« The Risk Pool | Main | Dependency Ratios: one last time »



"One reader points out that Detroit did support the Clinton healthcare initiative, so perhaps their failure to speak up right now around is strategic"

Detroit initially supported the Clinton health care plan, but gradually backed away under pressure from the Republican coalition.


This is a terrific article. I recall being angered by a Jack Welch appearance on George Stephanopoulos' show about 6 months ago. Welch talked about the huge burden that health care and pension costs had placed on companies, who therefore couldn't compete--his point being that they needed cost relief of some form. Which of course, ends up coming from labor, not management.
I recall thinking at the time that it was not so many years ago that management had made these very decsions that are coming home now to roost--yet where is the distribution of pain to management bonuses?
Malcolm has given far greater precision to not only that issue, but to, as he notes, issues of global competitiveness and social health care policy. One of your classics.
I think in the broadest sense what Malcolm Gladwell writes most and best about is the general topic of how it is that human beings organize themselves to deal with the issues of being social creatures. Those issues include (famously) crime, but also risk spreading, treatment of the homeless, caste systems, and the relationship of education to corpocracy.
Write on!


An article from the Detroit News outlining what GM is doing about health care costs.


Malcolm,I must say your Dependency ratio notion is still very much valid.The point is how to fix it and what to do with our social security system as a whole....
any take on G.W.Bush's proposal of 'personal savings account" stuff?


One small point that I thought you would eventually come to in the article, but never did, is the fact that, while spreading the risk of obligations across a larger group (i.e, a national healthcare system) does not eliminate fluctuations in dependency ratio, it certainly tames them. The dependency ratio of the United States is obviously not a constant, but it is infinitely more stable than the dramatic fluctuations in the dependency ratio of individual companies due to technological advances or market shifts. Sure, the baby boom has shaken up the ratio in the U.S., but G.M.'s changes and problems are much more drastic.


As an experienced accountant at a couple larger global companies, turned Co-Founer of a start-up, I can tell you a secret that many accountants dare say.

"Accountants rely upon future estimates (guesses) to determine today's profits and today's guesses have an impact on tomorrow's profits, as a result."

The problem of pensions and healthcare for retirees is quite complicated, relies on statistics, and is very prone to manipulation when assumption are changed. Only when a company has its back against a wall (e.g. in bankruptcy), is that truth finally revealed. Maytag is a good example of this.

I do hope that the answer isn't that new companies (without these burdens - today) are better, and more profitable, than old ones as that's an oversimplification.


Malcom, are you mentally retarded? McGurky was NOT suggesting an impossible "pig-flying feasibility" alternative to your proposal! (S?)he was pointing out that the number of workers, part of the questionable statistic you used to explain the pension crisis, is irrelevant, and demonstrating that through example. What matters, in order to pay retirees, is the (pre-pension payment) profits! THAT WAS THE POINT! And then, you missed that he was obviously referring to profits *before* paying out pensions. Pension obligations by definition don't affect this!

God damn, why do people even listen to you?

Malcolm Gladwell

Is it possible, do you think, to register an objection without being rude?

Aaron Taylor

Mr. Gladwell, you're a very interesting writer but you get a number of things wrong here.

- First, let me associate myself as politely as possible with those who are pointing out that the number of current workers are irrelevant to the ability of a company to meet obligations to retired workers. The money for these obligations come from profits (revenues minus current expenses). While the dependency ratio makes for a good rule of thumb, it's only a pithy substitute for the really important ratio: dollars of revenue per retiree. As productivity increases, fewer workers can generate more revenue, thus supporting more retirees.

(Unfortunately, increases in productivity and the revenue it generates are notoriously variable and impossible to predict over the long term. Thus the use of the dependency ratio rule of thumb.)

- If every company turns its pension obligations over to the federal government, the money to pay those retirees still has to come from somewhere, ultimately the U.S. private sector. Companies may relieve themselves of a huge burden in the short run, but it will eventually all have to be paid back in taxes.

- Notice that so far I've only talked about pensions, not health insurance. You are conflating the two, which is a BIG mistake. At the macro level, there is no cost savings from moving pension obligations from the private sector to the public sector, unless the government decided to shortchange the pension recipients. But there are potential savings by moving health insurance to the public sector through rationing or (possibly) economies of scale.

- While the cost of health insurance has risen unexpectedly quickly over the last 30 years, the costs of the pensions were fairly predictable and totally within the control of GM and other companies and their unions/workers. You can't blame the companies for the health insurance costs, but the pension costs can be blamed squarely on the companies and their poor management.

DISCLAIMER - The preceding has been the opinion of me alone; not anyone else and certainly not my employer. :-)

aaronleetaylor over at yahoo dot you-know-what

Mr. David

I remember reading an article in (I think) the Atlantic a couple of months ago that takes the point you make here further. The author predicted that rising healthcare costs would force Wal-mart to abandon somewhat its Republican, can-do ethos (which is apparently what prevents many companies from coming out against medical coverage for all citizens) and instead demand universal healthcare. Once Wal-mart declares that it can't afford to pay for its employees healthcare plans, several other companies will likely follow suit, and the chances that it will pass in Washington will be much greater.

(Of course, I may have misunderstood the whole thing, but as I understood it, Wal-mart is domino #1, with the Big Three presumably close in line behind.)

Michael Byrnes

It seems that many of these objections, rude or otherwise, are missing the point.

A company can only turn a profit if its expenses are lower than its revenues. As Gladwell points out, pension obligations are part of a company's expenses.

Unlike other expenses, pension obligations are offset, at least in part, by contributions from the current workers. As the number of workers goes down, a greater percentage of the pension obligations are shifted to the company. Its "reward" for increased efficiency is greater pension expenses.

The real problem with the "dependency ratios are irrelevant" arguments is that there are limits to how efficient a company can be. And those limits are set by the marketplace competition. GM has competitors, and the competition affects the prices they can get for their cars, the quality with which the cars must be built, etc. If GM finds a revolutionary new way to build more and better cars with half of the workforce, they will benefit, of course, but Ford and others will see what GM is doing and figure out a way to do it better.

Gladwell's Bethlehem Steel example is tells the story. Before bankruptcy, they were in the position of having to compete with younger companies with much lower pension obligations. (Ironically, those obligations were created by Bethlehem's long history of success). Those companies could be protitable with revenues that, after pension expenses were subtracted out, would still leave Bethlehem in the red. So Bethlehem goes bankrupt, reorganizes, has no more pension obligations and - presto - is profitable.

The real issue that Gladwell ducked is the 401 pound gorilla. He argues that single company pensions are not sustainable because of the dependency ratio, and thus we should have national pensions. I think conservatives would whole heartedly agree with Gladwell about single company pensions, but would argue that the answer lies with 401k and other defined contribution plans.


The dependency ratio is applicable to economies but not to firms.
The dependency ratio makes sense in economics because the individuals on one side of the ratio (current workers) support those on the other side (dependent workers).
At the firm level, the firm itself, and not the firm's current employees, 'support' retired employees.
As suggested by Aaron, comparing revenue or operating profit might be more relevant.


Michael Byrnes

"comparing revenue or operating profit might be more relevant."

The greater a firm's pension commitments to retired workers, the greater percentage of their revenue must be spent on retired workers. That means less profit, and makes it very hard for a large, old, established business (with big pension commitments) to compete in the same marketplace with a younger company that does not have to kick a huge portion of its revenue to its retirees.

Dependency ratio matters because it has a major effect on a firm's bottom line.

Sarah  Tantillo

Dear Mr. Gladwell, I am a huge fan. While the preceding discussion about dependency ratio is intriguing, I am wondering if you have considered pensions from another angle--how their lack of portability affects career paths, particularly those of K-12 educators, and how it affects the quality of the educational system in the US. K-12 pensions are not portable to other fields, not even to higher ed. This means that once teachers have vested in TPAF (Teachers' Pension and Annuity Fund, in NJ), they are less inclined to take the risk of leaving the classroom to try working in a non-profit or in a university (which would put them into TIAACREF at the starting line again). Professionally, then, people feel as though they have no choice but to stay. Many become bitter and can be heard saying things like, "I've got 10 years left," as though imprisoned, which in some senses they are. This situation limits their professional growth at best and damages the lives of children at worst.

Malcolm Gladwell

I think that there is a simple way to clarify a lot of the debate here over the usefulness of dependency ratios in describing the fortunes of individual companies. Is it, in theory, one of the most useful statistical indices of how well a company is faring? Probably not. As many of you have pointed out, what really matters is simply how profitable a company is. But I'm not interested in formulating a general theory of economic health. I was, in "The Risk Pool" specifically interested in the fate of old line automakers, steel companies and airlines. And in that specific case are dependency ratios useful ways of understanding their problems? Absolutely. The world is far too complicated to capture in a single New Yorker article. It is possible, though, to make sense of one part of the economy. And it is useful, I think, to try and do so using a lens (demographics) that too often gets overlooked in our understanding of economics.


I agree with your article however I think some of the confusion comes from how you laid out your argument. Dependency ratio is another measure of fixed costs. GM's cars cost more because of the pension benefits that GM has to pay. Since their cars cost more, GM sells fewer cars. GM then has to lay off workers due to declining sales which increases the number of pensioniers. Its a death spiral.

Anil Pal

No one so far apears to have mentioned the fact that GM incurred the pension obligations in exchange for lowered current compensation for its employees in years past. In other words, the past employees effectively loaned GM capital. If GM had invested this capital in future production capability (automation, technological advances, worker training) rather than management bonuses, it would now have a productivity advantage over its rivals (or else they would have to have invested similar amounts of capital, and incurred comparable costs to repay the investors) and the ability to pay out the pension obligations from the return on the invested capital.

malcolm gladwell

For the record, Bloom and Canning (nor I) never argue that the sole reason for the Irish economic miracle is that their dependency ratio fell, or that countries with lots of working age people and few dependents automatically succeed. Ireland's success was simply aided (albeit powerfully) by demographics among other things, and the so-called "demographic dividend" can only be exploited if a country also pursues intelligent economic and social and political policies. I thought I stated this clearly in "The Risk Pool." Certainly Canning and Bloom do, repeatedly, in their writing, for those inclined to look before they leap to conclusions.


Malcolm, don't let the nitpickers get you down. It's great to see a famous author responding to comments.

Gabriel Roth

The problem with the dependency ratio argument still exists even when we confine the discussion to struggling industrial-age companies like automakers, steel companies and airlines.

The dependency ratio argument says those companies are failing because they have too many retired workers relative to too few current workers. That's exactly half right. The fact that GM et al. have so many retired workers makes for high costs, which cut into profits. But the dependency ratio muddies the picture when it comes to the revenues side of the equation.

The reason the industries in question are facing revenue shortfalls is not because they don't have enough current employees -- it's because people don't want to buy their products, due to competition (from Japanese carmakers and budget airlines) or waning demand for their product (as with steel). If they wanted to, they could improve their dependency ratios by hiring a bunch more workers. And yet they don't, because they recognize that adding workers and increasing production wouldn't help them. Instead, they're laying off workers and cutting production (as with Ford's announcement earlier this month), making their dependency ratios even worse! Is that a foolish thing to do? If not -- if it's economically rational for Detroit to lay off workers -- then the dependency ratio theory has some explaining to do.

The demographic problems at the companies in question don't stem from a ratio (too many retired workers/too few active workers). They stem from an absolute: too many retired workers.


I've had the opportunity to hear Rick Wagoner speak to an internal audience of GM employees on a few occasions. Bemoaning GM's legacy costs, Wagoner claims to have asked numerous times for "assistance" on health care from the current administration in Washington. While he did not go into detail, I assume he (and the auto industry lobby in general) are calling for some mild version of national health care. Wagoner says that the White House and Republican members of Congress -- who presumably cater to the desires of big business -- consistently turn a deaf ear to GM's concerns and this issue in particular.

It's so sad when he tells the story, I almost want to cry. Because if the Republican Party doesn't give a damn about the financial viability of American corporations, who the hell will?

Jane Galt

I'm afraid I don't understand how nationalised health care would help. Don't we have nationalised health care for GM's retirees at least as generous as that which is available abroad? Why hasn't Medicare solved GM's problem? And how would putting everyone under 65 into the scheme help GM pay for its retirees? Or are we talking about Canada/North Korea style "no one may have more generous health care than what the government provides" plans only?


"The demographic problems at the companies in question don't stem from a ratio (too many retired workers/too few active workers). They stem from an absolute: too many retired workers."

I think Gabriel makes an interesting point, and I'd like to take it a step further. One thing we need to keep in mind is that on a nation-to-nation basis, the US industries that Malcolms is discussing have one thing that China, Japan, and Africa don't have...large numbers of retired workers WITH pensions. A favorable "national age balance" doesn't necessarily reflect how many of the retirement age individuals are supported by private industry (instead, they're supported by their families, which further strains the economic prosperity of the nations in question).

But what happens in 30 years when international competitors' employees have retired and become a "fixed expense"? All things in time with respect to the global economy.

I suppose my point is that Irelend compares more comparably to the US in terms of measuring "success" by dependency ratio than other countries, whos industrial development post-dated ours.


Interesting article. Here in Malaysia and Singapore employees are required to deposit on a monthly 11 percent of their basic pay into a centralised provident fund. Similarly employers are also mandated to deposit a matching amount to the fund. The fund cannot be utilised until the employee reaches 55. The account created in the fund is permenant and employees are free to move from one job to another while still keeping their fund account. The advantage of this is that even if the employer gets bankrupt there is no risk to the employee as the retirement fund is protected with a central organisation. similarly companies do not need to set aside any particular money for future benefits. In a sense it is pay as you go.


"I'm afraid I don't understand how nationalised health care would help. Don't we have nationalised health care for GM's retirees at least as generous as that which is available abroad? Why hasn't Medicare solved GM's problem?"

Hundreds of thousands of GM retirees and their spouses are too young to qualify for Medicare. These are people who started working in the plants at the age of 18 and retired at 50. And when they reach 65, they usually continue to make use of their private health insurance plan because it fills in the many "Medi-gaps" of the national system (like dental and prescription). I'm not familiar with national health care systems elsewhere. Why should I assume that Medicare is as generous as they are?

The comments to this entry are closed.

My Photo


  • I'm a writer for the New Yorker magazine, and the author of four books, "The Tipping Point: How Little Things Make a Big Difference", "Blink: The Power of Thinking Without Thinking" and "Outliers: The Story of Success." My latest book, "What the Dog Saw" is a compilation of stories published in The New Yorker. I was born in England, and raised in southwestern Ontario in Canada. Now I live in New York City.

    My great claim to fame is that I'm from the town where they invented the BlackBerry. My family also believes (with some justification) that we are distantly related to Colin Powell. I invite you to look closely at the photograph above and draw your own conclusions.

My Website


  • What the Dog Saw

    buy from amazon


    buy from amazon

    buy from amazon UK


    buy from amazon

    buy from amazon UK

    Tipping Point

    buy from amazon

Recent Articles

Blog powered by Typepad