The idea that the government should stand ready to provide work to all who seek paid employment but cannot otherwise find a job has been advanced by a number of authors in recent years. The term employer of last resort (hereafter ELR) has been used by authors such as L. Randall Wray (1998a,b), drawing on the analogy with the central bank willing to provide reserves to the banking system as a label for such an idea. Others have used the term job guarantee (hereafter JG) for a rather similar idea (e.g., Mitchell 2001). (1) The proposals under the heading of ELR and JG are similar, and this paper discusses both sets of proposals. However, the term ELR is generally used in this paper as this gives more of the flavor of the proposals than the term JG does: for example in terms of the jobs which would be available under the proposals.
The term employer of last resort encompasses the following ideas:
The government offers employment to anyone who seeks work but would otherwise be without a job. (2)
The wage for such jobs would be set at a "base" level (fairly close to the minimum wage where such exists and somewhat above the pre-existing level of unemployment benefits). (3)
The wage bill for those on an ELR scheme is not to be paid by raising tax revenue but rather would increase the government budget deficit (or reduce the budget surplus). In the ELR literature, this is often linked with what has been labeled the tax-driven money (hereafter TDM) view. High-powered money (hereafter HPM), or base money, is viewed as created by the state; public expenditure involves (high-powered) money creation while the payment of taxes involves money destruction. The government can, though, issue bonds which would serve to remove "excess" money. (4)
As indicated in the titles and sub-titles of papers and books on the ELR and JG (e.g., Mosler 1997, Wray 1998a) these schemes promise both full employment and price stability. The central concern of this paper is to examine whether they would be able to deliver on such a promise.
The paper begins by considering the notion of functional finance (Lerner 1943), which forms an important element of ideas on ELR. This is followed by a section which considers the nature and role of money as envisaged in the TDM approach which is often associated with the ELR proposals. Section 4 considers the budgetary costs (and deficit implications) of ELR proposals. It is argued that while the ELR budgetary costs may be relatively small, this would also be the case from any public sector employment program. The question is raised in the next section as to whether there would be jobs of a type which could fit in with the ELR proposals and what the nature of these jobs might be. The following section considers the extent to which ELR would involve underemployment and unemployment by another name. The possible inflationary implications of the ELR are next considered. This has two aspects: first to consider whether inflation would result from unemployment falling below any form of supply-side inflation barrier (such as a NAIRU, or non-accelerating inflation rate of unemployment) and second to consider whether the use of a "base wage" would bring price stability as claimed. In the next section, two notions of unemployment (and corresponding ELR workers) are considered. It is argued that the notion of the unemployed or ELR workers as a buffer stock is misleading. It is also argued that ELR workers would retain many of the characteristics of being an "industrial reserve army."
Functional Finance and ELR
The idea that governments should use the fiscal stance to support a high level of demand (rather than seek to balance the budget) is a long-standing one which can be put under the heading of functional finance (Lerner 1943). The basic justification for the use of a budget deficit to support high levels of employment is that private sector demand is inadequate to generate such levels of employment. (5) This can alternatively be expressed by saying that private savings would exceed private investment at the level of income which corresponds to this high level of demand. Private savings are S + (M - X) where S is domestic savings and M (imports) - X (exports) is the trade deficit and equal to the inflow of capital. The required budget deficit G - T is then equal to private savings minus private investment, each corresponding to the high employment level of income. Hence G - T = (S - I) + (M - X). If the right side of this equation would be positive at full employment, then the left side necessarily would be positive, in other words, a budget deficit would be required to sustain full employment.
An early advocate of the view that budget deficits should be used (as required) to reach and sustain full employment was Michal Kalecki (1944). He discussed "three ways to full employment" in terms of government spending and subsidies to mass consumption (leading to a budget deficit), stimulation of private investment, and redistribution (consideration of the foreign trade sector was left for another paper in the same volume of essays). He argued that there were limits on the stimulation of private investment for aggregate demand purposes but that the other methods were capable of securing a level of aggregate demand consistent with full employment. Thus manipulation of aggregate demand can take the form of fiscal policy (via budget deficits) and influencing savings and investment behavior. But Kalecki also argued that "the gap to be covered (to stimulate aggregate demand to reach full employment) may be so large that public investment will soon become entirely, or at least nearly, useless. In such a case it would be absurd to restrict the government spending program to public investment when a higher standard of living can be achieved by devoting a part of this spending to increasing consumption. The general principle must be that social priorities decide the nature of the government's spending programme" (368). He also argued that "the proper role of private investment is to provide tools for the production of consumption goods, and not to provide enough work to employ all available labor.... Both public and private investment should be carried out only to the extent to which they are considered useful. If the effective demand thus generated fails to provide full employment, the gap should be filled by increasing consumption and not by piling up unwanted public or private capital equipment" (371). While the general aim is to achieve full employment, there are different (three) ways of reaching the general objective, and the balance between the different ways depends on social priorities. Further, the achievement of full employment may be hampered by inflationary pressures, though "inflation will only result if effective demand increases so much that a general scarcity of labor or equipment (or both) arises" (361). (6)
The ELR proposals differ from previous ones such as those made by Kalecki in two respects. First, the ELR provides a mechanism for the continuous achievement of high levels of employment: a person made unemployed today is promised a job immediately. The proposals of Kalecki would be to seek to forecast the levels of government expenditure and taxation (say for the year ahead) required for high levels of employment. But the forecasts may go awry and government plans may not come to full fruition. The use of the term employer of last resort and the analogy with the central bank as "lender of last resort" suggests that ELR jobs would be immediately available to anyone who would be otherwise unemployed.
Second, the "make work" component of employment is provided at the "base wage." "Mainline" public sector jobs pay wages which correspond to the skills required and bear some relationship with pay for comparable jobs in the private sector. (7) Further, it is because the "output" of a mainline public sector job is deemed worthwhile by the government that the job is provided. For the ELR jobs, while the output of the job is (generally) beneficial, the job is provided to mop up unemployment rather than for the benefits of the output. If there were a high level of employment because of high private sector demand, then the ELR jobs would not be provided. But the mainline public sector jobs would be provided in general whatever the level of private sector demand.
The ELR proposals have the advantage of providing a fiscal "fine tuning" whereby jobs are always available, though this is subject to doubts raised below on the availability of appropriate jobs. It has the disadvantage that these jobs are provided at the basic wage only. Further, jobs are provided in order to "mop up" unemployment rather than to let "social priorities decide the nature of the government's spending programme" (Kalecki 1944, 368).
It can be noted that "in Lerner's functional finance approach to government spending, full employment is the abolition of all unemployment up to frictional unemployment. In Lerner's model, frictional unemployment consists of what is now known as 'frictional unemployment' and structural unemployment. Functional finance, according to Lerner (1951, 192) is ineffective in solving what is commonly referred today as structural and frictional unemployment" (Kadmos and O'Hara 2000, 11). Further, "Abba Lerner considers the difference between structural and cyclical unemployment to be paramount when discussing the relevance of functional finance" (Kadmos and O'Hara 2000, 11).
Now consider unemployment at any time to be composed of demand-deficient, frictional (search), and structural unemployment (recognizing that in practice it would be difficult to precisely identify each component and that they may overlap). The ELR scheme seeks to remove demand-deficient unemployment through the provision of the required aggregate demand, albeit that this demand is focused on ELR jobs. Frictional unemployment is apparently...