David R. Henderson  

Enough to Buy Back the Product

Is Russia's Government Hostile... Murder: A Socratic Dialogue...


I was on an email discussion this morning with some free-market economists and some economically literate fans of free markets. One of them surprised me with this statement:

I always tell my students the Henry Ford $5 a day story in 1914, a great example of management looking out for labor and a "win win" formula for success. It stopped Marxism's two big arguments: that capitalists exploit workers and that capitalism alienates workers, forcing them to produce products they can't afford themselves. But when Ford overnight raised workers pay to $5 a day, it was the first big example of sharing the profits -- and Ford workers could buy Model T's for the first time, thus destroying the Marxist argument of alienation.

There's nothing necessarily incorrect in his actual statement above. But there is something misleading. It may well be the case that Ford workers could buy Model Ts for the first time because of the increased wages, but he makes it sound as if that was Henry Ford's motive. That's unlikely to have been Ford's motive because the number of Ford workers was probably well under 1% of the number of potential buyers. Would it really make sense to pay more than necessary to get workers so that your number of potential buyers can increase slightly, especially when the downside is that with the higher wages come higher costs and your larger potential market is likely to shrink?

I think it was a fortunate accident that Ford's workers were able to buy Fords. The reason I've read for Ford's success is that Ford was plagued with high turnover and high absenteeism and paying above market reduced turnover and absenteeism substantially. One can imagine this being particularly important in a relatively new industry and certainly an industry with a relatively new production method--the assembly line--because there was probably a bigger gain than normal from keeping people on the job who gradually got very good at it.

If you still think the "pay them more so they can buy the product" argument makes sense, then consider what Henry Hazlitt wrote on this issue in his 1946 classic, Economics in One Lesson:

Some sponsors of the theory seem to imply that the workers in each industry should receive enough to buy back the particular product they make. But they surely cannot mean that the makers of cheap dresses should get enough to buy back cheap dresses and the makers of mink coats enough to buy back mink coats; or that the men in the Ford plant should receive enough to buy Fords and the men in the Cadillac plant enough to buy Cadillacs.

So a question for the person I quoted above, based on Hazlitt's quote, is this: Were Cadillac workers more alienated than Ford workers?

A few years ago, I used Hazlitt as a supplement to my main text and we covered this chapter. Remember that my students are largely military officers and disproportionately Naval officers. One U.S. Navy student had a great example to make Hazlitt's point. He pointed out it's not a good idea to pay workers making aircraft carriers high enough wages that each of them could buy an aircraft carrier.

Here's an earlier discussion of the issue.

Comments and Sharing

COMMENTS (17 to date)
Michael Makovi writes:

Also, the "buy back the product" argument is literally incoherent. Wages enough to buy the product OVER WHAT TIME PERIOD? Should a person who makes a car be paid enough to buy a new car every year? Every day? Every century?

In fact, everyone is paid enough to buy the product, if you just assume a sufficiently long time horizon. Every worker who contributes to building an aircraft carrier IS paid enough to buy an aircraft carrier, if he just works long enough. (Which may be a thousand years, but who's counting? Not the socialists. They literally don't consider this issue, as far as I know.)

Garrett writes:

Boeing workers should make enough to buy airliners.

fralupo writes:

@Michael Makovi,

I think you're not considering consumer behavior. Consumers have their own preferences and these change with income. A person earning half your wage isn't going to match your consumption habits (just at half the speed) even if they were your clone.

Intuitively, someone living paycheck-to-paycheck may not be willing to wipe out their savings in exchange for a machine that adds costs to their lives and has a declining resale value.

In terms used in economics:

a) People earning low wages usually have higher discount rates than people earning high wages and so purchases of expensive durable goods will seem more attractive to people as their wages rise.

b) Going back to first year economics a consumer's indifference curve for automobiles may intersect with their budget constraint at q=0, and this is less likely the more income someone has.

Hunter writes:

The Ford example is one that drives me crazy. As Dr. Henderson noted the wage increase was to retain experienced workers. This helped keep the prices low enough that the workers who didn't work for Ford were able to buy the product.

Michael Makovi writes:


I think adding consumer preferences just makes socialism even worse.

For example, some socialists will advocate worker control of the firm. The problem is, many workers don't want to control the firm. They'd prefer to make certain, guaranteed wages rather than uncertain, varying profits (and losses). Whether the firm makes a profit or a loss this firm, they'd rather just receive their salary.

And as many professors can attest, having to serve on a governing committee can be awfully bothersome. Many workers would probably prefer to avoid the burden of having to help run the company.

So I think there's a general problem with socialism, viz. that many consumers don't want what socialists think they ought to want.

john hare writes:


a) People earning low wages usually have higher discount rates than people earning high wages and so purchases of expensive durable goods will seem more attractive to people as their wages rise.

I would argue that low wage people should do as you say. An unfortunate number of them live paycheck to paycheck due to poor choices. Rent to own big screen TV instead of cash for a smaller or used unit for example. In my case, I bought a Silverado back when making $5.40 an hour, stupid and still embarrassing if I think about it. .

Mr. Econotarian writes:

We should keep in mind that the Ford assembly line was VERY different from traditional hand-crafted "carriage" manufacture. He needed workers who were very on-the-ball, because a small screw-up could "stop the line". And it is likely that assembly line work, though more efficient, was probably very boring to do one small task hundreds of times per day compared with doing a wider range of handcrafting at a more relaxed pace.

Ford also had a kind of "secret service" who investigated the "moral character" of his employees. He paid a lot, but he expected a lot as well.

I believe Ford did sell is $5 day as some kind of benificence to his employees, but it was only the improved productivity of the assembly line that could allow it, and the stringent requirements of the assembly line also demanded it to bring in the best workers.

Paul Zrimsek writes:

My favorite argument against the "pay them more so they can buy the product" version is: even in the best-case scenario where every penny of the wage increase goes towards a Model T, all Ford has done is find an indirect method of giving away cars.

Matt writes:

Tim Worstall had a bit of fun with this a few years back:

"Car production in the year before the pay rise was 170,000, in the year of it 202,000. As we can see above the total labour establishment was only 14,000 anyway. Even if all of his workers bought a car every year it wasn't going to make any but a marginal difference to the sales of the firm.

"We can go further too. As we've seen the rise in the daily wage was from $2.25 to $5 (including the bonuses etc). Say 240 working days in the year and 14,000 workers and we get a rise in the pay bill of $9 1/4 million over the year. A Model T cost between $550 and $450 (depends on which year we're talking about). 14,000 cars sold at that price gives us $7 3/4 million to $6 1/4 million in income to the company.

"It should be obvious that paying the workforce an extra $9 million so that they can then buy $7 million's worth of company production just isn't a way to increase your profits. It's a great way to increase your losses though."

Jon Murphy writes:

If the goal is to pay workers enough to buy back the product, and many minimum wage workers work in fast food (or other food service jobs), where the products are often $1 or just a little more, wouldn't that necessarily imply reducing the minimum wage?

Khodge writes:

It'd be cool to own my own aircraft carrier and it would flesh out the resume quite nicely but I was unable to run a flower shop so I think I'll pass.

pkn writes:

Uneducated as he was, Henry Ford understood production. With a turnover rate approaching 300% in his work force the slowdown on the assembly line as they trained new workers was very costly. Hence the $5 wage. It ended turnover and production increased accordingly and profits increased.

Michael writes:

Might it have had a psychological impact?

Cars, unlike airliners, were a new product that Ford workers could surely see a use for in their own lives. Getting paid enough to actually being able to afford to buy one might improve workers' job satisfaction.

Hasdrubal writes:

Jon Murphy:

If the goal is to pay workers enough to buy back the product, and many minimum wage workers work in fast food (or other food service jobs), where the products are often $1 or just a little more, wouldn't that necessarily imply reducing the minimum wage?

The funny thing is, $15 minimum wage proponents suddenly remember the actual reason why Ford raised wages when talking about low wage jobs. (And assume it was the wage, not the worker the wage was attracting that led to the improvements.) It's one of those great stories: It can mean anything you want it to depending on how you tell it.

Jon Murphy writes:


Agreed, although there is an aspect of the argument minimum wage proponents who try to tout the minimum wage as an efficiency wage miss: the efficiency wage increases productivity by increasing unemployment.

An efficiency wage is not much more than a self-imposed price floor, whose job is to limit workers shirking on the job. The idea is, if a firm pays above-market wages, then a worker will work harder for fear of losing his job and being forced to take less pay somewhere else. If all firms pay this efficiency wage, a worker will work harder because there is now unemployment in the labor market; there are others who are willing and able to take his job.

So, when minimum wage proponents make the efficiency wage argument, they are implicitly allowing (indeed, calling for) unemployment to increase.

mike davis writes:

Nothing like hijacking one of David’s posts for a short rant…..

The “pay them enough to buy what they make” is an obviously silly argument. But our econ 101 answer, “pay them their marginal revenue product”, isn't helpful--it doesn't do much to explain Ford's wages and leaves students vulnerable to bogus stories like this.

As an opening sentence in a long novel on labor markets “w=MRP” is only slightly better than “It was a dark and stormy night.” Maybe it’s true, maybe it’s not. But it’s boring and misses all the color, passion and drama that are part of the story of real labor markets. Things like efficiency wages, various types of human capital, signaling and the non pecuniary aspects of work matter much more than the first derivative of a profit function with labor as an independent variable.

There’s nothing wrong with teaching simple optimization. These are useful life skills that every freshman or MBA student should have. But when teaching labor don’t leave out the sex and violence parts of the novel just because it’s hard to think up good multiple choice questions.

…rant concluded. (BTW not accusing David of ignoring this stuff in his teaching and writing.)

Dave writes:

[Comment removed pending confirmation of email address. Email the webmaster@econlib.org to request restoring this comment and your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Comments for this entry have been closed
Return to top