Christine Benz: So, Professor Merton, you've talked about some shortcomings that you see in traditional defined-contribution plans. Let's discuss some of those?
Robert Merton: Sure. One has to look back a little bit of the history in connection of DC plans. They were traditionally for most people a supplemental addition for retirement, usually for people who were somewhat higher-income, who needed to save more than they would have in their company plans, which were typically defined-benefit plans.
And as supplements, we look at them in a different day than we do if they were at the core of retirement, which is where they are now looking like are going to be employed. And I would say that the limitations of them are that for the vast majority of people, vast majority, the challenges, the questions, or the decisions they have to make are very complex decisions. They are decisions they've never had to make in the past; they are decisions they can't make in the present or are not equipped to.
Benz: So, decisions like how much to save, how to asset allocate?
Benz: Things that most people don't have experience with?
Merton: Absolutely, because if you think of that most people came from some kind of a retirement plan, a corporate retirement plan like a defined-benefit plan, where they never had to make any decisions, they never even knew anything about rate of returns and never even thought about it.
Benz: All being done behind the scenes?
Merton: All being done behind the scenes by their employer or the agents of their employer. Now we're saying to them, "OK, now it's your responsibility, and here is the problem you need to solve." The problem you need to solve is that you're going to have contributions made as a function of your income over the years you work, and those contributions have to be invested in such a fashion, so that when you reach retirement you will have sufficient resources to provide for a good standard of living for the rest of your life. That's a very complex set of decisions.
We who are professionals who do this for a living every day find this a very daunting challenge. The idea that millions of individuals are going to solve this problem or will be required to solve that problem is a bit daunting. And yes, as you said, they can get advice, and they should. But then they are in the role of whose advice do they take, where, and what do they do. If the advice given to them is that of "Well, what do you want to do?" frankly most people don't know, particularly when they are much younger, someone like yourself, you have so many years before you will retire, it's an abstract concept.
So, to say what do you want for retirement? The typical answer that I have seen is, "Tell me what I probably going to need or want in retirement." They throw it back to you because they really don't know. So, that responsibility for this complex decision is pushed around.
I mean this is as if you're a brain surgeon or professor or auto assembly line worker, it doesn't matter. And indeed on top of that not only it is something that we are not equipped to do, but it's also something that most people don't like to do. There are plenty of people in the finance profession, when it comes to their personal savings for retirement, they turn it over to someone else rather than do it themselves. So that's the first thought.
The second is that the goal in retirement is to achieve enough to live on, not to get partial pieces. So, to do a good job of managing that money to try to achieve the proper goals, you need as much information as possible about all the assets of the individual. Historically, there are sort of three, as we'd say in the United States, legs of a stool through the retirement. (In the United Kingdom and Europe it's a little more sophisticated. They call the three pillars, but it's a same thing.)
These are the three sources of retirement funding. We have government. In our case in United States Social Security, employer plans which is the second, and then personal savings. And it's a mix of all of those to get to us our goal. Ideally, we would want to know all three. The third one is difficult, at least at this time because we just don't have the information.
Benz: On individuals' personal savings?
Merton: On individuals' personal savings and what they would do. I think we will have more in the future. For the moment though, we want to certainly take account of all as much as we can of sources from the first two legs, mainly government and employer-provided plans.
But we also want to take account of a very important asset, future contributions. This is because if you want to think of it this way, if someone is 28 or 30 it might be that when you look at their assets for retirement, maybe 10% of their assets are in this investment, the other 90% are future contributions.
Benz: The human capital piece?
Merton: The human capital piece.
Merton: So, just let's think about that. If we only focus on the financial piece and we say we put a 100% of that in equities, when you just look at what happens to that, it looks very risky. We say to ourselves, "We had a decline of 30%; equities have lost 30%."
But if we look at it more holistically and say, "Really it's not that bad. They only have 10% of their retirement assets in this, because the other 90% is future contributions," then saying a decline of 30% is really only 3%. If you add to that the Social Security which is of different importance depending at the level of income--obviously Social Security benefits are bounded, so, for lower-income people it's a bigger importance than for higher-income people--that's a very important piece.
Then finally, what we at least try to do is that many people, because of this transition the way retirement is done, are moving from having been in a defined-benefit plan of their company, to a defined-contribution plan, which means there will be a number of people, particularly people who are maybe in their 40s or even in their 50s, who have a accumulated a very substantial benefit in their defined-benefit plan.
We need to have that in order to do the right thing for them; that's a very key asset. If it's with the same employer that's accessible. So, the spirit is, we would like to have, if possible, as much information about the dedicated retirement assets as we can in order to make an intelligent and informed decision on how to get people to their goal in the most efficient and effective way possible.
Benz: And how to find the right asset allocation. So, I think a lot of us looking at the industry might say, "Well, a target-date fund helps get participants halfway there at least in terms of keeping them from making those terrible decisions like switching into a technology fund in late 1999 or something like that. But do you think that target-date funds, as currently conceived, have some significant shortcomings?
Merton: Yes, I do think they do. I know this was in an attempt to try to put some portfolio management in there--it's automatic and the investors are not engaged--and that's a good step. The difficulties are multifold, however, in the implementation. For one thing, the target-date funds put fixed weights solely as a function of age or time, independent of the individual's circumstances, how much they've accumulated, what their salary level is, and all these other factors, and independent of how market conditions change.
So, you could have two twins, same age, but who have very different backgrounds in their other assets and who really should be doing something different. You're not making that adjustment there, that's one part. The second part is that target-date funds really don't have a goal. They don't have a target. They don't have a target in sense of something that we can understand. The target we normally think of is you'll need a certain amount of money or as I would prefer to say, a certain amount of permanent income, by the time we retire to deal with your retirement.
Target-date funds do not set either of those. All they do is describe what's called a glide path, which is an asset allocation, a function of time up to some date. And so, it's really describing the means without saying where you're going. It's very hard to think of something that's close to being optimal for people that doesn't even have a target.
Then the last piece which is extremely important in my view--and this is common not to just target-date funds but to most defined-contribution funds--is that the targets themselves if they have them, outcomes, are distribution of wealth, a certain amount of money accumulated. Rather than what I believe is the appropriate target which is at retirement a stream of income for life, protected for inflation, if we think about standard of living--which is what people really care about certainly in retirement--we don't define a standard of living by how much money we have. We usually define it by a flow of income necessary to sustain ourselves, and it turns out that the risk differences between what's risky with respect to wealth or value versus what's risky with respect to income are very, very different.
As someone said it it's an analogous to using the wrong currency to measure gold. So, if you are an investor from Japan and you care about what you can buy in Yen, it isn't very helpful to have targets set in terms of a dollar return because it's a fact that the currency is uncertain. And it's much the same here, because interest rates change a lot, because how long people are likely to live changes a lot, and because inflation can enter. These three elements introduce a great deal of difference between a certain amount of money and a certain accumulated wealth.
One quick example would be, if somebody who was lucky enough to have $1 million in a bank account, it wasn't that many years ago when they could get 4% or even 5% in a certificate of deposit, a standard old thing, and they were able to get in that case $40,000 or $50,000 a year in income or more. But then you have today, you're lucky if you can get $2,000 a year on the same amount of wealth.
So, if you just looked at wealth, managers will say, "Look, I preserved your capital." But if you look at welfare or on what investors can live, they could live on $40,000 or $50,000 a year, they can't live on $2,000. And this is a real world example, it's not hypothetical. So, those, I think, are the major drawbacks I see to target-date funds. I would agree that perhaps they are along the way, but they're a long way from what we need and what we can actually do.