Network News

X My Profile
View More Activity

Klein smackdown watch: health-care and wages

EPI's Larry Mishel has a new paper arguing that changes in health-care premiums do not drive changes in wages. He names me as someone arguing otherwise, and he's right to do so: Looking back over the paragraphs he quotes, they certainly sound like I'm saying wage changes are simply the product of changes in premium costs.

As Mishel says, that's obviously not true. Health-care costs aren't big enough to drive wages. For instance: This year, health-care costs will grow very slowly because we're in a recession. But that doesn't mean wages will rocket upwards. Quite the opposite, in fact.

That doesn't cancel out the fact that there is a much larger tradeoff between premium increases and wages than workers understand. Imagine a worker making $55,000, and whose employer pays for a $14,000 health-care plan for the worker and his family. If health-care costs grow by 10 percent that year, that's $1,400. Assume the worker pays a bit of his plan's costs out-of-pocket and the employer is actually spending $1,100 keeping up with premium increases. That's $1,100 that's going to that worker's compensation, but not to his wages. Over the years, that sort of thing adds up, and workers never even know that it's happening. That essential ignorance is a serious problem in the health-care system.

But as Mishel says, it certainly doesn't add up to the entirety of all wage changes for all workers. If I gave folks the opposite impression, that was poor writing on my part. Apologies.

By Ezra Klein  |  January 7, 2010; 3:45 PM ET
Save & Share:  Send E-mail   Facebook   Twitter   Digg   Yahoo Buzz   StumbleUpon   Technorati   Google Buzz   Previous: Use your debit card!
Next: For people who like scary stories


When costs go up, employers pass the cost on to employees.

When costs go down, employers pocket the difference.

Why is this so hard to understand?

I think that Wyden's plan to eliminate the employer exemption, credit employees for whatever premiums employers are paying, and expand the exchanges to include a medicare buy in is the way to go.

Posted by: srw3 | January 7, 2010 3:57 PM | Report abuse

So you admit that lower health benefits do not produce a 1:1 increase in wages. Now consider that the CBO said 82% of the revenue will come in the form of higher income taxes. Their assumption must be pretty close to a 1:1 relationship (since the Cadillac tax is 40% whereas the average income tax rate is 20%). Are you prepared to repudiate the CBO estimate? I am.

Posted by: bmull | January 7, 2010 4:02 PM | Report abuse

The fallacy is deeper than that. The wage/health insurance = compensation is to mistake an accounting identity for an existential reality.

In the more vulgar versions of classical liberal economics compensation is set right at the level of marginal productivity and so you get a one to one tradeoff between health care insurance costs and wages. But mostly this seems to have been used as a blind by employers not wanting to pay for health insurance to start with.

If you throw away your EMH blinders you see a different story. It takes a certain amount of money to recruit, train and retain a worker. Some of that money is carried on the books as compensation some isn't. Is the cost of heat in an employee considered compensation? No. How about the cost of the employee lunchroom? Probably not. On the other hand I had a job where all the appliances and supplies in the break room were paid for by subscription, if you didn't pay monthly you were not supposed to use the fridge. A few years latter I moved to a department on another floor and all those costs were covered. Did that make my now free use of the fridge compensation? To me yes, that was $3 a month in my pocket. But I don't think it hit the department books that way.

Is a bus pass compensation? An employee parking space? Probably not. How about a car service in Manhattan? Ask Tom Daschle. A company picnic? Nope. A catered lunch for executives? No. A country club membership? Yep. Can you deduct that portion of the food and drink incurred by a guest? Maybe yes, maybe no, depends on how good your accountant is (and if necessary your defense attorney).

In the world we actually live in employers incur all kinds of costs to retain staff. It might cost millions to build a pleasant workplace within convenient commute of neighborhoods with good school districts. Is the extra property tax that might be incurred considered compensation? Nope. Could that location induce an employee to choose to work there in lieu of better wages elsewhere? Of course, that kind of thing happens all the time.

The attempt to carve one part of the total cost of supporting an employee and calling it "compensation" and pretend that it and it alone has some fixed relation to marginal productivity is nonsense. Some jobs require you to bring your own tools, most jobs have at least an implicit dress code, does any economic text book consider that a tariff on compensation?

"When costs go up, employers pass the cost on to employees.

When costs go down, employers pocket the difference."

Except in the tens of thousands of instances where this doesn't happen. A spike in heating costs for your office building might be passed onto employees generally in the form of job losses. But maybe not, if demand for your product is steady and your operation working somewhere near its efficiency limits, the employer might have to eat the difference. The claim that this is different when the invoice is labeled "Aetna" rather "Podunk Utility District" is just a case of people reifying an accounting concept into an economic operating principal.

Accountants get paid good money to run a razor in between business expenses and employee compensation, that the numbers end up on two different lines in the ledger doesn't have any real world meaning, at least until the boss uses increasing health care costs as an excuse not to increase your take-home, or perhaps to cut back on something you consider to be a benefit that has some time-money value.

I suppose there are plenty of employers who have let this accounting practice become their own reality, but workers don't need to buy into the idea that the tradeoff is just some iron rule of economics.

Posted by: BruceWebb | January 7, 2010 5:01 PM | Report abuse


Read the comments from folks on the original posts-- the same criticisms were made there. It simply wasn't poor writing IMO-- the graph that you posted multiple times was similarly false evidence as the excerpt that you now seem to be recanting.

More simply, while there may be a case for some wage growth to be lost to health care cost growth, you overstated the case in multiple posts.

You were clearly arguing for a close to 1:1 subsitution, which was the primary objection in my comments. But I remember a few others that were good as well.

Posted by: wisewon | January 7, 2010 5:13 PM | Report abuse

I don't think this negates the near 1:1 substitution.

Total compensation trends are clearly driven by conditions in the labor market, both in the aggregate and for various groups within the labor market (high skilled/low skilled, etc).

That doesn't mean that health cost growth doesn't offset wage growth. Firms provide health care to workers because they are being nice. It is part of the overall compensation package. If health care costs were cut in half, and some employers pocketed the difference, smart employers could steal all of the good employees simply by offering higher wage and/or better benefits. A firm that pays its workers on average less than it thinks they are worth will likely find itself stuck with mediocre workers.

Conversely, if in the 1990s healthcare costs were rising 12%/yr, salaries on the high end probably would have stagnated - sure, they still would have their health benefits, but salaries would have barely grown. Companies aren't going to just hose themselves by paying higher wages and simply accept a rapidly growing healthcare bill - if the worker's aren't worth that level of compensation, then the company doesn't have to increase wages. Or it could continue increasing wages but use co-pays, deductibles and higher premiums to shift some of the health costs onto employees - effectively the same as cutting wages, although less costly to the firm because it is hard to cut nominal wages and healthcare is tax advantaged in any case. If some other firm is willing to pay more for those employees than the firm thinks they are worth, well, more fool them.

Granted, it is difficult in practice to figure out precisely what a worker is worth, and how to precisely balance healthcare and other compensation, but in general market forces tend to push employers to the proper level - those who continously overpay have a tough time making money, and those who continuously underpay have a tough time finding talent.

Posted by: justin84 | January 7, 2010 7:05 PM | Report abuse

Ok Ezra, but doesn't that suggest that unions and other folks skeptical of the Cadillac plans are skeptical because they are justified in worrying that their employers will pocket the difference, especially if these kinds of decisions are so opaque? (

Posted by: StevenAttewell | January 8, 2010 2:45 AM | Report abuse

The comments to this entry are closed.

RSS Feed
Subscribe to The Post

© 2010 The Washington Post Company