A Fascinating Report from the Government Accountability Office

By Benjamin Zycher 

The bureaucracy’s analysis of green jobs leaves a lot to be desired.

No, really. Like a million monkeys pounding on keyboards, even the Beltway occasionally produces something Shakespearean, and fortunate indeed we are to be alive in the summer of 2013, as the Government Accountability Office (GAO) has issued a report announcing that the “[Department of] Labor’s Green Jobs Efforts Highlight Challenges of Targeted Training Programs for Emerging Industries.”
B-O-R-I-N-G, you say? You are oh, so wrong. GAO was happy to translate that bureaucratese right up front: “Of the $595 million identified by Labor as having been appropriated or allocated specifically for green jobs activities since 2009, approximately $501 million went toward efforts with training and support services as their primary objective…”


Yes, I quoted that correctly: of the $595 million, over 84 percent was spent on “training and support services” rather than actual employment. A cynic would ask whether those employed in the provision of “training and support services” constitute the remaining 16 percent, thus transforming federal “green jobs” efforts into a permanent exercise in tail-chasing. But I am not a cynic, and I refuse to engage in the kind of mudslinging that has made the modern policy environment so toxic. I insist that we stick to the issue. Let us review what kinds of employment are viewed by the feds as “green,” and then examine a few numbers before turning to the question of whether “green jobs” programs can expand employment significantly even in principle.
With respect to the definition of “green goods and services” and attendant employment, the Bureau of Labor Statistics offers this formal definition:

[Green] goods and services [are those] produced by an establishment that benefit the environment or conserve natural resources. Green goods and services fall into one or more of the following five groups: (1) production of energy from renewable sources; (2) energy efficiency; (3) pollution reduction and removal, greenhouse gas reduction, and recycling and reuse; (4) natural resources conservation; and (5) environmental compliance, education, and training, and public awareness.

A few observations:
What leads to environmental improvement? Strictly speaking, all productive activities “benefit the environment” by increasing aggregate wealth, and it is greater wealth – a bigger economy per capita – that increases the social demand for improved environmental quality and that provides the resources necessary to achieve it. A substantial scholarly literature exists on this topic, examples of which can be found here, here, and here. As economies start to grow with expanding industrial activity and the like, they tend to display lower environmental quality due to increasing emissions, even with rising per capita incomes. But as those incomes rise above about $8000-$12,000, individuals seem to demand better environmental conditions, and political systems tend to respond with more stringent regulatory constraints on polluting activities.

The required use of wind power exacerbates air pollution problems because the cycling of the backup units results in inefficient operations. New peer-reviewed research shows that biofuels create a similar array of pollution problems.

Accordingly, were the BLS to think this issue through, its definition would become more nuanced. The production of goods and services as driven by market forces is “green” in principle, and the arbitrary distinctions made by officials and bureaucrats as to which goods and services are “green” or not are irrelevant, except in cases in which productive activities create adverse environmental impacts – “externalities” – not reflected in market prices.
Thus has BLS endorsed economic growth stronger rather than weaker, without quite realizing it. Notwithstanding all the casual assertions by government officials that, for example, more regulations limiting greenhouse gas (GHG) emissions would “strengthen the economy,” the real issue is the value of the environmental quality achieved versus the value of the resources consumed in the effort. In summary: an immediate cut in U.S. emissions by half would yield a reduction in global temperatures of 0.1 degrees Celsius 100 years from now. Because annual temperature variability is greater than that figure, the predicted effect could not be measured reliably. How much should we be willing to pay for that?
What is “conservation”? With respect to the production of goods and services that “conserve natural resources” and that therefore supposedly are “green”: does BLS – or, for that matter, any federal agency – have a useful definition of “conservation”? It appears to be the case that the answer is no, although the EPA offers a definition of “sustainability,” a related concept, that is deeply problematic.
From an analytic perspective, “conservation” must mean a shift in the consumption or production of some resources from the current time period into a future one. Since no natural resource – say, natural gas – is being depleted this year or over some short time horizon, market forces do and always have “conserved,” that is, the market is reserving the vast bulk of the natural resource base for the future, even apart from the effects of exploration and technological advance. (That is why not all the recoverable oil in the world was consumed decades ago, even though the market price of crude oil always was greater than zero, which is to say that using it would have yielded value.) The argument that productive activities are “green” if they “conserve natural resources” actually is an assertion that market forces yield too little “conservation” or too much resource use in the short term. Suffice it to say that the analytic basis for that implicit premise remains entirely obscure; for a summary discussion of the powerful incentives imposed by market forces to preserve resource capacity for the future, see my AEI Energy and Environment Outlook on wind and solar power. It may be the case that various tax policies or other government actions result in too little conservation, but then it is those policies rather than market outcomes that are insufficiently “green.” What is the basis for the implicit premise that the same political competition that yields government policies resulting in too little conservation somehow will produce new policies with the opposite effect?

A case for government ‘efficiency’ standards must be based upon some reasonable argument that markets undervalue such efficiency (and that government policies in practice will yield an improvement in the productivity of resource use). It is not quite clear what that argument would be.

Is “renewable” energy green? Notwithstanding the “green” virtues of renewable energy widely assumed in the conventional wisdom – and therefore worthy of considerable skepticism – the environmental problems caused by renewable energy are substantial. Consider wind power as an example. The “rare earth” minerals needed for wind turbines are highly toxic heavy metals – examples are dysprosium, neodymium, praseodymium, and samarium – requiring great care in use and disposition. There is the problem of bird deaths, including large numbers of protected species. There are important noise and light flicker problems created by wind turbines, as well as the unsightly use of massive amounts of land. Most important, even in terms of conventional effluents, the integration of wind power into electric grids actually increases emissions of conventional pollutants because the unreliability of wind power creates a need for conventional backup generation capacity – coal and gas units – that must be cycled up and down as the winds prove weak and strong. In an engineering study of the attendant emissions effects for Colorado and Texas, Bentek Energy concluded that the required use of wind power exacerbates air pollution problems because the cycling of the backup units results in inefficient operations. New peer-reviewed research shows that biofuels create a similar array of pollution problems.
Is “energy efficiency” efficient? “Energy efficiency,” in the aggregate, is a reduction in energy consumption per unit of national output (GDP), and in a micro sense is a reduction in energy consumption per unit of output (e.g., miles driven) by heating and cooling systems, vehicles, lighting, industrial equipment, and the like. Such reduced energy use is not free, of course: the energy-using capital (e.g., autos) must be improved in some way to allow for the same quality of service delivered with less energy, and such improvements – an example is vehicle transmissions with more gears – are costly. Were that not the case, obviously, markets would provide greater “energy efficiency” without any prodding from government. “Energy efficiency,” therefore, is not the same as economic efficiency.
Accordingly, a case for government “efficiency” standards must be based upon some reasonable argument that markets undervalue such efficiency (and that government policies in practice will yield an improvement in the productivity of resource use). It is not quite clear what that argument would be. A pollution argument would not suffice because pollution limits and requirements for abatement equipment are the outcome of political competition in legislatures and regulatory agencies; however imperfect or opaque, there is no particular reason to believe that such regulatory requirements are too lax.
Another argument often made is that market prices do not reflect the defense cost of protecting overseas oil fields or the sea lanes used to import oil, but an attempt to allocate such costs of the U.S. force structure across many defense functions would be arbitrary. Moreover, the economic effect of increases in international oil prices would be virtually identical whether the United States imported all or none of its oil; “dependence” is irrelevant analytically.

Employment in the installation of ‘energy-efficient’ windows is ‘green,’ while employment in the installation of ordinary windows is not. Accordingly, subsidies for installation of the former yield an increase in ‘green’ employment, regardless of the larger reality that this definition is heavily arbitrary.

Another common argument is that the market discounts the future too heavily, a premise that is deeply problematic. The analytic basis for that assertion is far from clear, and there is no particular reason to believe that government as an institution has incentives to adopt a time horizon longer than that relevant for the private sector. Indeed, one plausible argument is that the time horizon for many public officials is the next election. Of course, to say that a given official views the next election as the “long run” is different from arguing that government acting collectively would display the same behavior; but the profit motive combined with the market rate of interest provide incentives for the market to consider the long-run effects of current decisions, while no similar constraint operates in the public sector, except perhaps notionally through democratic processes. In addition, such policies as campaign finance restrictions may have had the effect of weakening the constraints that political parties can impose upon officeholders. As the parties are long-lived institutions with some incentives to adopt time horizons longer than those of individual officeholders, the net effect may have been a tendency to discount the future effects of policies more heavily. Moreover, the corporate income tax – a government policy – yields incentives for much of the private sector to discount the future more heavily than otherwise would be the case, as investments must earn a higher expected before-tax return in order to yield the market rate of return after taxes.
There is the further matter that energy efficiency standards, by reducing the marginal cost of operating autos and equipment, encourage more-intensive use – this is the so-called “rebound” effect – thus reducing the actual ex post energy savings assumed ex ante to result from investments in energy “efficiency.” In short, it is far from clear that “energy efficiency” standards produce an improvement in aggregate resource use – that they are worth what they cost – and thus the size of the economy. Resource waste is at least equally plausible as a net result, and it is not “green.”
At this point, a few numbers are revealing. BLS reports that in 2011 (the last year for which data are available) total employment in the production of green goods and services was 3.4 million, of which 886,000 were employed by federal, state, and local governments. Recall from the aforementioned GAO report that “environmental compliance, education and training, and public awareness” jobs are counted as green, meaning that regulators, staffers, bean-counters, and other such attendant government employment is counted as “green.” Moreover, those employed by the private sector in functions required to respond to, deal with, and implement the rules and proposals generated by government also are counted as “green.” Is the private/government regulatory ratio 10 percent? Or one-to-one? We do not know, but some nontrivial private sector employment must arise as a response to the “green” government employment. Who knew that lawyering and lobbying could be so socially responsible?
The BLS estimate is that total green employment increased by about 158,000 jobs between 2010 and 2011. The real problem is definitional: of that total increase, 102,000 jobs were in construction, largely as a result of changes in the “energy efficiency” of construction materials. An example: employment in the installation of “energy-efficient” windows is “green,” while employment in the installation of ordinary windows is not. Accordingly, subsidies for installation of the former yield an increase in “green” employment, regardless of the larger reality that this definition is heavily arbitrary.

The industries of interest are ‘emerging’ only because of subsidies, which can support only a limited amount of employment unless they are allowed to become massive and permanent.

That the production of “green” goods and services must be mandated and/or subsidized is strong evidence of uncompetitiveness fully consistent with the GAO’s finding that five-sixths of the government dollars allocated for green employment were spent on training and support services. Unless government itself is to be the employer, or unless government is prepared to subsidize uneconomic industries indefinitely, “green jobs” programs cannot overcome that central reality. This is another way of saying that the industries of interest are “emerging” only because of subsidies, which can support only a limited amount of employment unless they are allowed to become massive and permanent.
Those subsidies must be financed with current or future taxation. Those taxes will have adverse employment effects, some of which are likely to be among those jobs that would be defined as green. More fundamentally, an expansion of green employment created by government policies actually would be an economic cost rather than a benefit for the economy as a whole. However counterintuitive, suppose that policy support for green industries had the effect of increasing the demand for, say, certain appliances. That clearly would be a benefit for the producers of such appliances, or more broadly, for owners of inputs in that production, including appliance workers. But for the economy as a whole, the need for additional labor in an expanding appliance sector would be an economic cost, as the resources used to produce the appliances would not be available for use in other sectors. Similarly, the creation of “green jobs” as a side effect of various policies is a benefit for the workers hired (or for those whose wages rise with increased market competition for their services). But for the economy as a whole, that use of scarce labor is a cost because those workers no longer would be available for productive activity elsewhere.
There is the further matter that an expansion of “green” sectors must mean a decline in some other sector(s), with an attendant reduction in resource use there; after all, resources in the aggregate are finite. If there exists substantial unemployment and if green employment is not highly specialized, a short-run increase in total employment might result. But in the long run – not necessarily a long period of time – such industrial policies cannot “create” employment; they can only shift it among economic sectors.
In short, an expanding “green” sector must be accompanied by a decline in other sectors, whether relative or absolute, and creation of “green jobs” must be accompanied by destruction of jobs elsewhere. At best, the BLS count of green employment is a gross figure that ignores the larger employment effects of government “green” policies.
That there are no free lunches is a shorthand way of asserting that larger reality, one that even the federal government seems to have recognized: BLS has announced that data on employment in the production of green goods and services will be collected no more. This is being blamed on the budget sequester, but it is easy to believe that this particular cut would have been made in some other program were the prospective data more promising. Perhaps I am a cynic after all.
Benjamin Zycher is a visiting scholar at the American Enterprise Institute.