Bryan Caplan  

Why Can't Labor Be More Like Housing?

Only a Biologist Can Go to Uni... I'm Proud to Be a Bad Person...
During recessions, demand for both housing and labor plummets.  But the two markets respond in very different ways.

In the housing market, we usually see dramatic price falls.  Lots of properties sit on the market for months.  But almost any property owner can swiftly sell by cutting his price.  In economic jargon, then, housing surpluses are largely voluntary.*  The lingering problem is just that some sellers stubbornly cling to pre-recession pricing.

In the labor market, in contrast, wages almost never fall.  Vast numbers of workers lose their jobs when demand goes down - and employers almost never offer the option to stay on at a reduced wage.  So in economic jargon and ordinary language, their unemployment is largely involuntary.  In every recession, lots of workers who are willing and able to work at the market wage genuinely can't find jobs.

Which market works better (or "less badly") during recessions?  "Housing" is the obvious answer.  Sudden price falls provoke angry complaints, but at least property owners can still sell by cutting their asking price.  And despite some inefficiencies, buyers gain by roughly the same amount that sellers lose.  In the labor market, in contrast, unlucky workers' labor income utterly vanishes - and lots of perfectly good labor goes to waste.

Now you could object that the so-called "obvious answer" overlooks the indirect effect on Aggregate Demand.  (See here, here, and here for earlier discussions).  When housing prices plummet, so does property owners' net worth.  Owners respond by cutting spending, which makes the Aggregate Demand shortfall even worse.  In the labor market, in contrast, wages stay steady, mitigating the Aggregate Demand shortfall.  Right?

Hold your horses.  This complaint doesn't even follow on its own terms.  If housing prices were as rigid as wages, property owners' net worth would indeed be higher on paper.  But their effective net worth would still sharply fall.  With flexible prices, they can sell their house for sure at, say, 30% less than they paid.  With rigid prices, they have a chance to sell their house at face value.  In which scenario are owners effectively poorer and stingier overall?  It's a tough question.

The same goes for labor markets.  With rigid wages, the value of workers' human capital is higher on paper.  But that doesn't mean that rigid wages actually make workers in general feel richer.  Instead, it means that employed workers feel somewhat richer, while unemployed workers feel much poorer.  In which scenario are workers effectively poorer and stingier overall?  Again, it's a tough question.

If housing prices were as rigid as wages, every recession would feature horrific involuntary vacancy rates.  By-the-book economists would say, "Housing prices need to fall."  Hard-line Keynesians would object, "That would just make the vacancy rate even worse by reducing Aggregate Demand."  Intellectual bystanders would shrug, "Well, economists disagree.  Who knows?" 

Fortunately, housing prices are flexible, so this dispute almost never arises.  But we can learn a valuable lesson from the absence of this dispute: Flexibility really is a good thing.  The direct effects of flexibility are good, and there's no reason to think that the indirect effects of flexibility are bad.  If there's any way to make labor markets work more like housing markets during recessions, we should try it.

* The main counter-examples are properties where the legal status is in dispute - most obviously, when a bank forecloses but the occupants refuse to leave.

Comments and Sharing

COMMENTS (27 to date)

I would say labor *is* like housing, at least more than you suggest.

Vast numbers of workers lose their jobs when demand goes down - and employers almost never offer the option to stay on at a reduced wage. So in economic jargon and ordinary language, their unemployment is largely involuntary.

This seems to imply that each person can only do one job, the job that their former employer paid them to do (or one very much like it). At a very similar wage. And there are literally no other unfilled jobs out in the economy. If none of that is the case (and it basically never is), I'm not sure what's so 'involuntary' about said unemployment in ordinary language (economic jargon notwithstanding, of course).

To your larger point, I'm not sure house prices are all that less sticky than labor prices. Your answer to my preceding paragraph will be that people have a reserve wage, have sticky fixed costs, and other factors/properties, etc. which essentially rule out going to work at McDonald's as a viable option for them. Fine, well another way to say that is that when they observe the market-price of their labor (the type of labor they think of as 'what they do' anyway) drop below that reserve wage, they 'take themselves off the market'.

Which, of course, is the same thing that homeowners do, in response to a slump in house prices. This need not even manifest as a house 'sitting on the market for months' - it might just be - and indeed very often is - a decision not to move or sell in the first place; 'now is not the right time'. Similarly, just as laid-off workers have stickiness re: the type of job they'll accept, preventing them from accepting a switch to McDonald's (and thus keeping them off the market), homeowners have stickiness re: the type of home (with an analogous). They don't think 'yes it's bad that I'll only get 60% of the peak price for selling now, but at least with that money I could go buy a tiny house in deep Las Vegas suburbia'. They think: 'gee, if I'd only get 60% of peak price, I shouldn't sell, I'll just stay here'. And again: understandably so.

It's true that in any recession there will be *some* forced/distressed home sellers willing to take big price cuts. But the same is true of labor! If you are implying that no significant # of people took lower-pay jobs, indeed different-career jobs, after being laid off during the recession, I think that's just plain wrong. So I'm not sure that the two markets are as different as you say in the first place. To the extent that they are, and that labor is the one that is empirically more sticky, the solutions that present themselves are things that incentivize that stickiness - such as cancelling unemployment insurance altogether, as well as perhaps many other social welfare programs. Is that what you propose?

MatthewH writes:

Query: if wages equal a worker's marginal product, does this imply that during a recession a worker's productivity decline?

Brian writes:

Wages are sticky in large part because that's in the employer's best interests. Why? Well, for a number of reasons.

First, wages are only a part of compensation, and keeping the same number of workers at a lower wage provides no relief on the benefit end. Fewer cuts overall have to be made by laying people off since a double benefit of reduced wages and benefits is obtained.

Second, wage cuts will cause some employees to leave, but the leavers are more likely to be the better employees, who are more confident in getting a another job and less likely to think they deserve a pay cut. The weaker employees will be less secure and hang on to the job no matter what. Ultimately, this means that pay cuts lead to uncontrolled loss of employees with a bias toward brain drain. This is not what an employer wants.

Third, by keeping wages the same but cutting the workforce, employers control who leaves (they decide who to lay off) and can ensure that they hold on to the most productive, most valued employees.

I don't see layoffs as the great evil you claim them to be. Yes, being laid off can be devastating to the person involved, but how else can workers be reallocated to more productive work?

BC writes:

MatthewH, I believe that the answer to your question is "yes". Although the worker may produce the same number of widgets per hour, the value of those widgets decreases during a recession since the demand curve shifts to the left. Since equilibrium price drops, marginal product in *dollars* per hour also drops. Hence, the need for a nominal wage cut. At least, that would be my Econ 101 explanation.

Chris H writes:

@Crimson Reach and Brian

I think a point you two might be missing is that you seem to be assuming that the need to cut wages is only happening in particular firms or perhaps particular sectors of the economy. In a recession however, almost all sectors simultaneously find themselves paying employees more than their revenues permit. As a result the laid off employees cannot easily find new work.

Think about it like this, a factory closing down in the US and going to China, thereby laying off a bunch of US workers, isn't really a big economic problem. That's just creative destruction in action and the economy as a whole is big enough and dynamic enough to integrate those workers back in. The only cost is the fairly short term trauma of losing one's job and the need for a job search. In a major recession where demand has dropped nearly everyone is cutting back and there is no where for the workers to be hired at. This is why you'll hear stories of hundreds of people applying for a small handful of not particularly high paying positions.

Now, there are structural things the government does that makes the situation worse, such as minimum wage laws and unemployment benefit extensions, but most of these can also be helpfully solved by looser monetary policy (unless they've decided to automatically index the minimum wage and unemployment insurance benefits to inflation).

And for your particular first point Brian, while that might contribute to sticky wage issues, I doubt it's a primary factor or else extended unemployment in late 19th century business cycles (before many aspects of modern benefits schemes were in place) and very low skill jobs with no significant benefits structure wouldn't see increased unemployment in recessions. Though I think your points 2 and 3 were an interesting and useful part of the explanations for sticky wages.

BC writes:

Bryan, I agree with trying to increase labor flexibility through deregulation. However, I think you underestimate the ease with which we could reduce nominal wage stickiness. I believe in your previous post, you even say that we just need to enact a "massive cultural shift".

That sounds as easy as changing human nature itself. You say that, "Employers almost never offer the option to stay on at a reduced wage." However, I don't believe that employers are the ones that resist wage flexibility; it's workers. Most people would be much more disgruntled by a 5% nominal wage cut with 0% inflation than a 0% nominal wage change and 5% inflation, even though both represent a 5% real wage cut. That is the source of nominal wage stickiness, i.e., it's a behavioral phenomenon. (As far as I know, real wage stickiness is much less of an issue.) I would say that people are even less aware of market-clearing wages. Consider two cases:

(A) Due to a boom, average wages for people in your profession increase by 5%, but your wages stay constant.

(B) Due to a recession, average wages for people in your profession decrease by 10%, and your wage is cut by 5%.

It would not surprise me if more people would be more upset in (B) than (A), even though in (A) they are underpaid by 5% and in (B) they are overpaid. People may not even know what is happening with average wages across an industry, but they see their nominal wage on every paycheck. Hence, they are more disturbed by a nominal wage cut than by not receiving a market or inflation justified nominal wage increase.

Regarding your previous post, laying off an employee can potentially create a sorrowful (or angry) ex-employee. Cutting an employee's nominal wages can potentially create an angry current employee. Angry current employees can disrupt morale, cause mischief, and do other negative-value things much more easily than angry ex-employees. So, I don't think employers have a preference for laying people off as much as they fear cutting nominal wages.

While we can try to educate everyone to focus on real wages and market-clearing wages instead of the nominal wage on their paycheck, changing human cognitive biases is no easy task. On the other hand, stoking inflation through monetary expansion enables real wage cuts even when nominal wages are sticky. Admittedly, the process can take time. Monetary stimulus, however, does not require changing human cognition. (There is also good reason to believe that targeting stable NGDP growth would reduce recessions and associated unemployment, and Scott Sumner did not pay me to say that.)

So, if one believes that nominal wage stickiness is a result of human cognition, then it would seem that there are other ways to combat its effects besides trying to effectuate a "massive cultural shift".

Milton Recht writes:

If one distinguishes between owner-occupied housing and investor-owned housing then I think the difference between home prices and wages during recessions is explainable.

For an owner-occupied home, the buyer will have imputed rental income and buyers have knowledge of a comparable home's imputed rent. At some lowered selling price, there is value to the buyer in buying and occupying the home and in getting the future imputed rent. For investor-owned rental properties, there is also a lower price where there is value for future rental income, but vacancy and rent non-payment risks also must be considered.

When an employee lowers his/her wage (selling) price, the employer (buyer) knows the comparable wage price of the general class of employees similar to the specific employee, but does not have enough information to know the actual wage price value of the specific employee. There is more uncertainty about a new hire's future performance value than there is about the future (imputed) rental value of a home.

If the variance in performance/productivity of a class (same education, experience, etc.) of employees is larger than the rental variance of a class (same size, type of neighborhood, etc.) of homes, then lowering a wage demand will have a less positive sales effect than lowering a home selling price.

If an employer by lowering wages increases the potential for greater employee turnover, the future rehiring risk (variance in employee performance) to the employer due to the productivity/performance uncertainty of new hires will increase the cost of new employees and inhibit the employer from lowering wages, even if the employer can find future new employees who would accept the job at a lower wage.

Martin writes:


"Query: if wages equal a worker's marginal product, does this imply that during a recession a worker's productivity decline?"

If less output can be sold and less output is expected to be sold, then a number of workers have been engaged in producing something which has had apparently no value to others and is not expected to have any value to others in the future.

Whether you believe that workers' productivity has declined largely depends on what you expect to happen to output in the future.

Martin writes:


"In which scenario are owners effectively poorer and stingier overall? It's a tough question."

It is a tough question.

If everybody believes that housing prices will be lower/higher, then they will be lower/higher. And there is really not any way to know whether those expectations were correct until producers have made their output decisions, where the output decisions probably take into account the expectations about wealth and thus also housing prices.

I don't really see any direct way out of this, and whether rigid or flexible prices are "better" seems to me to depend on whether you believe expectations are correct often enough to overcome any errors.

RPLong writes:

One partial explanation is that home prices eventually recover after the recession, while wages almost never do. I know that if I agreed to take a $10,000 pay cut in the middle of a recession, it would take me several years to work myself back into a position that paid the same as my old job.

Now imagine you're an employer considering whether to hire someone at a job that pays (say) $100,000. Do you hire the person who was recently making $100,000, but got laid off 6 months ago, or do you hire the person who went from making $100K to making $80K over the last six months?

It almost looks like the latter person was demoted.

Danyzn writes:

Housing is a durable good. The exact labor analogue would be a market in slaves or indentured labor, which is not what you're talking about. The labor market you're talking about is a rental market, so the correct analogy is with rental real estate. Moreover, labor is a factor of production and not a consumption good, so the most exact analogy is with the commercial real estate rental market. And there you do indeed see vacancy rates go up during slumps.

R Richard Schweitzer writes:

Really now?

Housing is a nonproductive set of assets. It can be seen as a form of consumption; most clearly when the use of housing is represented by rents.

Labor is productive; to the extent that it is actual labor and not just activity that precedes consumption.

Even taking the available pool of work capacity as a "commodity" does not make it comparable to housing. The functions of housing and labor in an economy are distinct.

Zachary writes:

Historically, when we had -0.5% inflation and a stable gdp, workers were not only willing to take wage cuts in recessions, in normal times they were willing to take very modest or even no raises at all! Prices fell instead. I see the artifact of persistent rising nominal wages as a mere expectation norm caused by our designed monetary policy. There is no good reason historically or theoretically that price rigidities should be innate to labor markets.

Eric Evans writes:
Now imagine you're an employer considering whether to hire someone at a job that pays (say) $100,000. Do you hire the person who was recently making $100,000, but got laid off 6 months ago, or do you hire the person who went from making $100K to making $80K over the last six months?

In an industry where pay cuts are common (relatively speaking), that consideration means nothing.

RPLong writes:

Eric, can you give an example of an industry in which pay cuts are common? I thought the whole point of Caplan's post was to discuss downwardly sticky wages...

Chris H writes:


I agree in most industries most of the time pay cuts aren't common, but that concern you cite doesn't seem likely to be in effect during a major recession. Consider that in a recession willingly taking a pay cut might actually signal being dedicated or a team player especially given that it signals your previous boss trusted you enough to take a pay cut without ruining morale (or at least not having readily available alternatives, which doesn't speak poorly for someone in a recession when hiring is low). In normal economic times, yes I think an employer would likely view a pay cut as a likely demotion. But in a major recession (which pretty much everyone knows when that happens because that's all anyone will talk about), I think that concern isn't as prominent. Thus the morale concerns of pay cuts seem more likely to be what employers are fearing.

Thomas DeMeo writes:

Buying/selling a house is more like an initial hiring than an ongoing wage relationship. The price floats but you eventually commit to a set price.

Adjustable mortgages are somewhat like flexible wages. It seems like some countries make this work very well, but under strict controls. But we also saw some destabilizing abuse of the concept among some banks in the US. There might be an analogy in that too.

Greg Ransom writes:

Every employee in the construction company my neighbor worked for in SoCal got a 50% pay cut.

Wages don't adjust in the sectors controlled by government and the unions,

Greg Ransom writes:

The income of many self-employed and small businessmen dropped dramatically in SoCal.

If you define wages as an economic textbook does, there are many sectors with very flexible wages.

Greg Ransom writes:

When employers cut full time staff and increase part time staff with lower benefits, that's a wage cut.

Read the paper. It's happened everywhere.

Hana writes:


Any industry with a high component of labor with commissioned sales people, secondarily companies where total compensation has a high bonus/incentive percent.

So group one would include Real Estate, Automobiles, Furniture and Appliance, Advertising, and so forth.

Group two is more company dependent. One of the software companies I worked with provided compensation in salary, stock options, and bonuses. Salary rose significantly only with promotion, stock options were a function of management level, bonuses were dictated by results of the company. In normal (good) years bonuses were approximately 15% of total compensation for the year. In poor years bonuses were less than 5% of total compensation, and in one case were 0%. In very profitable years bonuses came to 30% of annual compensation. Wage flexibility was therefore built in to the total compensation, but not reflected in the salary itself.

The concept of lowering wages to retain more employees is misguided.
1) Employers, in the absence of labor agreements protecting seniority, will eliminate the least productive workers first.
2) Managers are able to release bad hires without having to address their poor hiring decisions.
3) Companies can revamp their strategic direction by eliminating employees and divisions that do not make long term sense for the direction of the companies.
4) High calibre employees will be recruited away. Companies that lower wages will experience the flight of their best employees further limiting their ability to survive recession and thrive post recession.
5) Employee morale is a function of the sense of pride in the work produced. If there is a slowdown, and fewer employees are needed, then reducing wages to keep more employees around is counter productive. The excess staff will reduce the overall morale, not increase it. This takes place on top of any anger that is felt by salary reductions. Having dead weight in an organization reduces the purposefulness of all employees efforts. As morale declines the most motivated employees will tend to leave earliest. Hardly the plan that leads to future prosperity.
6) Not being eliminated can be detrimental to the employee. Many of the hardest hit industries in a recession emerge from recession transfigured. (Steel, Automobile, Telecommunications, Printing, Newspaper, Textile, and the list goes on). By not reducing employment earlier, many workers have a more difficult time transitioning to a new type of employment. Assuming a 40 year work span, a three year recession, with a reduced salary, and in a dying/negatively transitional industry, represents a large hit on an individual's work career. Transitioning earlier through layoffs supplies better personal payout than waiting through low wage transitions.

On a personal level I have worked at a company that implemented wage reductions, rather than layoffs. It was implemented through a process called 'Negative Reviews'. The company used a standard 1-5 review process with 5 being the highest. The overall review was approved by your manager, your manager's manager and HR. Company philosophy dictated that no one was perfect therefore there could be no level 5 reviews. Ratings between 4-5 received no salary reduction. Ratings in the other ranges received between 10% and 30% reductions. There was a high management bias towards at least a 10% reduction for all staff. Some employees avoided the reductions through end or year promotions. Needless to say employee morale was crushed. No one felt better because there was no layoffs.

MingoV writes:

There are companies that avoid the fixed wage trap. Employee-owned companies sometimes link wages to revenues. Other companies could use employee contracts that link wages to revenues. These systems provide advantages other than avoiding bankruptcy during recessions. When wages are tied to revenues, there is a huge incentive to cooperate, increase productivity, propose new ideas, and get rid of mediocre employees.

Gordon writes:

If the majority of homeowners took on financial obligations based upon the current value of their homes, wouldn't the likely result be a great deal of stickiness in home prices when recessions hit? Comparing the flexibility of home prices and wages is comparing apples and oranges because so many wage earners take on financial liabilities based upon their current wages.

Hazel Meade writes:

I think this has come up in other contexts as well, but what I think the economics profession could really use is some accounting for the risk of becoming unemployed or experiencing a future economic downturn.

There ought to be some statistics on this that could tell a given worker in a field what his future probability of getting laid off it, given the statistical averages in that industry.

What's especially troubling is that most of the financial advice and the calculations of supposed risk aversion by ordinary people takes no account of the probability of economic downturns in the future. (I.e. the computations of how much you can "afford" to spend on rent per month.) What might seem like risk aversion by people stuffing excessive amounts of savings under the mattress, might seem like a rational response to the not unlikely probability the that economy will tank and the individual will be out of work, if given a different context.

My guess is the reason wages are sticky is largely because cannot easily downgrade their lifestyles to a reduced wage. Your mortgage payments and car payments cannot adjust downwards, and neither can utilities or food, which leaves you very little room for adjustment if you're living up to the maximum allotment that standard financial advice allows.

Peter H writes:

" Lots of properties sit on the market for months. But almost any property owner can swiftly sell by cutting his price."

This just isn't true at all.

When a house drops in value below the mortgage, the seller needs to get the consent of the mortgage company to sell in a short sale. If they can't get that consent, they can't sell below a price that satisfies the lien.

So for example if someone is upside down by $100,000 and has a steady job that allows them to make the payments, but lacks $100k in liquid assets, they will not be able to sell their home at all.

Floccina writes:

The signal to the employer is lower sales so he thinks he needs fewer employees to produce less product. The home seller thinks differently. It is hard for the employer to distinguish between lower demand due to a general decline in sales (ngdp)and lower demand due changes in preferences and even if he does and lowers his prices if other producers do not also lower their prices people still may not have enough money to buy as much of his products.

Joel Adler writes:

When a period of reduced demand for labor arises,employers are not likely to replace retired workers but ask a now smaller department to produce the same volume of output. While employers can't terminate current employees and replace them with lower cost new hires without exposing their company to a law suit,they can offer smaller bonuses and reduced benefits to the retained workers. If the need arises to replace workers due to natural attrition or retirements, the trend will be to offer a lower starting salary in the face of a weak job market.

Comments for this entry have been closed
Return to top