The Washington PostDemocracy Dies in Darkness

Nobel-winner Edmund Phelps on his plan to help low-wage workers — without raising the minimum wage

He was awarded the 2006 Nobel Prize in economics for his work on how to determine a nation's optimal savings level, on the foundations of macroeconomics, and on the relationship between inflation and unemployment in both the short and long term. He is generally credited with introducing the idea of a "natural rate" of unemployment.

In the 1990s, he worked extensively on issues of economic inclusion, culminating in a proposal for work subsidies for low-wage workers in his 1997 book "Rewarding Work." His most recent book, "Mass Flourishing," is about how grass-roots innovation drives drives economic growth, employment and job satisfaction, and why such innovation may be declining of late. We talked on the phone Thursday afternoon; a lightly edited transcript follows.

There have been dozens of studies on the employment effects of the minimum wage at this point. What's your take on that literature?

It's one of the subjects that the economics profession has really never achieved a consensus on. Statistical studies are all over the lot about the pluses and minuses of raising the minimum wage. There must be a level so high that, if you raised the minimum wage still higher, total employment would contract. Everyone recognizes that you can only go so far. But for now there's disagreement.

The advantage of work subsidies is that they would bid up the wages of low-wage people, and that same bidding for more low-wage people in the labor market would pull up their employment too. With the minimum wage, of course, the suspicion is that raising it will cut back on the number of low-wage workers that companies feel they can afford.

In preferring work subsidies over minimum wage increases, are you subscribing to the theory that the minimum wage creates a price floor on labor, lowers demand for it and thus creates a shortage of low-wage job opportunities?

I don't think it would lead to an outright shortage, but in most cases it would reduce hiring in the aggregate, and it would tend to shift hiring away from the lowest paid workers to the not-so-low-paid workers, because if you're going to have to pay more anyway, you will choose to hire the more promising candidates.

If you raised the minimum wage to $10.10 [which Sen. Tom Harkin (D-Iowa) and Rep. George Miller (D-Ca.) proposed and the White House supports], then some companies that might have otherwise gone on paying the minimum of $7.25 an hour would now say, "Well, if I have to pay $10.10, I'm going to hire high school graduates rather than dropouts."

The intuition of most economists is that both effects are going to be present. Companies are going to skimp a bit on labor in the aggregate when the minimum wage goes up, and they will tend to substitute more skilled workers for less skilled workers for the reasons I just outlined.

Do you think the increased labor cost to employers gets passed on to consumers in the form of higher prices?

We are talking about a smallish percentage of workers that'd be affected [EPI puts the number at 16.4 percent for Harkin-Miller -- Dylan]. So it is not nearly as consequential for business costs as some of the big hikes in the price of oil that we've seen over the decades. So I would suppose that costs would rise, sooner or later, depending on what else is happening, by one or two percent. But I don't think that is going to ignite expectations of inflation and require the Fed to go into a tight money posture.

Moving on to your proposal: a subsidy to employers who hire low-wage workers. A lot of people hear that and think, "Well, doesn't the Earned Income Tax Credit do that already?" Your proposal differs in that the money goes straight to the employer rather than the employee, but what is the advantage of it over the EITC? Or over the minimum wage, for that matter?

The main point is that raising the minimum wage seems to all economists to, at the very least, fail to raise employment, and we'd all like to see better inclusion of low-skilled workers into good, paying jobs. They need jobs that will support them financially, make them better experienced and make their lives a little more meaningful.

The low-wage employment subsidy idea is that the government would pay companies a subsidy for every low-wage person on their books. In the execution of that idea, the government must subsidize not only the employment of $7-an-hour workers, but also $8-an-hour and $9-an-hour workers, and so forth. You want to pull up the wage rates at the bottom, but you don't want the result that the people who were at the bottom are now earning more than people who were next to the bottom. You have to pull up the lowest wage more than the next lowest, and that less than the wage after that.

You ask about EITC. That has historically been heavily oriented toward mothers of dependent children, and although there have been some steps to broaden it, it remains a program that does not do much for those without dependent children, such as most single men. My thinking has always been that the worst problem we have with regard to lack of inclusion is the terribly low labor force participation rates and terribly high unemployment rates of young men, especially young men in ethnic minority groups, and in particular young black men.

You have little representation of young black men in the business sector, so you have children growing up in disadvantaged neighborhoods who don't hear discussions at the dinner table about what goes on at in business. It's almost as if we have two nations.

I've long felt that this is an untenable situation, and should be remedied in a reasonably cost-effective way, and I think that graduated low-wage subsidies are just that sort of system.

One common criticism of this kind of system is that it can create perverse incentives. People might be less likely to stay in high school if they don't need it to make a decent living, for instance, and businesses might be more reluctant to give folks in this position raises if raises brings with them a lower subsidy. How big a problem do you think that is?

The problem that some young people might not stay in high school to get a degree is a real one. Prior to the subsidies, getting a high school diploma meant the difference between, say, $10-an-hour and $7-an-hour, and afterwards it won't. So there's a loss of incentives for young people to stay in high school for the extra training, the enhanced literacy and so forth. If we institute a low-wage employment subsidy program, that's one thing we want to watch, its effects on the demand for years of education. But just because there is theoretically such an effect doesn't mean it's of a large enough magnitude to be a a crucial point against having such a program at all.

Perhaps the low-wage employment subsidy scheme would also reduce the incentive of a worker to move up the ladder, to get promoted to a higher position, and you don't want that. You want them to pull themselves up. But I'm not sure that the ambition to do better, to climb higher is just a matter of that wage differential. There could be all sorts of reasons to want to get promotions. It's natural that people compete to go up the ladder, to express their determination and exercise their ambition. There might be greater security up to a point, it's maybe more pleasant to be able to give orders than to be always on the receiving end of orders, it's nicer to have more responsibility, to have the scope to show what you can do, and so forth. There are a hundred factors behind the aspiration to do better.

So I could imagine someone looking at the job market today and thinking, "Even if we subsidize low-wage employers, it'll still be tough for low-wage workers and they may not have enough bargaining power for this to really help them." Do you agree with that, that this is mainly helpful during upturns but downturns are a different story?

I certainly do not think it will help only in upturns. I think it would help in downturns as well, when companies are thinking, "Oh my gosh, maybe we have to take an ax to parts of our workforce, because it looks like this slump is going to go on for a while." In that case, the fact that every low-wage worker is not just a producer, but also someone who brings subsidy money with him or her to the company, helps to encourage the employer to hang on to them, at least for a while, to see whether things stabilize and maybe start turning around.

To take an extreme case, suppose the subsidies were constructed so that companies don't have to pay for the lowest-wage people, and the government simply paid them, or paid the company to pay them. In that scenario, if that was sufficient for the companies to attract such workers, I would expect that the firms would downsize very little in the event of a downturn, because those lowest-wage workers weren't costing it anything but floor space. It does have a buffering effect.

You sometimes hear people arguing that we need both the minimum wage and wage subsidies, that they serve as complements. What do you make of that argument?

I disagree because under wage subsidies, we can expect that that wage will be bid up by companies competing more strenuously for each low wage worker because each low wage worker will bring a subsidy that goes to the company. The market competition passes along some of the subsidy to the workers. So if we think that $11 is the right minimum wage, maybe the wage subsidy program could easily deliver $11 as an effective minimum, and there'd be no need for a statutory minimum wage law. If there are companies already paying everybody $11, they wouldn't be affected by a minimum wage law all about $10. But if there is a high minimum wage, it will be very important to enact work subsidies of low-wage employment.

I think the argument is more that, if you implemented wage subsidies now, much of that would be captured by the employer rather than passed along in the form of a raise to workers. If I'm an employer paying $11 an hour now and the government steps in and says, "We'll pay $7 out of that if you'd like," I'll just pay out $4 and pocket the savings.

That could be the immediate impact. The immediate impact might well be that the company is delighted to find that wage rates are still low, and now it has the subsidies coming in. But someone in the company will say, "Why don't we hire some more of these low wage people? We passed them up before because we hesitated to pay them $7 an hour, but now we can pay them $9 while reducing our own cost to, say, $6 thanks to the $3 subsidy." Then a bidding war breaks out among employers, all of whom would like to step up somewhat their use of low-wage labor, because it's now a bargain, taking into account the subsidy. So that's got to buoy up both employment and wages.

Would the work subsidies make the difference between low employment and high?

Since "Rewarding Work," which I stand by, I've been struck by the utter principlelessness that we find in Washington, where any appeal to state aid is on an even basis with other appeals, and it's just a matter of what society feels like doing. Some people get their houses blown away, that's a shame, so let's give them a few tens of billions of dollars. It doesn't matter that they happen to be in the upper half of the population.

That's a corporatist idea, that everybody's entitled to social support. It wasn't that way in Mussolini's time, but there was an anti-market feeling, that everyone is entitled to be protected from the market. That has metastasized into what I call the "new corporatism," where it isn't just a few anointed social groups like factory labor and certain communities that are deserving, but everyone is deserving of support by the state. The message is, "Regrettably, we can't help everybody, but we'll get to you."

It creates this fantastic business in your town of lobbying and patronage. There are just so many claimants for so many things. Businesses now think they'll make more money if they get one or two government contracts, so it's better to orient themselves toward pursuing government contracts. That will earn more profit over the long haul, especially if you allow for risk. That will earn more profit than if we devote ourselves to some five year project that may or may not deliver an innovation.

So you get this orientation of businesses away from creation of things and visions and all that, and into lobbying, rent-seeking, etc. And once a large established corporation has the special interest money it wants, then it's that much harder for newcomers to break into the industry with a new innovation, and to the extent that new entrants have been neutralized, then any innovation becomes unnecessary, because they don't have to worry about innovations from outsiders. They've built a fortress around themselves, and so they don't need to innovate anymore.

I think, in a nutshell, that's a significant part of the story of what's happened in the American economy. I think American society gradually begins going to pot starting in the 1960s, continuing in the 1970s, and the best evidence of that is that there was a tremendous slowdown of total-factor productivity (TFP) growth, and we haven't pulled out of that yet. If we pulled out a decade later, it might just be one of those things that lasts a while and then we're back in business, but that hasn't happened. We've had steadily slow productivity growth, with the exception of the years of the Internet build-out, which petered out abruptly in 2004.