PsyFi Search

Thursday 6 October 2011

A Yen For Yield: Redux

Uncertainty On the Brink

Following on from A Yen For Yield we’re left with the rather odd position that a quest for yield through investing in higher yielding, but risky, assets seems quite at odds with the possibility of a deflationary environment where risky assets are likely to valued downwards. This is what comes of looking at behaviour as well as economics.

The underlying problem is uncertainty. No matter what anyone tells you, no one knows exactly what tomorrow will bring. And, in particular, if you stand on the brink of retirement that means no second chances if you get things wrong.

Foolish Predictions

We face two particular problems. The first is that the world is full of people making predictions and telling us what’s going to happen next. The second is that they can’t possibly know what will happen next because of all those darn quantum butterflies. In fact the biggest danger is that we listen to these siren calls and adopt a polarized view. For many of us this is a risk we can recover from, but if you’re at the end of your working life this may not be an option and, unfortunately, we’re living in peculiarly difficult times.

If we take the research from the San Fransisco Fed we examined in the previous article, which shows a close correlation between the demographic profiles of Japan in the 1990s and the US in the 2000s and an equally close correlation between the stockmarket performance over those same periods, we can leap to the conclusion that the future is set in stone: we can look forward to a decade or more of deflation, with equivalent declines in stockmarkets, as the demographic bulge of retirees sells down their equity holdings to fund their retirements.

This vision implies that stockmarkets would be a poor place to put our money, not just because of the capital risk but also because deflation is not a friend to firms: many companies would actually be forced to cut their prices and we’d likely see significant lay-offs as the only way of cutting labor costs is to remove staff: for the reasons we examined in Money Illusion, people are fixated on their nominal wages and fail to recognise that’s it’s the nominal price of things that’s important.

Brazilian Butterflies

However, there’s a nasty sting in this particular tale. The problem is that this vision of a deflationary future, in which even low yielding government bonds would be a good investment, is by no means certain. Correlation is not causality and there’s no guarantee that the deflation seen in Japan will be replicated elsewhere. The economy is a complex adaptive system, and if that darned butterfly in Brazil starts flapping its wings at the wrong time, we may end up with a very different result and if you’re invested in low-yielding bonds when inflation takes off you’re looking at one hell of an impoverished retirement.

Polarization, the effect that people tend to take extreme views about these things, runs rife in the general media as described by Daniel Isenberg. Individuals rarely get paid for writing articles that sit on the fence, refusing to take a position. Fortunately for me, I don’t get paid for writing this stuff … and I’m sitting on the fence.

Debias or Die

Debiasing ourselves from this type of polarized position is critical and often very difficult, but one way we can do this is by looking at a simplified decision tree. Let’s take just two options: (a) the next decade will be broadly deflationary and prices will fall in real terms and (b) the next decade will be broadly inflationary and prices will rise in real terms. Of course there are lots of intermediate or even more extreme possibilities, but we’re only worried about the general idea here, not the exact outcome of unpredictable events.

Now it’s impossible to deny the fact that either (a) or (b) is possible. We can each assign our own probabilities to these events and these probabilities may even vary daily or hourly. What we can reasonably say, though, is that anyone assigning a 0% probability to either of these options is smoking some substance that renders their opinions worthless.

What this exercise does is force our recalcitrant brains to step back from a polarised position and to take a more nuanced view of the overall situation. Which means that adopting any retirement investment strategy which assumes that one of these scenarios is certain is worse than questionable: it’s an example of extreme behavioral bias.

Options, Options

On the deflationary side fixed interest, government bonds, usually in the form of a retirement annuity, can offer a long-term, virtually guaranteed income. Many lifelong equity investors habitually underestimate the value of this – another example of behavioral bias, fear of regret, the sadness that would engulf them if the Dow raced off to 100,000 while they earn a nominal 1.5% a year on an annuity. As Mitchell and Utkus explain:
“Behavioral factors may also explain the low demand for annuities in retirement, including most importantly loss aversion. This arises because some retirees may worry about potential losses to heirs in the event that they die “early”, since annuitization typically eliminates the possibility for bequeathing these funds. Adding to the problem is that retirees may heavily discount future benefit coverage in the event that they life a long time in retirement”
In retirement, though, when mistakes in investment strategy can’t be recovered, certainty is worth paying for, especially when we look at the extended old-ages many of us will experience. However, if we look at the inflationary possibility it’s clear that a retirement income based solely on fixed income assets isn’t a safe option. Although stocks tend to be depressed by deflation they tend to be buoyed by inflation; although the difference between companies that can make real returns in an inflationary environment and those that are simply riding the general price increases is often hard to distinguish.

Behavioral Retirement

So the idea by the researchers at the Cass Business School that 25% to 50% of funds at retirement should be allocated to stocks, and then gradually reduced, seems like a reasonable balance in what is an uncertain world. Of course, you can back-test this type of approach until you run out of computing power, but the results will almost certainly be invalidated by the next nectar-hungry butterfly that flaps past your window.

In a deflationary environment you’d generally expect to see the equity element of a portfolio marked down – even more so, perhaps, if we think about mass sales of stocks by capital hungry retirees. Nonetheless, the Japanese example shows that pensioners – or at least their fund managers – aren’t content to simply accept low incomes. The flow of Japanese funds into higher yielding currencies was simply an outcome of that quest for more income.

Now Japanese stocks have traditionally not paid significant dividends, and started their decline from extraordinarily high valuations. If we look closer to home, as most home bias suffering retirees will surely do, then a more likely beneficiary of a deflationary environment will be higher yielding stocks. Of course, these will be risky and capital will flow out of them on whims and notions, but such stocks offer the opportunity of upside in both the inflationary scenario – as their prices increase in line with inflation – and the deflationary one – as their prices are buoyed by behaviorally challenged investors desperately seeking income.

Of course, this isn’t certain. There is no certainty. After all, if there was, then tomorrow and tomorrow and tomorrow would be perfectly predictable. And where would be the fun in that?

Related articles: A Yen For YieldMoney Illusion, Investment Forecasts - Known Unknowns


  1. Hi Timarr,

    Great follow-up post. Note that while I personally have a deflationary Japanese template bias for long-term macro reasons, including demographics, over indebtedness and historic extremes in important societal benchmarks like income and wealth distributions around the world, I don't base investment decisions on this premise. Rather, we have chosen to avoid the long-term buy and hold approach at this time, regardless of individual company fundamentals, because our statistical models suggest that long-term returns to stocks are likely to be poor over the next 10 to 15 years.

    Please see for our most recent study.

    In the meantime we are surfing the intermediate term 'beta waves' in global asset classes using a momentum/trend based framework which has worked well for us so far.

    Love your stuff!

  2. Rather, we have chosen to avoid the long-term buy and hold approach at this time, regardless of individual company fundamentals, because our statistical models suggest that long-term returns to stocks are likely to be poor over the next 10 to 15 years.

    Sadly I have to agree with that; cyclical bear markets tend to punish long-term holders if you can't wait them out. Active link here to estimating future returns. Well worth reading as a salutary counter to the more persistently positive purveyors of a permanent equity premium.