Life After Full-time Work Blog

Learn about preparing for life after full-time work through posts from Don's upcoming book.

#70 Three Goals, Three Instruments

This post looks at the main kinds of financial goals we have for retirement, and why each goal needs its own financial instrument.


Putting it all together, we have identified three potential financial goals in connection with retirement finance. (Obviously I’m ignoring the bequest motive, taxation, and so on.)  Two of them concern investment, one of them concerns longevity. Roughly speaking, they sound like this:

  • The investment goal of safety: arrange my assets so that they give me the amount of money I need, at the time I need it.
  • The investment goal of growth: arrange my assets so that they continue to grow.
  • The longevity goal of not outliving assets: stretch my assets to ensure that they don’t run out while I’m still alive.

These are separate goals. We’ve seen in many stages that investment safety and growth are at opposite ends of the spectrum. It’s obvious that the longevity goal is also separate, because it isn’t in fact an investment goal at all. It’s totally compatible with either growth or safety or both. You can go for growth, you can go for safety, and the investment instruments involved never ask you “How old are you?”


There are many forms of investment instruments available. Some are focused exclusively on safety; some are focused exclusively on growth potential. Think of those as basic or pure instruments. Some combine the two, influencing a mixture of growth and safety; think of them as combination instruments.

In Post #68 ( I invoked one particular philosophy with the words: “As long as I have X, I’m happy to go for growth with the rest of my pot.” Here X could be a number, or it could be an outcome you specify.

In principle, it makes sense to start by considering the basic instruments, the ones at the ends of the spectrum. That makes it easier to match your choice of X, by fitting the proportions of the two sets, also called the allocation across them. And if, later, circumstances change and you want to tweak either X itself or the allocation to the basic safety and growth instruments, you (or your professional) can do so directly.

It may happen that it’s more economical to use a combination instrument, perhaps because that instrument is more readily available or more easily traded. So you include a combination instrument that gives you some safety and some growth. But it may be tough to use it, all by itself, to get either as much safety as you seek, or as much growth. And it may be that the combination instrument can’t move enough towards the safety end or the growth end of the spectrum, if and when you desire a tweak.

For example, a combination instrument may be essentially a mixture of one-third safety and two-thirds growth. By changing the allocation to the combination instrument, you can move towards safety or towards growth; but there’s some level of safety below which you can never go, and some level of growth above which you can never go, as long as that’s the only instrument you use.

So, while the combination instrument may be useful and have other attractive characteristics, it may be desirable to have some element of pure safety and some element of pure growth, the allocations to which can be separately dialed to give you the X that you desire.

That, by the way, is the common sense explanation of Nobel Prize winner Jan Tinbergen’s dictum that the number of instruments you use should be at least equal to the number of goals you have.


Sidebar: I know that most people have never heard of Tinbergen, so let me tell you why he’s a giant.

Nobel Prizes started in 1901. Except, that is, for the Economics Prize, which started in 1969. So, when the first Economics award was made in 1969, every living economist was eligible. When I mention this to American audiences, I compare it to the start of the Baseball Hall of Fame. The first intake, in 1936, included the ultimate baseball giants, the greatest in history: Ty Cobb, Walter Johnson, Christy Mathewson, Babe Ruth, Honus Wagner. Well, in 1969 it wasn’t Samuelson or Kuznets or Hayek or Friedman or other famous names who got the first Economics award. They got their awards later. One of the two people honored in 1969 was Jan Tinbergen.

His work was in macroeconomics. But along the way he enunciated a principle which, like the best of insights, seems like common sense – once you hear it. Someone has to say it, first. Only then is it obvious. Here’s what he said (though it seems impossible to get an exact quote):

Achieving a number of economic targets requires the use of at least an equal number of instruments.  


Now, here’s how to apply Tinbergen’s principle to retirees. They have three financial goals, as we’ve seen on numerous occasions. And at the risk of overkill, I’ll explain them again.

One is longevity protection. Quite simply, they don’t want to outlive their money.

Another is growth. For most retirees, their ambitions are less than 100% funded, and they therefore need to take some investment risk.

Third – ironically, as has been pointed out – they’re very risk-averse.In particular, they don’t want to be suddenly told, “Oops, too bad, last year the markets didn’t perform, this year you’ll have to turn down the spending dial.”No! They realise that growth-seeking is risky, but they want some advance warning of the need to turn down the spending dial.

And Tinbergen reminds us that, if we have three goals, we need (at least) three instruments.

We’ve already discussed accommodating safety and growth by using at least two investment instruments.

The longevity goal isn’t an investment goal. A direct application, then, of Tinbergen’s principle is that you shouldn’t try to solve the longevity problem via an investment instrument. Not that you can’t do it, but it’s very expensive and therefore inefficient to try to do so (though this isn’t the place to go into detail).


One final consequence that I want to mention relates to measuring outcomes or performance.

Again, it’s common sense that is often ignored.

Suppose you desire safety and organize a portion of your retirement pot to give you safety. For example, you get a portfolio of index-linked instruments, or perhaps bank instruments, because X, the definition of safety, is for you to have certain amounts of cash flow on certain dates.

The measure of success is whether or not this actually achieves the safety you specified. It’s almost a “yes or no” question. If it achieves X for you, it worked. If not, it didn’t, even though it may have come close.

If there are many possible arrangements, all of which achieve X for you, then it makes sense to select the one with the smallest cost. It may even be that some arrangements give you a positive investment return. Great! But that’s a side effect, not your main goal.

What doesn’t make sense is to measure the return that your safety-oriented arrangement gives you, and integrate it into the return that you achieve with your growth-seeking portfolio, and say “That’s the total return from my pension pot,” and compare it with the total return from someone else’s pension pot.

That’s because you’ve really divided your pot into two parts, each with a different goal (safety and growth). You can sensibly measure whether the safety part achieved X, and at what cost, and what the return is on the growth part. But it doesn’t make sense to add in the growth return when you measure the success of the safety part, and it doesn’t make sense to reduce the growth return by the cost of the safety measures, when you measure the success of the growth part.

It’s amazing how often this simple piece of common sense is ignored in practice, not just by financial professionals but sometimes even by academics, when they say that dividing your overall pot into different buckets doesn’t affect the overall return. It’s true mathematically, but the overall return is irrelevant when you have different goals you’re trying to achieve with different portions of your pot.



Typically you’ll have three goals: investment safety, investment growth and some form of protection against the financial impact of a long life. This suggests the use of three different instruments, one for each goal.


I have written about retirement planning before and some of that material also relates to topics or issues that are being discussed here. Where relevant I draw on material from three sources: The Retirement Plan Solution (co-authored with Bob Collie and Matt Smith, published by John Wiley & Sons, Inc., 2009), my foreword to Someday Rich (by Timothy Noonan and Matt Smith, also published by Wiley, 2012), and my occasional column The Art of Investment in the FT Money supplement of The Financial Times, published in the UK. I am grateful to the other authors and to The Financial Times for permission to use the material here.

2 Responses to “#70 Three Goals, Three Instruments”

  1. JJ Woolverton says:

    Don, I took math in school along with geometry, trigonometry, physics and calculus and have gone through my whole working career not having to solve for “X”, and now, in my retirement years, you are asking me to solve for X — pretty cruel.

    Agree, not to measure performance against others, however, given the typically competitive environment we are always in, this makes it difficult not to compare. Unfortunately, we tend to compare against individuals who are better off than we are and, therefore, we always feel less than successful.

    • Don Ezra says:

      Sorry about X!!! Re comparisons … sure, we live in a competitive world. Two quick thoughts. (1) My remarks were about what to measure, rather than about comparisons with others. My point is that the “growth plus safety/insurance” total return has no meaning, as it combines two totally different (indeed, opposite) goals. So, if you want to compare, then compare like with like: compare your growth portfolio against someone else’s growth portfolio, not your total return against someone else’s. (2) I wonder if comparisons of investment performance really matter to the client. Certainly they matter to the financial professional, as a sort of mark of esteem and excellence. But if I’m the client, it’s my lifestyle that matters to me, not the recent returns that went into it, which are just one ingredient in the overall meal.

Leave your question or comment