Learn lessons of the past, and from long term running of numbers

Examine history of markets, societies, technology, and the economy. Focusing here and now on markets, here are a few things to help that study:

Up until the last few months, with very few exceptions, when financial companies and experts talked about history’s lessons in economics and investing, they reached back no longer than 30 to 40 years, … implying at least that what was before is largely irrelevant because all you need to know can be learnt from the study of these most recent few decades. I use a software tool that can illustrate, e.g., how various investor responses at bear market times resulted in widely different consequences later. Below are three screenshots from this program, indicating the historic time frame used (notice that I picked the 1981-82 bear, … when the drop was the biggest), the variables you can manipulate, and one possible outcome, 26 years later (in two versions: first you’ll see the inflation-adjusted numbers, then the nominal ones).

I think this kind of illustration is very instructive, … more convincing than pure mathematical calculations showing the importance of diversification, or general statistics of various, but separate, aspects. Similarly, you can learn a lot from online tools (like, e.g., this one from Dynamic Funds and another from Fidelity Investments), or a cute presentation from Invesco Trimark about important aspects of how markets actually behaved at difficult times during the last few decades.

In the next table, I collected screenshots from ORC regarding an accumulation portfolio dilemma. Let’s say a 25 years old inherits $100K and wants to invest it to fund his/her retirement at age 65. (He has about an 80% chance to reach that age, and she has a 88% chance.) Let’s also assume that s/he will add $5K new money, indexed at 3% per year, annually to the portfolio. S/he feels our current situation is comparable to the Big Depression of the 1930s, wants to decide if s/he should invest in stocks at all, and if yes then what percentage of the portfolio.

In each row of the next table you’ll see the same charts made by ORC, showing the growth trajectory of this ‘$100K starter plus $5K indexed annually’ scenario would have taken if it were started in 1900, 1901, 1902, etc., at a certain stocks/bonds composition as indicated in the first column. (The Dow Jones index is used to represent the equity growth, year by year, and also US data are used for fixed income and inflation rates.) The only difference within rows is that a particular year’s trajectory (the one that is indicated in the top row) is highlighted from the hard-to-distinguish mash of the many trajectories. You might be able to see the thick black trajectories in the small-size pictures of the table as well, but it will be easier if you click on the table-cells (the small charts), because that way a larger size version of each chart will pop-up, … unless you disallowed pop-ops in your browser, of course. While our focus is now on the thick black lines, you may also want to know what the green, blue, and red lines mean. They do not represent any particular year’s trajectory; instead, they show the level of assets as time progressed in the ‘lucky’, ‘median’, and ‘unlucky’ cases. Median means that half of the yearly trajectories (the thin lines) were above and half below this level, while the other two levels refer to the size of portfolio which separates the top and bottom 10 percentages of annual trajectories from the rest.

equities/bonds	1929	1930	1931	1932
100%/0%	Assets at age 65: $1,723K	Assets at age 65: $1,482K	Assets at age 65: $1,681K	Assets at age 65: $2,310K
75%/25%	Assets at age 65: $1,637K	Assets at age 65: $1,493K	Assets at age 65: $1,678K	Assets at age 65: $2,100K
50%/50%	Assets at age 65: $1,520K	Assets at age 65: $1,455K	Assets at age 65: $1,603K	Assets at age 65: $1,850K
25%/75%	Assets at age 65: $1,346K	Assets at age 65: $1,346K	Assets at age 65: $1,450K	Assets at age 65: $1,566K
0%/100%	Assets at age 65: $1,133K	Assets at age 65: $1,181K	Assets at age 65: $1,250	Assets at age 65: $1,289

As you can see, the total assets at age 65 varies significantly between $1.13 million to $2.3 million, depending on both the starting year and the portfolio allocation. Notice also that as you increase the equity percentage, the end total is growing too, but (i) not without exception, and (ii) with ‘diminishing returns’, that is by increasing the total with less amount by the same increase in the equity percentage. We are not looking here at volatility of the various portfolios, but this apparent ‘diminishing return’ combined with stocks’ higher volatility indicates that even for this long term accumulation portfolio some balance between stocks and bonds would be a better choice than pure equities. The two most glaring lessons for this young person appear to be: put a large chunk of your investments in stocks despite the harsh reality of what is going on with stocks, and rebalance the portfolio regularly, because you want to be part of their come-back whenever it happens, and (ii) be aware that your final score depends to a large extent on luck, … since you cannot know right now for sure if today resembles more which one of these years. There is one more thing worth noticing in the charts: most of the time the thick black lines are below the blue ones, and especially if equities were not at least 50% of the portfolio, they performed really-really poorly, compared to other time periods.

Let’s look now at a different situation, in which a 55 years old person plans to retire at age 60. S/he has now $600K, and will add $10K to it annually until retirement. This person wants to draw the equivalent of what is $25K today from the portfolio from age 60 to age 95. (Statistically speaking, his chance of living beyond that ripe age is 6%, and hers is 13%). Picking the same four years as in the previous case, we are interested in knowing, of course, what the level of total assets at age 95 would have been for these unfortunately timed retirements, but the even more vital question is if the nest egg could have lasted that long at all. Look at the striking differences in the table below.

equities/bonds	1929	1930	1931	1932
100%/0%	Assets at age 95: $0 - run out @ age 75	Assets at age 95: $0 - run out @ age 79	Assets at age 95: $15K	Assets at age 95: $3,229K
75%/25%	Assets at age 95: $0 - run out @ age 83	Assets at age 95: $0 - run out @ age 91	Assets at age 95: $472K	Assets at age 95: $2,450K
50%/50%	Assets at age 95: $0 - run out @ age 94	Assets at age 95: $212K	Assets at age 95: $552K	Assets at age 95: $1,444K
25%/75%	Assets at age 95: $259K	Assets at age 95: $284K	Assets at age 95: $276K	Assets at age 95: $384K
0%/100%	Assets at age 95: $167K	Assets at age 95: $39K	Assets at age 95: $0 - run out @ age 90	Assets at age 95: $0 - run out @ age 86

Without equities in the portfolio (last line), two of the four cases ended up too early with no money left, and the other two cases were not far ahead either, having just a bit in them by the end. The most shocking differences among the years are still at portfolio allocations where equities played a significant role. With high equity levels, it’s either heaven or hell, … and would you dare to bet if our current situation and timing is more in line with any one of these years? No promise or much likelihood of course that any of these scenarios will be repeated, still it’s hard not to conclude, I think, that (i) for this close-to-retirement person, having a high percentage of stocks can be lethal, and that (ii) modest level of stocks appears to be the best bet against the wild vicissitude of history and markets. Notice that there is one line only in the table with a relatively even outcome through the years, and without a single case of running out of the money too early. Notice also that, except for a very few of these charts, the thick black line is below the blue one, and sometimes even the red one, … indicating that people retiring in those years (or to be more correct, five years after these particular years, … since that’s how we run the scenario, remember?) were really not spoiled by history.

Both of these illustrations above, just like any illustrations ever, can be criticised for various things. Let me address two main lines of potential criticism. One is that there are more than stocks and bonds to build a portfolio, and the other is that one is not obliged to keep any fixed asset-allocation as was supposed here.

It would be more real and beneficial to calculate with different asset types in addition (or even instead of) the stocks and bonds used here, … like cash, GICs, annuities, etc. There is a lot of validity in this criticism, but it doesn’t invalidate completely the lessons I suggested. The main factor disregarded by this criticism is inflation. We should remember that inflation is an essential factor from time to time and it can kill portfolios consisting too much of low-risk assets. In the above retirement calculations, the annual income amount demanded from the portfolio changed year by year according to the actual historical inflation numbers, lending it more credibility again than when it’s disregarded or assumed to act evenly at some rate. If one deals with a specific plan, there must be more factors included, but here we focused narrowly on the role and nature of having equities in portfolios, and for that purpose the equity_or_bond simplification is acceptable, I think.

As for the second criticism, … it’s a tricky issue, leading to far-reaching arguments about portfolio management, and predictability, especially in a timely manner, of markets (or anything, really). You can read a lot about these issues on this site and elsewhere. Let me sum my position this way: there is no final conclusion regarding the superiority of passive over active asset management, or vice versa. It’s probably basic human nature that we feel sometimes strongly that we understand well and foresee things, which we actually don’t. Especially challenging is to forecast the temporal aspects of things (when exactly, for how long, at what pace things will stay/change). Because of these challenges, I’d feel uncomfortable with recommending too much active management, either in terms of portfolio allocation changes, or in terms of picking ‘winner’ assets. There is a lot of anecdotal evidence to discredit inflated claims and hopes, and also a lot of science and research from various fields, to the same effect. Consequently, I don’t expect significantly different results from what history metes out for us, so to speak, even though I wouldn’t give up completely the chance of ‘being right’ with making choices and predicting things either. On the other hand, if someone says they believe in and want only passively managed low-cost portfolios (of indexes and ETFs) I can completely agree with that. Just be aware that while you will not risk losing unnecessarily compared to ‘the market’, you will not outperform it either. (Or at least not by much; .. more about that in a separate future post.) Study ORC and other historic sources to get an idea of what you can expect.

Lest I implant some misunderstanding regarding ORC, I’d like to emphasize that these illustrations above represent only a very limited use of its capabilities. The software does much more that creating this kind of charts, from so simple assumptions, … although these ‘aftcasts’, as Otar contrasts them with ‘forecasts’, are what the whole thing is based on.

Finally, let me show some very simple Excel spreadsheet calculations of my own, again for an accumulation portfolio. It’s a simplistic and speculative piece about alternative futures. There is not much historic reality or precedent here, except that I will start the story with something that happened to many people recently. Let’s assume for this case that at the end of 2007 you had $1,000 in equities (100 units/stocks at $10 each) plus another $1,000 in an income asset that guarantees 3% return per annum. Equity markets have declined 50%, therefore you already lost 23.5% of your investment portfolio by the end of 2008 (consisting now of roughly 1/3 equities and a 2/3 guaranteed 3% income generating portion). You think we probably are at the market’s bottom, and the market will come back to the original level by 2020, … a cautious but not very cautious assumption, I’d say. It has happened a few times that markets didn’t go back to previous levels in such a time frame, although most people expect that it will do it much sooner (and they might be right, despite my doubts, of course). Here are some alternative action scenarios to consider:

Alt.1-Do Nothing
Alt.2-Sell all stocks and invest proceeds in the 3% guaranteed income asset
Alt.3-Rebalance the portfolio to the original 50%/50% -> which would require you to sell some fixed income assets and buy 53 units of equity.

Let’s assume that things do not go well on the stock market in 2009, and in fact there is another 50% drop in prices. (Unit prices at the $2.5 level by year-end.) Provided you were a risk-taker who rebalanced at the end of 2008, you will face the same choices at the end of 2009: do nothing, throw the towel in, or keep rebalancing, … that is buy more equity units. Let’s assume that you can handle (both financially and emotionally) the pain of suffering losses; we can see your choice as …

Alt.4-Rebalance again —> which would require you to sell some fixed income and buy 81 more units of equity.

What if in 2010, instead of a finally turning tide, we are to see another halving of equity prices, … meaning that we’d be down to the $1.25 unit price level. (Notice that this is an assumption so devastating that nobody has gone this far, not even the ‘gurus of doom’ shown in previous sections. However, this is quite close to what has happened, during the Great Depression.) Provided again that you are a firm believer in rebalancing (and that your spouse will have not secured yet your being declared insane and therefore unable to deal with your finances) you may pick again the same action:

Alt.5-Rebalance again —> which would require you to sell some fixed income and buy 134 more units of equity.

Let’s see how the numbers will work out by 2020 if unit prices will really go back to the original level of $10 by then, or if they go back only to $7, $5 or $4, … how much will you end up with then? (Plus, just for the indication of price sensitivity and upside potential, I added an ‘optimistic’ last column where I assume $12 unit prices.)

Calculated outcomes in 2020:

Let’s be a bit even more cautious, and assume that the return of equity prices happens by 2030 only, … and run the same scenarios again. The numbers change according to the table below:

These are, of course, overly simplified and extremely pessimistic assumptions, just to show that even in these terrible scenarios - provided one can wait long enough, and provided there will be at least some partial come back in equity prices - sticking to strategic asset-allocation might make sense. Rebalancing, that is selling what performed (relatively) well recently and buying what performed the worst, however painful, is the essence of it. The potential disadvantages compared to a ‘no action’ or ‘flee equities altogether’ approach’s consequences are less than the potential advantages if things do turn out to be just a little bit better than those extremes. If, on the other hand, things do turn out to be really that bad, then perhaps a 30 or 40 percent disadvantage in the total portfolio will be the smallest of your problems, … scenarios like these conjure up situations (hyperinflation, social collapse, etc.) where we may have more important things to be concerned about than this monetary loss. Notice that even in the worst situations, the future totals are not less than what is the value of total assets today ($1.530), after the already historic loss of 2008. Not a big consolation, I know, but still … something.

Of course, an easy way to criticise these wild and oversimplified calculations is that in the more fortunate outcome scenarios there is no portfolio rebalancing during the improvement period, … so the equity/fixed income proportions do not even resemble the original 50%/50% by the end. Fair and important comment, no doubt. Why I think it’s still not useless to look at these numbers is that the act may help drive home the importance of looking at the long term, and widening the range of potential futures we are preparing for. Also, it may help turning attention from focusing on losses already suffered to thinking on what might happen in the future, and act accordingly.

I didn’t refer to Jeremy Grantham in the earlier section listing several of my sources of reference, because I didn’t know about him until quite recently. Content-wise, this interview with him would fit there quite nicely. Below is an interesting excerpt form the transcript of the conversation. After he explained that he thinks markets are relatively cheap but still can go further down, he explained how he approaches whether to increase or not clients’ exposure to stocks:

“Yes, I would say two-to-one, by the way, my instinct plus looking at the history books that it will go to a new low [in 2009]. So, this is the problem. We’re under weighted still, in an ordinary asset allocation account that has 65 percent in equities, we have moved up to 55. So, we’re still underweight, even though they’re cheaper than they’ve been and they’re reasonably cheap.

But in any case, it’s a very personal judgment of risk avoidance and how tough you are under stress. The worst situation that will befall probably quite a lot of people is that they exaggerate their toughness. The market goes down 30 percent from here to 600 and they panic, dump their stocks and never get back. And that’s the worst outcome.”

If you watch or read the whole interview, you’ll probably agree that this seasoned and very successful veteran of investing appears to be the kind of active manager one can entrust with part of their assets. Unfortunately, you will not find many of them around.

Top

After weeks of brooding, post was finally published on March 8, 2009