MMP Blog #48: Is The Job Guarantee Necessary?

Share on facebook
Share on twitter
Share on linkedin
Share on email
Share on reddit
Share on whatsapp

Originally published April 29, 2012 on the New Economic Perspectives blog.

Over the past 11 months, or so, we have examined Modern Money Theory. This is the proper paradigm for analyzing all modern countries that use their own currency.

Some have wondered whether we can separate MMT from the Job Guarantee (JG): can one accept MMT while rejecting the JG?

To be honest, I find this to be a rather strange question. We have a modern theory of Bubonic Plague. Until medical science developed a theory of disease that can be caused by microbes too small to be seen with the naked eye, all sorts of explanations of the cause of the Plague were offered. The most popular was “bad air”. The preventative measures offered (quarantine of affected individuals, evacuation of cities, burning of the property of the deceased), and improvement of public sanitation actually were quite effective—indeed, in the richer countries the plague was almost banished even before the “germ” theory of disease was ever developed. (To be sure, there were false starts—such as killing cats, that had helped keep rats in check!)

We now “know” that the disease is caused by the bite of an infected flea, and that once bitten a victim can be treated with antibiotics. Although the Plague is still contracted occasionally (especially in the American West), it doesn’t lead to a plague because swift treatment prevents a pandemic.

OK, should we separate the theory of Bubonic Plague from the policy response? I think most reasonable people would say there isn’t much reason for developing a theory of disease unless you’re working toward prevention and treatment. I think almost everyone would castigate any researcher who was only interested in uncovering the causes of the Plague, but who then opposed a policy response on the argument that some deaths serve a useful purpose—disciplining the population, weeding out the weak, and serving as cautionary examples for the imprudent.

And, yet, such has been the mainstream economist’s response to the “discovery” by J.M.Keynes in 1936 that unemployment is caused by insufficient demand. We know—beyond any question whatsoever—what causes most unemployment. Indeed, we are more certain of the main cause of unemployment than we are of the main cause of the Black Plague of the mid 1300s (there actually is a bit of a controversy about that—contemporaneous reports of the disease that plagued Europe do not fit what we know about Bubonic Plague, but that is a digression). And so we have developed a whole approach to the policy response to unemployment—the Phillips Curve NAIRU—that actually says it is OK to ignore the disease of unemployment that afflicts somewhere around 200 million people globally even at the peak of the business cycle.

Think about that. Since 1936 we’ve had a large proportion of economists and policy makers who assert that unemployment on a massive scale is good policy—to discipline the workforce, to weed out the weak, and to serve as cautionary examples for the imprudent. All of this is in the name of maintaining price stability.

Now, to be sure, some of these advocates for unemployment believed that there really wasn’t much government could do, anyway. Like a household, government is constrained by its revenue. And so even if it wanted to lower unemployment, it could not “afford” to do so.

That is where Abba Lerner’s functional finance, and our MMT (derived from it) come into play. We know that is false. Anyone who understands modern sovereign currency knows—beyond any doubt whatsoever—that sovereign government can “afford” full employment. It could choose among a number of methods of reaching full employment, but can afford any and all of them. It can “prime the pump”, it can subsidize employment by private employers, or it can operate a JG program.

I want to make clear here that throughout this primer I have included all nations that issue their own currency in my definition of “modern money” or “sovereign currency”. In this I deviate from some other expositions. Some exclude countries that peg or manage their exchange rates. I do not. I see these as positions on a continuum—with a floating rate on one end and a pegged rate on the other. Clearly, the floating rate gives more domestic policy space. And that is the exchange rate system I advocate—although many governments are not in a political situation in which they can float.

However—and this is the important point—pegging or managing an exchange rate is a self-imposed constraint. Any country that issues its own currency can move to a floating rate—technically we know how to do it; politically it can be hard. But it can be done.

And that is what makes a sovereign currency issuer different from a household or firm. The household or firm cannot decide to move to a floating rate. It is a currency user. Its government, however, is a currency issuer—even if it chooses to peg. It can choose to float; the user cannot.

And, so, the currency issuer can ALWAYS afford full employment if it removes self-imposed constraints.

Personally, I think it is absolutely reprehensible that anyone who understands MMT would choose to oppose JG. Once one understands that we do not have to force millions to suffer involuntary unemployment, then the ethically defensible position is narrowed. Unemployment is in that case unnecessary, and anyone who advocates using unemployment as a policy tool must mount a strong argument in defense of maintaining unemployment.

We would not accept anything less if one were to argue against treating deadly diseases like the Bubonic Plague or Smallpox—and yet unemployment is, today, without much question an infinitely bigger problem than either of these diseases. Yet, we have most economists—including self-professed progressives—arguing for maintaining millions of people in a condition of involuntary unemployment.

Before turning to the JG, let us quickly review what we understand about modern money systems.

Recall that we first began with the logic of a sovereign currency: government names a unit of account, imposes a tax liability in that unit of account, and issues its own IOU in that same unit in its own spending. Government must spend first before the taxpayers can obtain “that which is necessary to pay taxes”. That is the logic of “taxes drive money”.

Next we use MMT as a description—how modern governments actually tax and spend. This is necessarily more complicated and must be nation-specific. Most nations have a nominal separation of their Treasury and their Central bank; and many impose constraints on each. It is common to prohibit the central bank from directly lending to the treasury; and to require that the treasury write checks on its deposits at the central bank. These self-imposed constraints then give rise to various operating procedures that are designed to allow the treasury to obtain those deposits it needs at the central bank, and minimize disturbances to banking system reserves.

There is no need to repeat all of the analysis here—I merely want to distinguish between explanations that explicate only the logic from those that explain how modern money “really works”, that is, descriptions of the operation of real world sovereign currencies.

Finally, we know that logically government faces no financial constraint, and that it spends first before it taxes or sells bonds. However, government can self-impose constraints on its spending (ie: balanced budget requirements or debt limits), and normally has a budgeting process requiring congressional and presidential approval.

But, logically, it can change those constraints, or even ignore them. Logically, it can “afford” anything for sale in its own currency. The problem is not solvency, but rather inflation and exchange rate depreciation. Too much spending could cause either of these to occur, under some conditions.

Again, this is not the place to explore the conditions in which more government spending would cause inflation or currency depreciation.

The important conclusions are two: a) taxes or other obligations create a demand for the currency; and b) the value of the currency depends on what one must do to obtain it.

As our moms always told us “if only money grew on trees”—then its value would be worth the effort of harvesting it. With a sovereign currency, it really isn’t a matter of “harvesting” but the ease of obtaining it does matter. When the economy goes beyond full employment, continued expansion of the government’s spending is likely to bid up wages and prices.

If government ups its bid for labor from $10 per hour to $20 per hour, it is possible that wages and prices more generally will rise. The “ease” of obtaining currency rises, and we can expect that its value falls. How close this (inverse) relation is will depend on capacity utilization and the institutional setting (pricing power of firms and workers).

And so we return to the JG. As you know by now, we have formulated the JG as a wage-stabilizing buffer stock. With its fixed wage and benefit package, and its infinitely elastic supply of jobs, it creates full employment without chasing bids for workers upward. It “sells” workers at a small markup over the program’s compensation. As a matter of logic, it cannot pressure wages once the compensation level is established as the nation’s base.

It is a “fixed price, floating quantity” model—much like commodity buffer stock schemes. It stabilizes the “price” of labor—the wage. And, yet, the labor is “fully employed” in the sense that anyone willing to work can get a job at the JG wage.

Some advocate demand stimulus over a JG.

OK, explain to me how pumping up the demand for higher skilled and educated workers—setting off a bidding war for them—will cause jobs to trickle down to the less skilled and less educated workers WITHOUT causing wages and prices to rise.

Others prefer a BIG (basic income guarantee—welfare payments to all, payments that are not means tested).

Again, explain to me how this will create jobs for the less skilled and less educated workers WITHOUT causing wages and prices to rise.

(In truth, advocates for both strategies ALWAYS simply change topics—they NEVER address this question. Which, I suppose, is OK, but in that case they cannot legitimately criticize the JG on the argument it will supposedly cause inflation—because their own strategies are far more inflationary, indeed ONLY create jobs for the unemployed by causing wages and prices to rise so that firms will try to substitute into lower skilled and educated workers.)

Some say that the inflationary impact of the JG comes not through the wage but rather through the “demand multiplier” because the new workers can consume. Think about that. This view is even crueler than the view that we need to keep millions of people unemployed to keep prices down. Because this view is that we need to ensure that millions cannot afford to eat, else they’d push up prices.

So it is not just unemployment that these critics want—they want abject poverty! A permanent, starving, underclass is necessary so that the better-off can enjoy lower prices that come from insufficient aggregate demand. That, too, is a reprehensible policy position. (And it is dynamically incoherent—a point I won’t pursue here.)

Throughout our discussion of JG I have always presented it as an “add-on”. We can retain (and even add) any safety net we want. We can retain (and even add) any other macro-stabilizing policy (monetary or fiscal) we want. They are still policy options once we’ve added the JG. We can have the BIG and we can have the pump-priming—if we find some need for them. Personally, I do not think we’ll need either. But we can implement the JG and see—if we still need to fight inflation or insufficient demand or exchange rate instability, so be it. We have all policy options available.

The only thing we eliminate is use of involuntary unemployment as a policy tool. With the JG as an add-on, we can still use all other policies at our disposal, but in terms of employment all these will do is to change the size of the JG pool: tight policy increases the pool, easy policy reduces it. But through the thick and thin, anyone can have a JG job instead of unemployment.

So, can we have MMT without a JG? Certainly! We already have modern money systems with sovereign currencies and without the JG. MMT is the proper way to analyze these. I believe it is a policy mistake to operate a modern money system without a JG—but that is what almost all countries do. MMT allows us to analyze them, and to offer policy recommendations.

But if we leave out the JG in our recommendations, we are seriously remiss in our advice.

I realize I’ve probably not explored this issue fully; let me see the comments and I’ll decide if we need a Part II.

Blog comments and responses

Leave a Comment

Your email address will not be published. Required fields are marked *

Share this post

Share on facebook
Share on google
Share on twitter
Share on linkedin
Share on pinterest
Share on email
Scroll to Top Skip to content