« The Economic Journal of Negative Research Findings | Main | Robust and fragile models, and fragility in the limit »

Comments

Feed You can follow this conversation by subscribing to the comment feed for this post.

"I can't think of any plausible reason why it wouldn't work."

O.K., so if you're cutting wages by 1% and increasing employment by 1%, why not also cut weekly hours by 1%? While you're at it, why not cut wages and hours per worker by 20% and hire 20% more workers, giving existing workers a four-day workweek?

Keynesian, you say? "The full employment policy by means of investment," he wrote, "is only one particular application of an intellectual theorem. You can produce the result just as well by consuming more or working less..." Keynes in a letter to T.S. Eliot, April 5, 1945.

See also Robert Skidelsky: http://www.project-syndicate.org/commentary/skidelsky41/English

Sandwichman: because my policy increases output, because there's 1% more person-hours of employment.

Make that: because my policy increases output, because there's 1% more person-hours of *government* employment.

Makes sense, but here's a try at some plausible disagreement: you might be a New Keynesian that blames price stickiness rather than wage stickiness, and that involuntary unemployment doesn't really exist; just people who are (unsurprisingly) vocal when their expected real lifetime incomes drop.

In this case, since real wages are not too high, you will be unable to hire 1% more labor at the same wage, never mind 99% of that wage. Of course, in a heterogeneous labour market one will always be flooded by lower-productivity labour seeking a higher wage, so there will always be applicants, just not of the same quality as the ones you have.

david: I think that's a valid counterargument, given your assumptions. If prices were sticky, but wages perfectly flexible, then a fall in AD would cause output and employment to drop, but there would be no involuntary unemployment in the sense of an excess supply of labour at the going wage.

But I don't believe those assumptions, and I don't think any Keynesian does either, even though some New Keynesian models make that assumption for simplicity.

Well, here's another, since I expect someone else will complain if I don't: 0.99 × 1.01 is not exactly 1, so the government payroll will not remain exactly the same... ;)

(but this is a field which should really have a Fisher approximation rather than a Fisher equation, and nobody cares, so whatever)

Besides this. Alright, say that you're a New Keynesian who believes in the Lucas critique and has rationally microfounded agents. But also believes in wage rigidity, both real and nominal. This is explained by blaming assorted labour market artefacts, like morale, Akerlovian gift exchange, etc. that mess around with nominal wage setting.

For a government employer, all these reasons still apply (i.e., the private-sector labour market, being rational, is not adjusting because it has good self-interested reasons not to adjust). So there will be a loss that applies to the government. So this is not a free lunch.

"because my policy increases output, because there's 1% more person-hours of *government* employment..."

Ummm, Nick... what makes you so sure that productivity wouldn't also be increased by a reduction of hours per person while holding total person-hours constant?

"The days are gone when it was necessary to combat the naïve assumption that the connection between hours and output is one of direct variation, that it is necessarily true that a lengthening of the working day increases output and a curtailment diminishes it." – Lionel Robbins, 1929

Nick - didn't Bob Rae do something more or less along these lines when he was NDP premier of Ontario in the early 1990s? Cut everyone's salary by requiring people to take rather than cutting jobs?

It's sort of reverse interest group politics - annoy loads of people a little bit rather than annoy a few people a lot. Not a good way of getting yourself re-elected...

david: I couldn't find my calculator! (And I would only have confused myself, and some others, if I had tried to do it right).

I like your new argument better than the other one. It's a very different argument than the one that is normally used by Keynesians to argue that wage cuts in a recession won't work. There, wages are treated as a policy variable, and the argument is a demand-side argument. Yours is a supply-side productivity argument. But it's internally consistent.

Sandwichman: I am putting forward an argument that, from Keynesian grounds, a 1% cut in wages, and a 1% increase in employment, holding hours constant, would be a good thing.

You aren't addressing my argument. You are arguing that work-sharing would be a good thing, because it would increase productivity.

"You are wrong! It would be much better to give every worker free gym membership, because that would make them healthy and happy, and that would increase productivity!"

(I'm not serious, but you are off-topic.)

Well, the level of real wages could still be the policy-relevant variable. As per standard New Keynesian recommendations, one could argue that the govt. should reduce real wages by 1% and increase hiring by 1%, albeit through inflation rather than cutting nominal wages. The private sector never has the option to engineer inflation, of course.

Frances: IIRC, Bob Rae didn't increase government employment *compared to what it was previously* when he (sort of) cut wages. But you could well argue that he increased government employment *compared to what he would have done otherwise*. I.e. he kept it constant but would have cut employment if he couldn't have (sort of) cut wages. So, maybe it's an example of what I'm talking about.

(Oh, and nobody get into Bob Rae bashing either. I have taken Frances' comment in the spirit of "is this an example of the policy you are talking about?".)

Rae Days, I remember them well. It applied to schools too. All the PA Days for that school year were made unpaid, but there was still a half-week left, so we got an extra half-week for March Break, Wednesday off included. Victoria Day was a four day weekend, AIRC. In my school board, for the next decade after that we always got the Friday before the official March Break off in the schedule as a PA Day. I believe parents appreciated being able to get away early if they were travelling.

Sadly the practice has since stopped.

David has got my critique down. Why exactly is cutting the wages of 100 people in order to hire one extra person a good thing? Furthermore, as someone who has been through the Public Service of Canada's hiring process up to the interview stage, I cannot imagine using that system as an employer of last resort.

I love the Public Service because it's very transparent and logical in its decisions, actually gives you rejection notices and once you figure out the logic it gets much easier to push an application. But don't ever say it's quick.

Actually after all that persistence it was delightful to be summoned for an interview in downtown Ottawa four blocks from the Parliament Buildings and realizing how few get that far.

To be fair most private sector employers who don't make you man a cash register are worse at getting back to you and outright slower by integer multiples than the federal Public Service is. My standard of a good employer is one who has the decency to give you a form rejection e-mail to say you didn't get the job. I understand if you don't want to have an awkward conversation but at least have the decency to inform me it's over.

"You aren't addressing my argument." "I'm not serious." I'll resist the strong temptation to connect the two statements with ellipses.

But you can't be serious! Ever heard of Alfred Marshall? A.C. Pigou? Lionel Robbins? John R. Hicks? John Maynard Keynes? John Maurice Clark? Sydney J. Chapman? Maurice Dobb? Paul Douglas? Dorothy W. Douglas? (Not to mention Karl Marx!) All of the above explicitly treated wages, HOURS and output and inextricably interrelated. If I'm "off topic" then you're not talking about economics.

That should read AS inextricably interrelated.

Determinant: "David has got my critique down. Why exactly is cutting the wages of 100 people in order to hire one extra person a good thing?"

Think macro. Think output. Think GDP. Think useful goods and services. You are thinking of wages as if they were transfer payments. "Why is cutting the dole for 100 people so you can afford to pay the dole to one extra person?" the only good there is the minor benefit from redistributing income. I'm talking about *employment*. Employing one more person is a good thing because/if it will increase *output*. David's answer is a good one precisely because it questions my assumption that output will in fact increase.

"David's answer is a good one precisely because it questions my assumption that output will in fact increase."

But Sandwichman's is "off topic" precisely because it questions Nick's assumption that output will in fact increase!

Sandwichman: This post is not about work-sharing. I am holding hours per worker constant. You are off-topic. Do not hijack this post. Stop now. Not everybody shares your interests.

In the aggregate, is there really a fundamental difference between supply-side productivity and demand-side arguments? For the disequilibrium Keynesian, the question really is "what is the least costly way of correcting the disequilibrium" and systematically altering the real value of all nominal assets is just one of many options. Direct nominal price adjustment is also one of those options, but it may not be the ideal one. Hammering aggregate intertemporal preferences aka deficit spending is another option. In some hypothetical universe, perhaps distributing buttons a la Whip Inflation Now might work...

The concept of downward nominal wage rigidity is tantamount to asserting that workers don't noisily object when their real wages creep downwards, whereas they do if their nominal wages decline. Aren't both arguments about productivity under given circumstances?

This proposal works only to the degree that you believe there is an output gap, or idle resources, in which case you should go ahead and hire the unemployed at full salary.

On the other hand, if you believe that hiring the unemployed at full salary would crowd out private spending, then you don't believe there is an output gap and hiring more labor would also crowd out other output.

It is a bit like policy by committee, where you take a little bit of two mutually inconsistent policies and try to get a consensus policy as a result.

It's a good thing I think of wages as transfer payments then.

Fundamentally I don't think this policy will work because it doesn't solve a balance-sheet recession. Keynes described a balance-sheet recession in the General Theory somewhat clumsily. He didn't make the ultimate leap that a balance-sheet recession, a demand depression, is caused by a shortage of money. In order to really, truly solve such a problem we need to expand the money supply. It's the fact that the money supply is finite that gets us into recessions in the first place.

In logic models it's a memory system with a negative edge-sensitive state change. To get out of recession you need a positive shock; an expansion of the money supply. Once you expand the money supply then you need to spend it.

Your solution certainly takes care of the spending bit but it doesn't expand the money supply. Without expanding the money supply you won't get the positive shock. You've got the transmission mechanism but nothing to transmit.

In a balance sheet recession economic actors believe it is better to pay down debt than invest in additional productive capacity. You need to shock them into believing otherwise.

david: I have some sympathy with that perspective.

rsj: "This proposal works only to the degree that you believe there is an output gap, or idle resources, in which case you should go ahead and hire the unemployed at full salary."

That would presumably be the Keynesian's first-best policy. But suppose you couldn't convince the government that deficits didn't matter? Or, suppose you were advising a provincial government, that couldn't print money, and so faced a hard budget constraint? (Or, a Eurozone country, for that matter.)

Determinant: "It's a good thing I think of wages as transfer payments then."

? If an increase in transfer payments didn't increase output, GDP, and national income, Keynesians would not want to increase transfer payments (except it might redistribute the same amount of income). They want a policy that increases output. My policy (david's argument aside) would increase output.

"He didn't make the ultimate leap that a balance-sheet recession, a demand depression, is caused by a shortage of money. In order to really, truly solve such a problem we need to expand the money supply. It's the fact that the money supply is finite that gets us into recessions in the first place."

Now you are sounding like a monetarist :-). A true Keynesian wouldn't say that. In any case, why wouldn't my policy work, even if the money supply stays fixed? I just showed you. Government output expands; the private sector should be unaffected.

"That would presumably be the Keynesian's first-best policy. But suppose you couldn't convince the government that deficits didn't matter? Or, suppose you were advising a provincial government, that couldn't print money, and so faced a hard budget constraint? (Or, a Eurozone country, for that matter.)"

Well, the first best policy would be to exit such an arrangement :) But barring that, if I was in a government that didn't have control over interest rates, but was able to levy taxes, then I would modify your proposal and raise taxes, using the funds to hire additional labor.

It is just like your proposal, and would work for the same reason, but you have much more control -- e.g. raising taxes on the wealthiest to employ the unemployed -- rather than just singling out government workers to bear the brunt of the transfer.

Particularly if the wealthiest members of your society are not government workers. That would be the logical conclusion, right?

Anyway, Nick, William H. Hutt had the same idea in 1945: "Full Employment and the Future of Industry." (And it's not about "work-sharing") But he didn't think it was Keynesian. He was an outspoken critic of Keynes.

I mean the same idea as yours: more hours = more output = larger GDP equals more employment all around.

rsj: are you thinking of the balanced budget multiplier? Or the idea that the wealthy have a lower marginal propensity to consume? (Maybe both).

Yes -- I think my last comment didn't go through. Take a look here:

//econ.jhu.edu/people/ccarroll/BSLCPIH.zip

Extract:

"This paper will argue, however, that such responses, and a wide range of other evidence, are consistent with a version of the LC/PIH model in which consumers face important income uncertainty, but are also both “prudent,” in Miles Kimball’s [1990b] sense that they have a precautionary saving motive, and “impatient” in the sense that if future income were known with certainty they would choose to consume more than their current income. Under these conditions, consumers may engage in what I call “buffer-stock” saving behavior. Buffer-stock savers have a target wealth-to-permanent-income ratio such that, if wealth is below the target, the precautionary saving motive will dominate impatience and the consumer will save, while if wealth is above the target, impatience will dominate prudence and the consumer will dissave."

"...The model’s most surprising feature is its implication that, even with a fixed aggregate interest rate, if consumers are sufficiently “impatient,” average consumption growth will equal average labor income growth, either for individual households or for aggregate consumption. This is true even though the consumers in the model behave according to the standard Euler equation which has been widely thought to imply that consumption growth depends only on tastes, and not on the growth rate of income. The problem in previous work has been in the common assumption that the second-order variance term in the log-linearized version of the Euler equation can safely be ignored. In fact, this variance term is an endogenous equilibrating variable: it will, on average, take on whatever value is required to cause average consumption growth to equal average income growth."

rsj: actually, I think you are right. Conceptually, the reason why it works is (I think) identical to the Keynesian rationale for the balanced budget multiplier. It is exactly as if you had placed a special 1% tax on government workers, and used the proceeds to increase government spending. According to the simplest Keynesian model that should have a multiplier of 1. Which is what my argument says.

But, on third thoughts, there is one difference between my policy and the standard balanced budget multiplier. In the ISLM model, the balanced budget multiplier is less than one. That's because the increased income increases the demand for money, so if the LM curve slopes up, the multiplier is less than one. That's because the (stock) demand for money is assumed to depend on *full* income (C+I+G). You could argue that the demand for money instead depends on C+I+T, because you don't need to hold money to pay for government services (though the government might). Let me simplify: my policy would presumably have no effect on the demand for money. Existing government workers demand less, and the newly hired demand more, so it's roughly a wash.

Well I'd argue it depends what these workers are doing. In your 1%/1% scenario, you effectively get more productivity for $ spent, and the wage terms for existing employees is weak anyways with elevated employment so why not. Part of the key with emerging from recessions is providing _real_ economic growth and governments are right to continually review whether work is productive, as is true regardless of time period. According to a Keynesian, they should always be ready with an ordered list of productive-ish things on which to spend money when labour capital become available, like bombing the heck out of Nazis or whatever.

Energy efficiency seems like one that should be considered, in the absence of armed Germans.

Also it's a bit of a false question. Wage terms are weak for the employed due to high unemployment. Does that mean we should slash their wages? Actually no, according to Keynes, governments just borrow more, no? But with one caveat: if there is waste in the system it will be unproductive and ineffective regardless any output gap, which is the Medicare/Medicaid problem the US is currently facing. JMHO as a non-economist looking through the looking glass.

To add: one way I had always interpreted Keynes is that governments should try to keep wages of employed workers flat and bring unemployed workers up to where their employed equivalents are, through government-sponsored programs. Basically, fill in the output gap hole with borrowed money. But lest we forget that there was likely a lot of built-up waste in both the private and public sectors before a recession, so slashing spending in certain oversubscribed areas is warranted if it wasn't done already, so long as it's made up with other more noble pursuits.

Interest post thanks.

Isn't this exactly what the UK is attempting to do?

Public sector wages are frozen for two years, CPI inflation is at 4.2%, public service pensions contributions are being signficantly increased. This adds up to a significant real terms wages cut, (around 10%).

Whilst total public sector employment has been cut as well, the reduction must be smaller than had real public sector wages increased in line with inflation.

Not sure whether this would be expansionary compared to unobserved counter factual where real wages went up but total employment went down.

I agree that this would probably work to increase output, but I don't think it's Keynesian.

It's the government using its market power to buy more stuff for the same price (reminds me of when some big retailers in the UK simply told all their suppliers that they expected an immediate 2% price cut or else they'd stop buying. Most caved in)

Because the government is a big purchaser, this might have some macroeconomic effects - wage deflation, which may or may not be positive depending on your models, and increased real output.

But I don't think it's Keynesian in any way - the Keynesian model depends on government borrowing to cancel out private attempts to save, and you have excluded that by assumption. It's just ordinary government intervention.

(yes, there's a correlation between supporting government intervention and believing in Keynesian models. But that doesn't make this Keynesian)

I do think this plan would have a monetary effect - it would increase velocity, to the extent that the newly-employed have a higher marginal propensity to consume than the already-employed. Does that count as monetary?

Nick, FWIW, this policy sounds completely sensible to me just from an ECO 1000 macro perspective. People save a higher % of their income as their income increases. So if you have 100% of the people putting in a 90% work week (and getting paid 90% of their salary), you'll have less saving and more spending than if you have 90% of the people putting in a 100% work week (and getting paid 100% of their salary). This generates more demand for stuff and the economy expands.

Now I think you're suggesting that people put in a 100% work week for 90% of their salary - but isn't that a stronger assumption than you need for your results to go through?

"It's sort of reverse interest group politics - annoy loads of people a little bit rather than annoy a few people a lot. Not a good way of getting yourself re-elected..."

Yeah, but Rae's mistake was annoying the "loads of people" who had voted for him. I suspect a government of a less orange hue wouldn't have that problem (at least not to the same degree).

It's interesting that explicit public sector wage cuts seldom come up in austerity discussions (at least not until all hell breaks loose, i.e., Greece), we're always presented with a trade-off between higher taxes or fewer services. Given those alternatives, I would have thought wage cuts would be the first thing you would start with (regardless of whether you want to you the savings for stimulus or to reduce deficits - though politically, I could see it being an easier sell if you commit to using some of the savings for stimulus).

Wouldn't the effect of this policy also depend on debt level of workers and this debt deflation? Let's assume that everybody in economy has the same wage and that everybody has debt ratio at 130% of his income - even the unemployed (maybe they were fired only shortly, or they negotiated payment delays etc.).
.
Now if you cut wages by 1% you will (ceteris paribus) increase real debt of all employees by 1,3%. So now everybody needs to save that 0,3% more to keep their payments and consumption at the same level as before. So this lowers total aggregate demand by 0,3% even taking into account consumption of newly employed people.

Great idea! :)

But why increase workers at approximately the same wages that are cut? After all, unemployment is not evenly distributed. For a 1% cut of higher wages you can hire more than 1% of workers at low wages. :) And why stop at 1%? I am sure that governors, legislators, cabinet ministers, justices, etc., would be happy to take a temporary 10% wage cut until full employment is restored. Right? ;)

Hmmm.....I'm trying to remember what I read about the French experiment a few years back; they had an official policy of encouraging firms to hire more workers but make them work fewer hours per week. That sounds like the same thing....(I'm assuming that they kept hourly wages fixed.) Anyone recall more?

Do I sound like a monetarist? My position is that there can be both demand-side and supply-side problems. The shift between those is a state-space transition which requires a shock. Normally we want the economy to be in a supply-limited, non-money limited space where investment increases productive capacity and the money market clears. But since 2008 we have been in a state where the money market does not clear and we have a persistent output gap.

Once you stop assuming that aggregate demand cannot be a problem (Supply-side economics) and supply cannot be a problem (Keynesian economics) and realize that it is entirely possible to transition between states with the corollary that there is more than one type of recession, it's actually quite easy to synthesize Monetarism and Keynesianism together.

Back to the point, Nick has imposed a 1% tax on government workers to expand government employment. So what if we imposed a .00005% tax on all workers to increase employment? Or better yet just printed the equivalent in money and spent it? To my mind it works out to the same in the end.

Part of the problem with "Inflation" is what we mean by it. Most people mean price increases greater than their personal money supply. It means they get poorer in real terms. But what if prices stayed the same due to deficient demand and we printed money anyway? Then we can revive the economy through spending.

The current proposal is an attempt to reduce unemployment and increase the velocity of money without increasing the money supply. 'Cause inflation is always bad, in the current received wisdom. Sorry, this where I get off the Received Wisdom Train.

Frances Woolley wrote, "Now I think you're suggesting that people put in a 100% work week for 90% of their salary - but isn't that a stronger assumption than you need for your results to go through?"

That was precisely Sandwichman's "off topic" point. Thanks, Nick, for commenting over at Ecological Headstand, where the topic is not off topic and I would welcome continuation of the discussion.

http://ecologicalheadstand.blogspot.com/2011/08/intellectual-progress.html

JV: The wage cut would worsen the debt/income ratio of existing government workers, but would improve the debt/income ratio of the newly hired. My guess is that it would be roughly a wash.

Min: in principle, if you followed this policy to its logical(?) conclusion, you would reduce *every* government worker's wages until there were no longer an excess supply of that particular type of worker.

Leigh: "It's the government using its market power to buy more stuff for the same price..."

Not really. In normal times, when there's no excess supply of labour, it would be monopsony power. Sliding down along the supply curve. But a monopsonist cannot both cut wages and increase employment at the same time.

But I'm assuming the economy is in a recession so *any* employer, large or small, should have the power to cut wages (absent david's sort of counterargument).

Frances: you gotta remember that old ECON1000 distinction between average and marginal propensities to consume too! If everybody has the same *marginal* propensity to consume, the redistribution of income wouldn't affect total consumption demand. If the rich have a lower mpc than the poor, then it would increase total consumption demand.

Nick: "If the rich have a lower mpc than the poor, then it would increase total consumption demand." - yes, that's what I'm assuming, and I think that's pretty much what the empirical evidence says.

Running through this thought experiment, and not being as up-to-date on my macro as I need to be for this, is there internal consistency between the govt and private market for labour? What is the distinction between labour that works for govt and labour that works for the private sector? If they are nearly equal in terms of their labour-value I imagine the govt would get less than 1% improvement in productivity because people will leave to work for the private sector.

Assumption: A Govt employee is paid their worth. They would be worth $$$ employed in either govt or private sector.
Step 1: Gov't drops wages 1% and goes out to recruit 1% more people from the unemployed.
Step 2: Private sector observes a reduction in wages and therefore a) reduces the wages currently paid, or b) offers the govt employee a job with only a 0.5% cut.

I recognize that the private sector is only interested in hiring (or firing) people based on the demand for their output, but then how is the "output" of the govt employee valued? Why does a 1% increase in labour result in a 1% increase in output - or is that irrelevant? In which case, at the extreme how does this differ from cutting govt wages 30% (or whatever is required) and hiring everybody who is currently receiving social assistance?

Peter: that is the fundamental question. Why does it work for government wages and not private sector wages? I tried to address that question in the post, when I talked about government output and employment being assumed exogenous in the keynesian model. So it makes sense to talk about a *policy of cutting government wages and increasing government employment and output. These are all variables the government can choose. The government cannot choose private sector employment and output. These are endogenous variables, determined by the model. And the model says that private sector wages cuts will not (under certain assumptions) increase private sector employment and output.

Wait...the beginning of the post asked us to park our politics at the door. But then it brings up a political-economy explanation. Um.

I believe it boils down to the fact that Keynesian models imply that private sector actors avoid losses or deficits strenuously while governments can tolerate them due to their superior ability to borrow, print money and the fact that G is used as a balancing factor anyway. A G deficit isn't assumed to be bad the way a private sector deficit is. It's an artifact of the model's construction.

G is constructed as just another form with the profit motive removed. The fact that G, that is the Minister of Finance can act this way does not mean that Public Servants on the ground will act that way. Their micro behaviour is just the same as any other individual's.

Mandos:

I believe Nick wanted to avoid the Bob Rae: Transformer type jokes such as "Dipper in Disguise. Liberals wage a battle to destroy the evil forces of the Conservatives."

Sorry, I was born in the early 1980's, Optimus Prime is my hero.

Mandos: well, it's a fine line, perhaps. But there's "political economy" in the sense of economic explanations of government behaviour (aka "public choice"). And then there's politics.

Actually though, now that the macroeconomics of the post has been thoroughly discussed (I especially liked david's comments on the micro-rationale for wage stickiness, and rsj's balanced budget multiplier parallel) I expect I can lighten up on the KEEP OFF POLITICS! rule.

So, if you want to have a *civilised* discussion of the politics, go ahead.

(I was just afraid the politics would drown out the macro.)

Yep. david's and rsj's comments were very good ones. I learned a lot from them. That's one of the really good things about blogging.

(I was just afraid the politics would drown out the macro.)

Talk about not seeing the forest for the trees! The "macro" assumes away the politics that inhabit and define the production process where the rubber of labor meets the road of output. Such "simplification" might be acceptable as a tentative first approximation, but it appears to be intended as a decisive last word.

Nick, you appear to know almost as little about the nature of statistics and the history of economic thought as you do about the labor process itself -- that black box where flesh and blood "workers", (not algebraic "L"s) crank out usable, exchangeable stuff (not algebraic "Y") in return for a paycheck.

Don't worry, though, the reification (spell checker suggests "deification") appears to be virtually universal.

Politics is the art of explaining macroeconomic policy to those without PhD's in Macro.

Peter: "Assumption: A Govt employee is paid their worth. They would be worth $$$ employed in either govt or private sector."

Does it make sense to speak of the worth of anything, including labor, -- as opposed to price -- in a disequilibrium or suboptimal equilibrium? Presumably gov't policy is intended to move the system to a (more) optimal equilibrium, no?

I'm very late to this one, so apologies if this has already been said. You said;

"In a Keynesian recession, private sector wage cuts will only increase employment if they increase the demand for private sector output. And, in general, they won't do that."

It depends what you mean by "in general." If the central bank is an inflation targeter, then wage cuts will increase output. Is that a big if? I don't see why, aren't most central banks at least close to being inflation targeters?

Scott: you are quite right. If the Bank of Canada already has inflation at the 2% target, then anything that fiscal policy did to increase AD would be offset by monetary policy.

I assumed that monetary policy "did nothing" (which can mean anything, of course, but for the sake of this post, which is about Keynesian macro, it's easiest to assume a traditional Keynesian perspective where "holding monetary policy constant" means "holding interest rates constant").

Another way to rephrase my question is: if the government did this, would the Bank need to take note of that fact in order to keep inflation at 2%? Or, if the Bank didn't know the government had done this, would inflation rise above 2% temporarily?

Nick, I agree, I just wanted to put it out there that central banks do inflation target, hence private sector wage cuts create jobs too.

Scott: agreed. If there's excess supply of labour because of (say) imperfect competition in labour markets, and not because of a shortage of AD, then wage cuts (which might be nominal rather than real wage cuts) could reduce the natural rate, and an inflation-targeting central bank should ensure that AD rises to meet the increase in AS.

The comments to this entry are closed.

Search this site

  • Google

    WWW
    worthwhile.typepad.com
Blog powered by Typepad