Brad DeLong's Semi-Daily Journal (2004)

I'm heartened to note that my retirement account is proceeding to grow according to this sage advice.

Always good to have someone advise you to do what you're already doing :D

Surely the spread of a 401k across bonds and stocks should vary according to an individual's place in life/work cycle? A 60/40 stock/bond split for a younger person seems way too conservative... Can anyone give me a better informed view?

I echo Roland's question. I'm 30, and I have a 80/20 stock/bond split. I always understood that young people should invest more aggressively. Is 60/40 a good rule for all ages?

I have the same question as Roland and Matilde. I'm 35 and have the entire 401(k) in stocks. Why is this a bad idea?

Brad quotes Economist: Ms. Munnell offers a common-sense solution...

Quick check on Schwab shows 10-year S&P average annual return 11%, intermediate goverment bonds 6%. So if you move 40% of your portfolio from stocks to intermediate goverment bonds, you'd loose, what, one-quarter of the returns? Some common sense.

Risk risk risk risk risk. If you're young, you can almost surely take more risk. If you're approaching retirement age, you probably don't want your entire portfolio in anything like equities.

Brad writes: If you're approaching retirement age, you probably don't want your entire portfolio in anything like equities.

First, the article says "automatically enroll eligible workers; set worker contributions at a level to take full advantage of employers' matching contributions; put 60% of money in stocks and 40% in bonds;". No age qualification, nor any qualification in regards to the business cycle.

Second, you do not have to take all your money out of the IRA at once. If you are going to live another 25 years after you retire, why stop investing and go to annuity at 62?

Third, the number 60/40 someone pulled out of his ass looking at the stocks/bonds correlations a quarter-century ago. If someone can point me to any work that would show 60/40 mix was risk/return superior over the last 10 years, I would appreciate it.

"automatically enroll eligible workers; set worker contributions at a level to take full advantage of employers' matching contributions; put 60% of money in stocks and 40% in bonds"

Ain't gonna happen. Bush & Co aren't doing this to fix SS, or to allow workers to have better retirements. The purpose is to generate fees for the Wall Street firms that will handle these accounts. Giving people a reasonable portfolio they can leave alone and let grow will not provide the big fees. So it ain't gonna happen that way.

I'd add another stipulation: no more than 5% of contribution (if that) in employer's stock. (See Enron, et. al.)

Roland/Matilde/Tyrone/Leopold: The 60/40 split would not be ideal for everyone, but optimization is not the point here. These ideas are geared at nonparticipants, and at participants who do inadvisable things like having their retirement funds completely in their own company's stock or (on the other hand) keeping it all in the money market. Key line at the end of the article: "Anyone could opt out of these choiced, but no one would end up with a stupid choice by default."

Linkmeister's is the best idea here - no one, no matter what, should have 401k money in their own employer's shares. Merely working for someone puts enough of your eggs in one basket. Specific splits are silly - even assuming all or most of your retirement funds are in 401k, a 25 year old and a 55 year old are looking at very different risk profiles, and thus very different risks. And if you have an IRA or non-sheltered funds, you have other incentives. (For one thing, you can move money around inside a 401k without tax problems, so you should have as much of your stock portfolio there as possible. Unsheltered funds can be put in municipal bonds to shelter them - you can get your optimal split that way.)

I'd wager that the *average* investor would have done much better with 60% stock index fund, 40% bond index fund than with the choices they actually made.

“Do we want people to coach Little League less and learn more about finance?”

Well yes. I vote for less baseball and more knowledge about personal finance so people won’t make stupid choices like buying a variable annuity for his retirement. Since most of us have to make important financial decisions in our lives, I’d put courses in personal finance ahead of economics courses unless you want to become an economist.

“A 401(k) plan can be a better deal than an old-style pension.”

Very unlikely. An old-fashioned defined benefits pension (like employees of the UC system get) really beats a 401(k). With a 401(k), you the retiree bears the market risk. With a defined benefits pension, the employer bears the market risk. In some circumstances, you could do better taking a lump cash out and then buying an annuity. Here’s one possible example for you University of California employees. The UC retirement system actuaries don’t use sex differentiated life tables. They use a set of life tables that over estimate the longevity of men, and under estimate the longevity of women. As a result, women get a better deal at the expense of men because they collect (as a group) the stream of cash flows longer. But if a man takes the lump sum instead of the defined benefit, he gets more than he would if the correct life table were used. He can then take the cash out and buy a (fixed) annuity that that does use the correct sex differentiated life table. Why does the UC do this? Answer the courts make them do it.

No 401(k) can possibly top defined beneift pension plans because of the massively higher costs that any individual defined contribution plan must deal with. first, the administrative costs of keeping track of all those personal accounts eat up a lot of money. Second, because random risks such as stock slumps and longevity issues must be borne individually, each investor needs to build in a margin of error to deal these risks that is greater than the average risks that a pension fund must deal with as A. Zarkov alluded to in his post. Such risk spreading is the reason that insurance policies are cheaper than having everyone self insure for fires, car wrecks, etc. Those of us concerned about retirement security would therefore be better off looking for ways to make defined benefit pensions portable (a task more important now that Rep. Gephardt is retiring and will not be around to champion this issue).

I agree with the concerns about complexity. Would the portfolio be rebalanced to reflect the appropirate mix of stocks, bonds, and whatever else, as a function of time to retirement? If you started your career at 18-22 with 60 or more percent in stocks, human inertia would go the wrong way after awhile. How about different risk profiles depending on property ownership and industry segment? If they are portable, how should they be rebalanced when the employee moves or changes jobs?

OK, maybe the property ownership and risk profile of industry part is a little precious. But it seems like there would have to be some default rules for changing the portfolio and serious monitoring of their implementation at least as a function of age. Otherwise people would choose theire way into portfolios that are too risky by default as they age.

There is a lot wrong with US social security, but it sure does spread the risk better than anything I can think of. Next best would be a funded program with investment pools run by quasi-public investment boards.

And I have a feeling that many people would tend to lose their inertia whenever there was a bull market, which might not be a good thing. It is easy to say "let the winners win and the losers lose" when most of the risk takers are high income professionals. But if this plan is for everybody then the lack of risk pooling could have serious consequences for society.

There is ample evidence that left to his own devices the average individual cannot manage his 401(k) account prudently enough to provide for his retirement. We need portable defined benefit pension plans. Yet the trend is exactly in the opposite direction. Few companies offer their employees a defined benefit plan. Even some of the few that do, such as IBM, have tried to stiff their older workers by forcing a conversion to a “cash balance account.” Both major political parties seem loath to confront this problem; we get only vague platitudes.

There is another important risk: inflation. Most defined benefit plans offer no or inadequate inflation protection. For example the University of California Pension Plan only offers up to a 2% inflation increase, hardly enough to get through a blast of inflation such as we experienced in the 1970s. Moreover it’s almost impossible to buy an inflation index fixed annuity in the US. Bodie and Clowes (“Worry Free Investing”) recommend a portfolio of TIPS (inflation indexed Treasury Bonds) as one way to cope with inflation risk. This might have been good advice 3 years ago, when they were cheap, but today the price is so high for TIPS that the yield is not enough to use for your retirement. There is another problem with TIPS, you have to hold them in a tax-free account, otherwise the increased principal (from indexing to inflation) is taxed just like an original issue discount bond—the so-called “imputed income.” I have no simple solution (as yet) to the inflation risk problem. Any good ideas are very welcome.

A 40-60 split is an average.

A better rule is to have your portion in stocks as something like 100 less your age. So at 20
you are 80% in stocks & at 60 only 40% in stocks.

If you retire at 65 you still have to figure on another 20-30 years of life. So keeping a significant share in stocks is still needed as
an inflation hedge. 22 years at 2% inflation means that price levels rise 50% before you die.

The main problem with many of these calculations
is that assuming averge returns can be very misleading.
If you have a poor stock market early in the
time period it can massively reduce your total return even if the average return assumption is still valid.

A. Zarkov, I disagree with your thinking:

"Since most of us have to make important financial decisions in our lives, I’d put courses in personal finance ahead of economics courses unless you want to become an economist."

The average citizen will get hosed by their political choices long before investment decisions. If you don't understand economics, you will have a very hard time making prudent voting decisions.

As an aside, I've been led to understand that there are very serious tracking problems with defined benefit portability. Is this problem why most politicians aren't fighting for it in your estimation?

___

Overall, IMHO it seems like reasonable advice for default investment positions. She could make the advice much more complicated to take in some of the issues being raised here, but the more complicated the less likely it will be heeded. For the average worker, this is reasonable start. With its major technological revolution and stock market bubble and collapse, we also probably don't want to base too much investment advice on these last ten years.

Not gonna happen. The automatically enrollment of eligible workers. Not gonna happen. That automatic enrollment is an automatic employer's expense- and they're in the business of keeping wage and benefit costs down. An employer wants to offer a pension/ 401(k) to attract workers but if the employees don't take them up on the offer, all the better for the bottomline. It is the simply brilliant, obviously should happen, but ain't gonna option.

I like Munnell's suggestion. As others have pointed out, the 60-40 split is not intended to be an iron law. It's just a common-sense default position. Clearly individuals may be wise to use a different allocation, based on age, other assets, etc., but you have to have some default. Why not this one?

As far as defined benefit plans are concerned, it seems to me that there are considerable risks associated with these. I'm not familiar with how the insurance scheme works, but haven't there been cases of retirees losing promised benefits when their former employers went bankrupt, or were merged? Haven't some companies (IBM?) switched plans and reduced benefits some workers thought they had earned? And isn't there a temptation, when it's time to downsize, to consider pension savings in selecting those to be fired?

Stan: I’m sorry but I have to disagree with the idea that your political choices will hose you before your personal finance decisions. Your financial well being is far more sensitive to the latter. On exception might be unionized workers. If they have a politician that fights for the union’s special privileges then that could be an overriding determinant of their income. Moreover I don’t think that economics courses will have that much of an affect on political choices. But people need to understand things like discounted cash flow and most importantly risk. It’s all about risk. They also need to be well grounded in the concept of correlation so they don’t own stock not only in the company they work, but anything correlated to their company’s business. They have to learn to diversify their lives.

Forget about portability, industry is abandoning defined benefit plans altogether because they don’t want to bear the market risk. And the politicians of both parties give industry what it wants. Look at the bipartisan support for the H1b worker visa program that had put so many well-paid American out of work.

Stephen Fromm is right that a 60/40 stock/bond balanced fund investor would have done pretty well. For example, Vanguard's balanced fund had a 10-year trailing average annual return of 9.3% as of the end of last year. This is a bit better than Leopold's comment would suggest, in part because the intermediate government bond return cited is lower than the broader bond market return -- 6.95% p.a. for the Lehman Aggregate Bond Index. To echo Stan's point about the information value of the trailing ten years of stock returns, though, I'm not expecting to see anything close to 7% returns on bonds over the next couple of years (depending on if/when the economy recovers to the Fed's satisfaction).

My view is that defined contribution plans are great for those disciplined and/or well-off enough to sock away large amounts, ideally starting early. (I agree with Bernard that there are significant risks to private defined benefit plans, too.) But a big part of the problem (that Munnell's suggestions largely address) is that the behavioral economists seem to be telling us that we shouldn't assume that everyone else is freely and rationally choosing to under-fund their retirements -- which, going back to the recent Social Security thread, seems to be a central and highly dubious assumption underlying Friedman's critique that mandatory participation in Social Security is unreasonably coercive.

The other big part of the problem is the large fraction of the workforce with no access to employer-sponsored retirement plans of any sort. And when A. Zarkov says "we need portable defined benefit pension plans," it does bring to mind a certain Federal entitlement program...

Re being hosed by political decisions before financial ones: a single person's vote, donation, etc. is infinitessimally likely to be pivotal in any given political situation and therefore infinitessimally likely to affect one's wellbeing.

A single person's investment and wider financial decisions WILL affect their future wellbeing in a significant way.

The default could be the '100 minus your age in bonds, rest in equities' rule suggested earlier.

Should the default require some degree of international diversification?

Tom Bozzo wrote, "For example, Vanguard's balanced fund had a 10-year trailing average annual return of 9.3% as of the end of last year."

Exactly. My guess is that the vast majority of small investors would be better off in a single Vanguard balanced fund (e.g. Vanguard STAR).

"...I'm not expecting to see anything close to 7% returns on bonds over the next couple of years..."

No kidding. I'm into shorter durations now. I can't see how nominal rates (e.g. intermediate duration rates) won't increase somehow in the next couple of years.

Spencer has a totally excellent point:

"The main problem with many of these calculations
is that assuming averge returns can be very misleading.
If you have a poor stock market early in the
time period it can massively reduce your total return even if the average return assumption is still valid."

Which I was going to also mention in my previous post, but typing and thinking at the same time is not one of my strong points. If you are in a cohort that is unlucky enough to hit a period where stocks are doing nothing or not much for five or ten years at the beginning of your career, then the magic of compounding turns into a curse of compounded low returns. The effect on your final asset value is bad in a nonlinear way. That is why talk about the 25 or 50 or whatever year average annual return of the stock market is misleading.

A good portfolio should include something other than stocks and bonds. For most high income people, it does. At least they have human capital.

So I suggest adding mandatory town hall meetings to go with the automatic account to help people sort through the options as their accounts evolve. Or forget the idea of individual accounts, at least unless it is a supplement to a basic retirment plan.

I didn't define "basic" did I? Well, maybe some one here has a suggestion for that, if they don't like the current system.

A.Zarkov, I agree. I needed to fill out my statement. It should read: The average citizen taking personal finance will get hosed by their political choices long before an average citizen taking economics courses will on their investment decisions.

A background in economics will provide more than enough information needed to make the types of financial decisions most investors will ever encounter. It will certainly provide plenty enough information to gain the most out of any investment advice being offered.

On the other hand :), having only a background in personal finance leaves voters open to fleecing by political interests and will likely lead to continuing poor policy decisions. The economic costs of our poor policy choices are much greater to the nation and the quality of voting than the gain of taking a personal finance courses over economics.

A.Zarkov, I meant to add that I believe you are reading too much into the support for H1B visas. There are detracters of H1Bs on both sides of the aisle.

I've not really followed any debates on the defined benefits issue but I do believe there would likely be more visible support without the perception of workability issues. The issues appear to be very large to me.

I do doubt it would be enough to really change things due to the risk shift. Hopefully information technology improvements and better cost disclosures will improve the efficiency levels of defined contribution plans.