Because a job guarantee program is a novel policy, we start with an overview of the program from the MMT perspective. Like most macroeconomists, MMT economists express concern with maintaining a stable price level and full employment when assessing policy.3 What makes their policy recommendations unique is that they recommend a JG program not only to achieve full employment but also to stabilize the price level. The JG, operated by the government through federal agencies, would unconditionally offer a guaranteed job to all who want one at a fixed nominal wage (Wray 1998c; Mitchell 2017; Tcherneva 2020). 2. See (Thompson 1982; Glasner 1989; Hendrickson 2018) for a related proposal, the job guarantee under monetary policy, which has many of the same properties as a nominal GDP target. 3. We will use terms such as "MMT policy," even though there is a distinction between the positive theory and the normative policy proposals. "MMT policy" should be read as shorthand for "a common policy proposed by MMT economists." 3 There are two different models of how a job guarantee program might work. The first would call for something akin to the New Deal’s Works Progress Administration and entail centralized administration and decision-making. The second, worked out in detail by Tcherneva (2012, 2018), calls for centralized spending grants to states, municipalities, and various NGOs and non-profits, which would then make their own decisions at a decentralized level. According to advocates of the program, the JG would primarily work to produce public goods, particularly in (green) community investment or community art (Forstater 2006; Tcherneva 2018, 2020). Tcherneva (2018) describes the MMT proposal as allowing for guaranteed jobs in the arts, building green infrastructure, performing child and elder care, and food desert relief. More specifically, Tcherneva (2018, p. 21) gives examples of typical jobs: I. “A local artist collective employs painters, actors, musicians, and stagehands to run year-round productions for the community. They organize school outreach programs, run summer camps, and offer free art and music classes and literacythrough-the-arts courses for special needs youth. They collaborate with local schools in offering art enrichment programs.” II. “A former coal mining community experiences city blight, mass unemployment, and a high incidence of health problems. The JG organizes a comprehensive program for restoring the natural environment using the abandoned coal mine, based on existing best practices.” Hence JG output will typically go unpriced since the program will tend toward producing public goods. Neoclassical economists may stress the need to finance such a program after looking at the government’s intertemporal budget constraint. Financing does not present a difficulty to MMT economists, mainly because, in their conception, governments with sovereign currencies can afford anything; their only constraint is an inflation 4 constraint (Mosler 1997; Wray 1998b; Bell 2000; Mitchell, Wray, and Watts 2019). A government can "afford" anything that does not generate inflation by printing money. Therefore, even though the size of the JG may shift endogenously as conditions in the labor market change, the government can afford it. To grasp the fiscal magnitude of the program, Tymoigne (2014, p. 526) argues that a JG program would cost somewhere between 2 percent to 5 percent of GNP for an economy that faces unemployment rates around 4 percent in booms and 11 percent in busts. While one could advocate for the JG merely because of its social safety net property, MMT economists emphasize its ability to stabilize prices. Mosler (1997, p. 168) argues, "In addition to eliminating involuntary unemployment, the [JG] policy can be shown to provide price stability" (emphasis added). Mitchell (2017, p. 60) agrees, writing that a JG "would provide a macroeconomic stability framework designed to deliver full employment and price stability" (emphasis added).4 Indeed, MMT economists pitch the JG explicitly as an inflation management tool to replace the Phillips-curve-based non-accelerating inflation rate of unemployment (NAIRU) system employed by central banks today. Theoretically, the JG would stabilize prices in two main ways. First, in the background, as an automatic stabilizer, workers migrate in and out of the JG sector endogenously, and the fixed nominal wage and guaranteed job promote price stability. By setting a wage floor and a guaranteed job, there is a floor on nominal income so that if a recession occurs, the government can bound deflation. Hence the government ensures that nominal income cannot fall too low. The outside option of a JG for workers and the large pool of JG workers from which to draw for employers helps discipline the wage bargaining process such that wages cannot rise as quickly as they otherwise might during booms 4. It is not that the JG is a good policy that has an extra side benefit of stabilizing policy. The JG is explicitly a tool for stabilizing prices. In a blog post responding to critics, Wray (2019a) "MMT does have another tool to maintain price stability. It is the JG approach to full employment. It has always been a core element of MMT." 5 and cannot fall as quickly as they might during busts. From our perspective, this is functionally the same as our current system, with unemployment insurance serving as the primary automatic stabilizer. Note that we discuss this and other automatic stabilizers in greater detail in Section 4. Moreover, although the JG program is the primary policy option for achieving stable prices, it is intended to complement other price stabilizing policies including regulation, addressing "bottlenecks," price controls, and using taxes to temper aggregate demand when necessary (Fullwiler, Grey, and Tankus 2019). For example, when economies are in boom phases, tax revenues rise as a consequence of greater profitability, which can automatically act as a mechanism for slowing private and moving workers to the JG sector, consequently anchoring prices (albeit with sufficient fiscal support). For this paper, the relevant price stability mechanism of the JG is the one purported to replace the dual mandate of central banks. MMT economists frequently contrast the standard non-accelerating inflation rate of unemployment (NAIRU) concept with their own innovation, the non-accelerating inflation buffer employment ration (NAIBER) (Mosler 1997; Mitchell 2017; Mitchell, Wray, and Watts 2019). Under the NAIRU approach, which is dominant today, central banks manage inflation via interest rates, using some form of Phillips curve reasoning to increase unemployment through higher interest rates when inflation gets too high and vice versa. Under the NAIBER approach, defined as the ratio of JG workers to total employment consistent with stable prices, the government manipulates some policy lever to ensure that the buffer employment ratio (BER) remains consistent with stable prices, analogous to the NAIRU. However, this mechanism should not be understood as Keynesian fine-tuning; Wray (1998c, p. 543-544) notes that the critical difference is that, rather than fine-tune the economy as a whole, policymakers would only fine-tune the BER. In contrast to the prevailing policy regime, the policy lever for manipulating inflation 6 resulting from wage pressure under MMT is not interest rates. MMT economists argue interest rates should be set at zero (Mosler and Forstater 2005). Hence MMT advocates instead for fiscal policy as the proper lever for manipulating the JG by using it to move workers from the JG sector to the non-JG sector. When inflation rises, the role of the fiscal authority is to manipulate fiscal policy settings "to reduce the level of private sector demand. Labour is then transferred from the inflating private sector to the fixed wage JG sector and the BER [the ratio of JG employees to the total labor force] rises" (Mitchell, Wray, and Watts 2019, Ch. 19). Mitchell and Mosler (2002, p. 250) concur: "If inflation exceeds the government’s announced target, tighter fiscal policy would be triggered to increase the BER, which entails workers transferring from the inflating sector to the fixed price JG sector. Ultimately this attenuates the inflation spiral."5 Mitchell (2017, p. 70) agrees: "The value of the JG is that the government always knows that if total spending levels come up against the real capacity of the economy, then they are able to tighten fiscal policy without creating unemployment. In normal times, the JG would be a very small program but essential to those who would otherwise be excluded by private employers." In a foundational paper, Mitchell (1998, p. 552) describes the mechanism similarly: [I]f the private labor market is tight, the non-buffer stock wage will rise relative to the BSE [JG] wage, and the buffer stock pool drains. The smaller this pool, the less influence the BSE wage has on wage patterning. Unless the government stifles demand, the economy will then enter an inflationary episode, depending on the behavior of labor and capital in the bar- gaining environment. . . In the face of wage-price pressures, the BSE/ELR approach maintains inflation control in much the same way as monetarism—by choking aggre5. MMT economists exhibit substantial optimism in the ability of policymakers to use taxes to manage inflation. See Wray (1998a, p. 8-10), Bell (1999), Nersisyan and Wray (2010, p. 14), Tymoigne and Wray (2015, p. 26-8), Wray (2016, p. 10), and Nersisyan and Wray (2019, p. 8-9), all of whom advocate using taxes as one policy tool to fight inflation. While these authors do not explicitly argue such, we assume they mean a discretionary increase in taxes since they never mention an entirely mechanical, rule-based tax system. 7 gate demand and inducing slack in the non-buffer stock sector. In private correspondence, Warren Mosler says that "if a shrinking ELR pool is not answered with demand reducing measures, other prices will rise relative to the ELR wage and old fashioned inflation can follow." These "demand-reducing measures" turn out to be some form of policy. "As the BER rises, due to an increase in interest rates and/or a fiscal tightening, resources are transferred from the inflating non-buffer stock sector into the buffer stock sector at a price set by the government; this price provides the inflation discipline" (Mitchell 1998, p. 552). Numerous similar explanations appear throughout the literature. From the perspective of MMT, the JG is superior to relying solely on either unemployment benefits or UBI for several reasons. First, the JG is seen as a better tool than either one because it has the unique property of allowing people to remain employed in some capacity and thus maintain work skills and habits while they continue to look for non-JG work. On the other hand, if unemployment benefits or UBI are used, then people may become idle and drop out of the workforce entirely. Second, while all three entail large government expenditures, only the JG has some prospect of producing public goods. Neither the UBI nor unemployment insurance has any hope of producing goods and services, except perhaps funding job search for the unemployed. However, this property does need to be traded off with the JG’s apparent production of public goods; presumably, JG workers will have relatively little time to search for private sector work. However, MMT economists do not preclude the continuation of unemployment insurance alongside the job guarantee program. Third, the UBI has no inflation-fighting properties, and MMT economists generally disparage the inflation-fighting properties of unemployment as socially unproductive since it relies on an “unemployment buffer stock” (e.g., Mitchell (2017)). As Mitchell (1998, p. 552) points out, "The disciplinary role of the NAIRU, which forces the inflation adjustment onto the unemployed, is replaced by the compositional 8 shift in sectoral employment, with the major costs of unemployment being avoided. That is a major advantage of the BSE [JG] approach."