About fifty million people throughout history have tumbled the fact that there are an infinite number of public sector type jobs that the unemployed could to: sweeping the streets, maintaining public parks, etc.
These advocates of having government act as employer of last resort (ELR) nearly always advocate PUBLIC SECTOR type work, rather than PRIVATE SECTOR type work. And there is an appealing logic here, namely that the output of the public sector is GIVEN AWAY rather than SOLD. Thus demand does not need to be raised to create said public sector type jobs. (At least that is presumably the logic employed: the logic is rarely spelled out.)
That would seem to mean that there cannot be any inflationary effect from said public sector jobs. Unfortunately there is a flaw in that argument, as follows.
If unemployment is above NAIRU, or the “natural level” or the “inflation barrier” level, as Bill Mitchell calls it, a straight rise in demand is far preferable to any sort of ELR job. So let’s concentrate on the scenario where ELR really comes into its own, that is where unemployment is at or below NAIRU.
If an ELR scheme involves just the ex-unemployed and no other factors of production like permanent skilled labour, capital equipment or materials, it will be hopelessly inefficient. On the other hand, the scheme CANNOT order up the latter “other factors of production” (OFP) from the regular economy, because if the economy is at NAIRU, no extra demand is permissible.
So ELR is in a bind. At least it is certainly in a bind if it takes the form of schemes which consist of new or “specially set up” employers, as was the case with WPA – the main “make work” scheme in the U.S. in the 1930s.
Alternatively, an ELR scheme CAN consist of allocating the unemployed to jobs with EXISTING public sector employers (as was the case with the Comprehensive Employment and Training Act (CETA) system in the 1970s). But there is a problem here, which is that public sector employers are under similar cost cutting and output maximising incentives as the private sector. That is, if a public sector employer can cut costs by hiring subsidised ELR people and have them replace staff that the employer actually has to pay for, then the employer will be tempted to do so. Indeed, this abuse occurred to some extent under CETA.
Thus it is necessary to have rules in place governing the system (as indeed was the case with CETA). And those rules need to ensure as far as possible, that those taken on under ELR would not have been hired but for ELR. Of course the latter objective can never be attained with perfection, but as long as the objective is more or less hit, then the benefits of ELR will hopefully outweigh the costs.
Now if the above rules ensure that public sector employers are induced to expand output by taking on ELR people (rather than order up more of that inflationary OFP) then presumably the same rules applied to private sector employers will have the same result: that is private sector employers, if offered ELR employees on the same conditions as they are available to public sector employers, will expand output (given an increase in demand) by taking on ELR people rather than by ordering up more of that inflationary OFP.
Ergo there is no reason to confine ELR to the public sector.