401k Fees are Not the Problem

Park Bench Crop Dēmos is one of those lobbyist think tank organizations. I do not know if there is a better term for that kind of outfit, but I think we all know what I am talking about. They raise money, issue reports, and generally attempt, constructively or not, to influence debate on one or more issues. There are hundreds, maybe thousands of them, filling a spectrum from far right to far left.

Dēmos  (the bar over the e tells you it is pronounced deemos and that they are hugely sophisticated) describes itself as “a non-partisan public policy research and advocacy organization.” A quick scan of their website will make it clear that they are towards the left end of the political spectrum. That is not where you will find me, but this is, last I looked, a free country. They can write about the tragedy of declining union membership and the subversive nature of Louisa May Alcott’s Little Women, and I can ignore them.

But last month they put out a “report” entitled The Retirement Savings Drain: The Hidden & Excessive Costs of 401ks. Sadly, it was not as universally ignored as it deserved to be.

The paper was a follow-up to one Dēmos put out last fall, The Failure of the 401(k): How Individual Retirement Plans are a Costly Gamble for American Workers. In case those titles leave any doubt, I will cut to the chase. Dēmos is fundamentally opposed to what investment professionals call defined contribution (DC) plans, 401ks, IRAs, Keoghs, etc.

If I am being honest, I have to admit I am not all that happy about them either. Making every worker his own pension fund manager means that almost all workers have incompetent pension fund managers. And although the newly retired could purchase annuities with their nest eggs, insuring that they will not outlive their savings, for whatever reason practically none of them do, meaning that a substantial portion will wind up broke in their final years.

But to paraphrase Churchill’s cliché, DC is the worst possible system except for all the others. Old style pension plans, known as defined benefit (DB) plans, are loved only in the abstract. In practice, private companies do not like being pension fund managers. Workers prefer to have direct control over the money set aside for their old age. And public DB plans are threatening to bankrupt governments throughout the country.

But Dēmos does not seem like the sort of outfit that would let the unpopularity of DB plans amongst the workers they are to benefit slow them down. And the unpopularity of them with those nasty employers is likely a plus. Their pet scheme, what they call Guaranteed Retirement Accounts, is a sort of Social Security 2.0, a government run, subsidized, and guaranteed DB plan.

It is a stupid idea for any number of reasons, including the epic unlikelihood of the government managing it well enough to achieve the guaranteed inflation plus 3% return each year without infusions from the Treasury. But, as I said, it is a free country, and even dopey proposals deserve to be contributed to the marketplace of ideas. This is an important issue of public policy and we should be able to have an honest discourse on it.

But Dēmos, apparently, cannot keep inside the realm of honest discourse. Their paper on the costs of 401ks is not a reasoned fact-based argument but a screed based on half-truths and sheer fiction. That a reasonable, if unexciting, criticism of 401k fees could have been made instead only makes it that much more unforgivable.

Their headline is that about a third of investment returns otherwise due to the innocent 401k saver is taken away in fees, for a lifetime total of $155K for an ordinary American household, enough, Dēmos tells us, to buy a house. As such things go, that is quite an attention grabber. Alas, Dēmos had to work very hard to concoct those numbers.

They start with the expense ratios for the mutual funds that make up 401ks, what most of us think of as the fees. The median expense ratio paid by 401k participants is, the paper tells us, 0.78%. (Although if you read the appendix that details the ordinary household calculation, it turns out they used an average fee of 0.835%.)

They then proceed to double this number by introducing sinister sounding trading fees disclosed only in “cryptic” and “arcane” statements in mutual fund documents. In fact, these fees are so obscure that Dēmos can only estimate them by assuming “given previous research” that they are equal to the stated expense ratio, thus the total fee level is twice what you thought it was.

That previous research is apparently a paper from the BC Center for Retirement Research. It used an equation I have some issues with to estimate that for the larger mutual funds, the kind you are likely to find in a 401k, total transaction costs were something like 0.50% annually. To be clear, the paper did not look at actual costs, it just projected what the authors thought those costs might be based on theory. And they still came up with a number a lot lower than what Dēmos uses, allegedly based on their research.

As it happens, mutual fund transaction costs are one of my few areas of actual expertise. I worked as a quantitative analyst at a large mutual fund complex for a number of years and did several studies of transaction costs using the firm’s proprietary trading data. And I can testify that 0.50% is on the high side, although just plausible as an average annual cost.

Moreover, the reason I did all that work back in the good old days of having a job is that my bosses were desperate to get that number lower. Dēmos blithely states that mutual fund managers “have no incentive to minimize hidden trading costs” and generally lumps them in with the other fees paid to evil Wall Street.

This is simply false. Mutual fund managers live and die by their performance results. Reducing transaction costs directly increases returns. Picking better stocks might increase returns even more, but guess which one is easier to do.

Of course, mutual fund managers have an obvious incentive not to reduce the expense ratio, even though that would also improve performance. But unlike most of the stated fees in the expense ratio, mutual fund managers do not get the transaction costs. Indeed, nobody gets the great majority of them, because they are not fees such as brokerage commissions, but what is called market impact. If you want to buy a stock, particularly in large quantities, you will drive the price up a bit and so pay more than the quoted last trade. That is legitimately a cost, but it is not a fee paid to anybody in particular.

More to the point, Dēmos’ implied premise, that there is a meaningful true return to a mutual fund with actual fees and transaction costs added back in, is nonsense. Yes, it is reasonable to imagine the same fund being managed at a lower expense ratio. (Although the manager could not do it for free.) But without transaction costs, that is to say without trading, there is no active fund and no stated return to be improved by adding back fees and costs. It is like saying that driving a car would be a lot cheaper if it were not for buying gas.

It could be argued that transaction costs, whatever they are, could be lower for the average mutual fund if it was managed better. But given the economic incentives involved, it would be unreasonable to suppose that they could be very much lower than they are now.

Moreover, none of this is really an argument against DC plans at all. It is an argument, and an old one, against actively managed mutual funds, which have high expense ratios and incur significant transaction costs. Index funds do neither and are widely available in 401ks and IRAs. (Indeed, the superiority of index funds and ETFs was the point of the BC paper.)

So why is so much of our DC savings in active funds rather than indexes? Basically, because the humans making the decisions, primarily the individual savers themselves, are choosing poorly. That bothers me, but not nearly as much as it bothers the folks at Dēmos.

The unstated belief of Dēmos and their ilk is that although they are sincerely in favor of freedom of choice in principle, if it allows people to chose foolishly, then that freedom has gone too far. I think I am with the great majority of Americans in believing in the freedom to choose to do even stupid things, providing I do not harm others.

And I think that at some level even Dēmos understands that it is in the philosophical minority on this. Instead of criticizing 401ks because they give American workers enough rope to hang themselves, a cogent and reasonable objection that would be widely ignored, they attempt to whip up populist sentiment against (largely imaginary) Wall Street fee gouging.

If this bit of anti-capitalist propaganda had stayed within the world of fringe websites and publications I would just ignore it. But it leaked out into our mainstream personal finance media, with predictable results.

MSNBC “reported on the ‘stunning’ findings in Demos’ new report.” US News’ My Money also bought the Dēmos numbers as did Time’s Moneyland. The Time piece calls Dēmos a “research group.” It does quote informed sources saying that the Dēmos report is nuts, but then judiciously says “Let’s just assume reality is somewhere in the middle.”

Some people say the Earth is flat, others say it is a sphere. Let’s just agree it is a cylinder and move on.

SmartMoney carried a post Will New Regs Solve 401(k) Fee Mystery? that led with the $155K number from a recent report from “Demos, the non-partisan public policy group.” The WSJ Total Return blog also reported the Dēmos figures, and without so much as a hint that some found them objectionable.

Paying for our retirements is both an individual and collective problem that, sooner or later, we will have to face. This is not helping.

No Comments

No comments yet.

RSS feed for comments on this post. TrackBack URI

Leave a comment

WordPress Themes