time to time with investing big lump sums from individuals who have suddenly come into wealth or decided to do something with that bulge under the mattress. But for most of us, this particular issue isn’t an issue.”
For most of us this is an issue. As cash flows are introduced (savings now, withdrawals later), the problem compounds. Time horizon is not the start retirement, even if you invest only in fixed income in retirement. Time horizon is the end of retirement.
When you introduce cash flows, how you arrive at your average returns plays a bigger role than what your average is, even if it is “average.”
@ Carl
Nice post. I’ll remind folks your closing advice:
Determine where you are today.
Define the destination.
Chart a course to get there.
The premise of your article actually points to the need for asset allocation and understanding unforseen obsticles along the way. The sports car metaphor was actually the best summary of this back and forth dialogue.
@ Carl and Dylan – Thanks for your comments, they make a lot of sense.
I am a young investor (and I have little to invest at this point). And I consider the future fairly volatile as to what my rate of return may become over the years (as well as many other variable factors). My basic plan is to save responsibly and diversify my investments. I have no idea what will happen over the next 40 years to influence my goals.
I think the point you’re making is extremely valid: whatever your goals and whatever your time line, it’s crucial to keep an eye on what’s happening in your personal financial world so you can make changes as necessary. Socking away savings and blindly expecting a certain amount of return will do you no good if you check your accounts in 20 years and you have half of what you planned.
In part this is the result of looking at terminal values. A better approach would be to integrate over both accumulation and discumulation phases which would be both more natural and blur the results over longer periods. The best terminal value likely occurs just before a crash and the worst terminal value likely occurs just before an upswing so once one blurs over this extended period the end result is likely more similar.
“stocks, due to their inherently more volatile and risky nature, provide a better long-term return than other asset classes because they HAVE TO”
No, they don’t “HAVE TO”. It is only necessary that people believe that they will.
What is true, is that the less people believe stocks will return, or the higher they believe the risk is, the less other investments will have to pay to get some of those investment dollars.
Ultimately, the reason the stock market goes up over time is the same reason our lives get materially better over time. Companies grow and become more productive and wealthier and the stockholders who own them share in that increased wealth.
But that is hardly guaranteed. Nor is it guaranteed that the companies one buys now will prosper even if the world’s economy does. Its perfectly possible other companies will grow up and the companies you “own” will whither an die. Just look at American auto makers.
And that brings us to the distinction between “risk” an simple volatility. Timing, needing money when the market is down for instance or buying at one of its peaks, is a risk. But it is not the only one. The market may never reach the levels of last fall again. At least not in our lifetime.
Absolutely fantastic post. My thanks both to the author and to J.D. for hosting it.
I have never agreed with the conventional advice that young people don’t need to worry so much about suffering big losses in the stock market. Young people should be scared to death about suffering big losses.
Why?
Young people have the most to gain from the compounding returns phenomenon. When they suffer losses, compounding appliesin reverse. Young people get absolutely killed by being invested in stocks at times of high valuations.
If we thought about investors as humans instead of as numbers, this would be obvious to all. We would set up the investing question as — What should a young person be concerned about? He or she should be concerned about getting a good start on the wealth-building project. You don’t do that by suffering big losses.
The other way of looking at the same question is to consider the conventional advice for retirees — to lower their stock allocations. Does that really make sense?
No. There is no evidence in the record that stocks have even turned out to be a bad choice for retirees when the stocks were purchased at reasonable prices. It is only when retirees buy overpriced stocks that they suffer busted retirements. So why not warn both young people and retirees to protect themselves from big losses (by focusing on valuations) and not worry so much about what the silly numbers in the silly books say? We invest for a purpose — to live more fulfilled lives. For the humans, it’s the returns obtained that matter most, not the numbers used in some theoretical mumbo-jumbo academic exercise.
I believe that a big part of the problem is that it is only a few personalty types that are generally attracted to do work in this field. It tends to be the economists and engineers and accountants who are described as “experts.” But they are truly “expert” only re a small number of the factors that affect stock investing in the real world. We need more investing experts with the skill sets of psychologists and novelists and journalists and detectives. The numbers stuff matters. But it has been wildly overdone during the Passive Investing Era.
Rob
I think this is a very worthwhile article and addressed thoughts I have been having that I haven’t been able to articulate in such a clear fashion. Everyone talks about returns, but what everyone really cares about are dollars (and what equivalent resources one can buy with those dollars). I also agree that while stock returns may revert to an average over 30 years, depending on the timing of those up and down returns two investments with the same “average” return can have dramatically different amounts of money at the 30 year mark. Everyone will have different amounts of reactions to this. Myself it makes me realize I’m more of a bird in the hand than “potential profits are limitless” gal. While not mathematically optimal but more palatable psychologically I started diverting 5% of my retirement money to government funds. Also after having my e-fund in place instead of increasing my contribution to retirement I’m going to use that money to pay down my mortage early.
What a great discussion!
@ Carl
I’ve been following the discussion and wondering “does it make sense to consider the investment risk (as presented in the first graph) inversely?”
I mean by this that we are safe to assume our long run average return is the average in the chart when initially placing our money into the market, but as our time horizon closes we look to the ever widening variance and determine what percentage of exposure to that variance we want to have in our portfolio (given our plan.)
I realize this in no way contradicts your prescriptions (correct course, pay attention – enough, etc.) but I’m trying to take part of the ‘correct course along the way’ strategy (that mitigates planning risk) a step further and identify how we can correct the course of retirement investment along the way given the investment risk identified above.
So essentially, I’m curious if the 1 year variance given in chart 1 is actually more useful information for considerations one year out from the time horizon of a given investment (and subsequently 3 year variance for 3 years prior, etc.) Not that this information should be only considered at THAT point in time, but rather that it explains why we shift to bonds as we get older. Secondly, isn’t this a way that the mitigation of planning risk helps with investing risk, namely that we are correcting portfolio holdings as we go along to BOTH preserve capital (planning goal) and increase returns while decreasing risk of investments (investment goals)?
I loved the post, this is exactly the kind of content I want to see more of on this site. Thanks Carl and JD! Long time reader, first time commenter.
@Michael
You should use the geometric not the arithmetic mean when calculating average returns on investments. This site has a good explanation: http://betterexplained.com/articles/how-to-analyze-data-using-the-average/
@aconciselife
I strongly agree that the original post is simply presenting the same data in two different formats. The second graph is derived from the first graph! The only real lesson to learn here is that its better to invest small amounts over time rather than large amounts all at once.
This post made me think again about my own retirement investments. I started my 401(k) in 2001, invested almost entirely in stocks as I was supposed to do as a young investor (late 20s at the time). Because I invested at a very high point in the market, the money I invested for the first 7 years of my career has been obliterated, probably worth about half of its value. That money will like never regain enough value to equal the modest gains had I put it into more conservative funds. There’s nothing to do now but try to save even more, though it’s frustrating to think of almost a decade of my economic life being wasted.
@John Buerger: The ultimate distinction – Wealth Building results from good financial planning – not just investment planning. That is not a reality that financial salespeople in the investment houses want their clients to hear.
Totally agree with John. A well thought out financial plan has to be in place first prior to any investments being made. Each year the plan must be reviewed for course corrections. I have been doing this long enough to realize client’s change their minds – Life Happens. Parents live longer than expected and need assistance, little junior takes six years to complete college instead of four, or heaven forbid the client has huge medical problems ten years from the initial retirement goal (yet has great disability/medical insurance due to planning). Investment returns have nothing to do with any of these planning dilemmas.
Carl- great job and keep this thought process moving. Education remains a high priority for the investing public, especially to counter the message from the Wall Street houses. More investors need financial plans period. Fees paid for a solid financial plan are a great investment.
This is an interesting discussion. I appreciate Carl bringing this discussion up and being willing to leave it open ended for discussion. This topic is important because contrary to what one of the above comments says, this is not all that common, “common sense.”
@aconsciencelife #38.
Your use of coin flips is a good analogy, but only as long as you are sure the coin is “fair”. i.e., That it is just as likely to come up heads as tails. Problem with the financial markets is that they are not “fair”. Politicians drastically affect market outcomes – even changing rules after the game ends (ask a Chrysler bond holder). In addition to politicians,good and bad actors within companies and even the market-makers themselves also affect whether the market coin is “fair”.
If only we could assume a normal distribution curve… but alas – there are unreliable people in the equation and mathematical models often fail as a result of this (ask LTCM).
@Craig
I read your post with much interest as it is almost identical in the emotion and words used by two 30 somethings that recently came to me for advice. In general I advise those that are just starting out with their saving/investing to invest very conservatively until they have built up a decent amount of money for this very reason. The simple truth that I have found is that for many people stock investing [whether in index funds or other strategies] is too emotionally taxing until they have a financial base built up. It is a lot easier to withstand drawdowns if you have significant reserves. That is why I think Carl’s idea about risk deserves much credit. Risk is really not a mathematical concept, but an emotional concept.
I will also add that investing as an activity is something that people get better at as they gain experience and knowledge. No doubt the last 10 years has brought many investors to their knees [the last year especially], the smart ones will use this to learn and become better.
All this is why I don’t believe passive investing [or as its called in another thread “the lazy man’s way] is particularly appealing. Certainly, in a bull market it has its appeal, but the bear markets is where investors are made!
Carl and I had a discussion of these concepts last week by phone. I am impressed whenever I see an advisor that is willing to rethink the indoctrination of “buy-and-hope” models that have clouded our understanding of risk. The five summary points are excellent and are not disputable.
Great post
Jeff
Like Monevator, I was curious about how this concept of risk would fare in a more realistic scenario, where you invest over the course of 30 years rather than in a lump sum.
It turns out that unlike Carl’s graph above, the dollar amounts converge over time (to about 8.2%).
I used 1929-1958 (+8.5%) for my worst years and 1974-2003 (+12.3%) for my best years since I’m not sure which periods Carl used and I had those numbers available. For every dollar invested evenly over the worst period, you’d have $10.54 (+8.1%). For every dollar invested evenly over the best, you’d have $11.16 (+8.3%). Even by Carl’s definition of risk, that’s pretty safe.
For comparison, putting your money into CDs would return about $4.15 over 30 years.
@Will,
Your analysis is flawed. You’ve looked at two periods defined by their average percentage return, but this ignores the importance of the sequence of returns (when the above average returns occurred relative to the below average returns). There were actually periods with lower average returns then your best period that would have resulted in a higher ending balance, and there were periods with higher average returns then your worst period that would have resulted in a lower ending balance.
What you saw as a convergence of ending dollar amounts was a result of the sequence of returns, not the averages. In reality, the same pattern of cash flow can vary across different market periods by hundreds of thousands of dollars, even millions, over a lifetime.
A good period of returns for a saver making regular contributions would be characterized by a heavier distribution of below average returns at the beginning with the higher than average returns at the end (when there is more money in the portfolio). This same period would be bad for someone in retirement, regularly making withdrawals, even though the market average for the period is the same. The same market average for another period could have had the higher returns at the beginning and lower returns at the end. In which case it would be good for the spender and bad for the saver.
If you really want to see the long-term risk, list out a long sequence of returns on a spreadsheet and simulate regular contributions. Then try reversing the sequence, and then randomize it a few times. Or test multiple rolling periods using the same regular contributions. The risk Carl refers to is real. It’s well documented by academics and throughout the financial planning profession.
@Dylan: the analysis wasn’t flawed, just limited to the examples Carl gave. To extend the metaphor of a plane flying off course, investing money over time is like sending off 30 planes. Even if all of them will go way off-course in 30 years, at any given time some of them are pretty close to where you want them to be.
But perhaps that’s stretching the metaphor a little, so let’s go back to the real data (no need for random numbers). For an investor-over-time, the worst 30-year period began in 1952 and had an annualized return of 5.14%. The best period started in 1970 with an annualized return of 10.42%. For every dollar invested, that turns into $4.50 and $19.56 respectively. Although quite a spread, that’s a far cry from the $9.51-$62.53 difference that Carl talks about. As an aside, the average was 7.5% annual growth for a final dollar amount of $8.80 compared with Carl’s 11.28% and $24.71.
Based on that, I’d say that investing over time reduces risk in Carl’s sense (that is, reducing spread) compared to investing as a lump sum. On the other hand, it reduces total returns (presumably since most of it hasn’t compounded for 30 years).
Since most people are going to invest over time, I think it’s more reasonable to look at these numbers rather than assume that people are investing as a lump sum.
Which isn’t to say that I disagree with Carl’s 5 conclusions. It’s a great idea to sit down every once in a while and make sure that your financial goals are on track.
@Will,
Perhaps I should have phrased it as your conclusion that “the dollar amounts converge over time” based on your analysis is flawed. This is not a lump sum vs. periodic contributions issue. In either case, risk increases with the passage of time. Whether you invest in a lump sum or make periodic contributions, the spread of ending dollar amounts will increase, not converge, over time.
Have a look at 40 and 50 year periods, you can use your same numbers and you will see what I’m talking about (you may need to choose different periods or stop at 38 years). It does not matter that that the numbers are different than those in Carl’s $1 best/worst example, that’s just an example. More time equates to greater uncertainty, not the common misperception that that risk goes down over time. I think that is the point of his post.
@Dylan: But it is a lump sum vs. periodic issue. When you invest periodically, you’re basically averaging together many 30- (or 40- or 50-) year periods, which keeps the end result closer together than you’d see with a lump sum investment.
The best and worst 40-year periods for periodic investment (1935 and 1960) leave you with $10.76 or $32.46. The same numbers for 50-year periods are 1959 and 1950 with $20.24 and $82.62.
This means that the best value over 30 years is 4.35 times that of the worst. Over 40 years, it’s 3.02. Over 50 years, it’s 4.08.
On the other hand, investing a lump sum over those same periods (in their own worst/best years), you’ll get a factor of 6.58, 3.85, and 15.2 (!).
As you can see, investing over time reduces the difference between the worst and best cases, sometimes significantly.
In any case, I think the term “risk” is misapplied here, at least in the general case. As Carl mentions, we don’t care about percentages as much as we care about dollar amounts. But we also really don’t care about dollar amounts except insofar as they meet our goals. If my goal is to increase my retirement fund by 1% a year, then my risk over 30 years is essentially zero, even though the spread between the best and worst possible cases might be very large. That indicates to me that risk and value spread are two distinct concepts.
@Will,
No one is trying to make the argument that lump sum investing is safer. You are missing the point being made, which is that the commonly held notion that stock investing risks reduce as time horizon increases is problematic because it relies on the wrong measure of risk.
Most people do really care about dollar amounts because they’re needed meet our goals. If we knew that we could hit a dollar target decades out in the future with stock investments, then that would be safer because we could simply aim for a target return, calculate a savings rate, and go. But we can’t, and further we have to go, the more off course we could wind up if we try.
Just because you got some growth does not mean you’ll be able to fund your goals and protect against inflation and longevity. Half-a-million won’t cut it if you need a million. Conversely, if you save ’til it hurts in order to aim for $1 million and markets leave you with $2 million. Your painful savings could have been reduced. These are risks.
There is a very real risk of running out of money in the future or making unnecessary sacrifices along the way. This is the risk that increases with time when you invest in the stock market. It’s a manageable risk if you know to look for it, and as Carl suggests, make necessary course corrections along the way.
If you want to run your own numbers to see this with periodic contributions, you must look at more than just two samples. The periods that lead to the widest ranges after 30 years may be different from the periods that lead to the widest range after 40 years or 50 years. I’m not offering this to challenge any notions about lump sum vs periodic saving. I’m trying to illustrate the point about measuring “risk” in periodic saving terms to help you see that the the same application of risk Carl refers to is still present.
Carl,
Spot on AGAIN! I really like the analogy of driving to Florida and ending up in Maine. (I think I’ll have to use it myself). I’m sure there is no single source of blame here, but I wonder if advisors have such a difficult time moving the focus away from investments, and towards the other “levers” is due to financial media. 24/7 there are multiple stations dedicated to focusing on investments over what really matters. I once showed a client that even if his account rose by 50% in the next 12 months, he was still in great danger of running out of money in retirement (and that cutting spending by 15% a month could work)…guess which one he’s hoping will happen!
Russell