The Environmental and R&D Analogues
There are two successful examples of regulation—understood in the broad sense of public measures addressing externalities—under dynamic and uncertain conditions. The first is the Defense Advanced Research Projects Agency (DARPA) and its offspring, the Advanced Research Projects Agency-Energy (ARPA-E).
They respond to the characteristic learning externalities that arise at the far frontier of science and technology, where for now there may well be no solution at all to a particular problem, and the search for one will likely end in costly disappointment. Worse still for the private investor, even when the search is successful it is unlikely that the daring pioneer can appropriate the returns from the discovery. The predictable result is underinvestment, from the standpoint of society as a whole, in research and technology. The second example is regulation by the EU and Ireland of Irish water quality and the Irish dairy industry generally. This case illustrates the distinctive difficulties associated with mitigation of environmental externalities. Even when solutions can be developed in principle, it is difficult to estimate the costs of applying them, especially since, to be effective, general measures must be adapted to highly differentiated local circumstances. The familiar result is regulation that, for fear of imposing intolerable burdens on regulated parties, is often too timid to be effective, or when resolute, regulation that is ineffective for failure to take account of local particularity. Neither case is perfectly congruent to the “good jobs” challenge. But the success of both is due to the emergence of common mechanisms of governance under uncertainty that, we argue, can make the good jobs strategy workable and accountable.DARPA AND ARPA-E
In discussions of industrial policy, DARPA, created in 1958 in response to the Soviet launch of the Sputnik satellite, is often and usefully invoked as a reminder that the knowledge economy was not created solely by private actors—entrepreneurs, venture capitalists, and technologists—responding only to opportunities signaled by markets.
Far from being a mere bystander, the state, acting through DARPA and related agencies, played—and continues to play—a fundamental role in organizing the research from which are hewn the building blocks of the information economy. Among its iconic contributions are the computer network protocols underlying the Internet, precursors to the global positioning systems, and fundamental tools and devices for microprocessor design and fabrication. The accomplishments of DARPA have inspired a number of research agencies on similar lines, of which ARPA-E—a program created in the wake of the financial crisis to foster innovation in the energy sector—is both the most successful and the most faithful to the procedures of the original model.Recent studies of ARPA-E examine in detail the institutional mechanisms by which such public entities can orient, coordinate, and discipline collaborative investigation at the outer edge of technical possibility. If those mechanisms are today commonplace or rapidly becoming so, it is not because DARPA's methods are widely emulated but rather because more and more organizations, public and private, are adapting to the high-uncertainty environment, once exotic, that shaped DARPA from the first.
ARPA-E's overarching goal in establishing programs is to eliminate “white spaces” in the landscape of technical knowledge: missing capabilities, just beyond the frontier of current technical possibility, which, if mastered, would clear the way for advances in an important domain. A program might, for example, aim to support the investigation of novel battery concepts with the potential to reduce storage costs by enough to make an environmentally attractive class of electricity grid designs economically feasible. At every stage in the organization of research—the definition of programs of investigation, the selection of a portfolio of projects advancing the program purpose, and the supervision of individual projects in the portfolio—ARPA-E treats goals as provisional, or corrigible in the light of experience.
As with the contracts among innovating parties discussed above, precise goals are the result of a search, not fixed from the first.ARPA-E's program directors (PDs) play a key role in the collaborative setting and revision of goals. PDs are hired largely on the basis of their promise in giving direction to an emergent area of investigation. For instance, a candidate with a background in geology will be hired to create a program in advanced geothermal energy. Once program goals have been framed, the PD does a “deep dive.” PDs and ARPA-E technical staff supplement and correct their own background experience with reviews of the scientific literature, site visits to universities and companies, commissioned external studies, and consultation with Department of Energy (DOE) research managers. PDs then test the practicality of the emerging research area in technical workshops involving leading engineering, scientific, and commercial experts. If the research plan (adjusted to reflect the exchanges at the workshop) passes review, a project is formally created as a component of the developing program.
Proposals for research within the projects are developed and executed in the same manner, with goals open to recurrent challenge and revision. Applicants first submit a concept paper: a short document explaining why the proposal is superior to alternative approaches and how it responds to foreseeable technical and commercial risks. Proposals that survive a first round of external review are developed into full applications and reviewed again, with the difference that applicants may rebut criticism by external reviewers. The winners, designated “research partners” or “performers,” then negotiate project milestones with agency staff.
The execution of the project is subject, in the argot of ARPA-E, to “active project management,” a process with a strong family resemblance to the information-generating regime in contracting for innovation. Its most conspicuous feature is the quarterly progress report that research partners must provide for review by PDs and agency staff.3 Missed milestones can touch off an intensification of site visits, conference calls, meetings, and written analysis of problems and possible solutions.
When projects struggle, milestones can be reset to permit an alternative to the failed approach. Milestones are added or deleted in fully 45 percent of the projects, not counting substantive modifications, which are said to be frequent. If recovery efforts fail, the PD sends an “at risk” letter warning of the possibility of termination. In short, the agency rejects the model of hands-off, bet-on-the-person-not- the-project administration preferred by many established and successful research funders, public and private, in favor of the continuous, collaborative review and adjustment adopted in biotechnology, advanced manufacturing, and venture capital.ARPA-E is too new to permit any evaluation of its long-term impact. The energy industry—where even demonstration projects require substantial investment, innovators immediately confront legacy providers, and regulation is more likely to constrain innovation than, as in pharmaceuticals, accommodate it—changes so gradually that large transformations only slowly become visible. But the available evidence does strongly suggest that ARPA-E is indeed choosing projects in the zone of uncertainty—where the positive externalities of research and development will be especially large—and using its information-generating regime effectively to make the most of its choices.
Expert disagreement about what is possible is a good working definition of uncertainty.4 If ARPA-E funds uncertain projects, it should select projects whose prospects the best experts—its reviewers—disagree on. This is what we observe. There is a very slight correlation between reviewers' ratings of projects and the likelihood that they will be funded. Selection is not based on a consensus view of project prospects. Perhaps more tellingly, holding the rating constant, the agency picks the project where the range of reviewer rankings is the greatest—where judgments diverge the most. Plainly, the PDs and the selection committee are relying on other information—rebuttals, observation of the research in workshop dialogue with peers, and much else besides.
Project selection and governance, moreover, do not seem to favor either scientifically oriented projects doing basic research validated in journal publications or commercially oriented projects doing applied research validated by patents or market engagement. Compared to projects in other branches of the DOE doing either basic or applied research, ARPA-E projects have a higher rate of patenting and the same high rate of publishing. Most strikingly, they are more likely than the specialized projects to produce both a publication and patent (Goldstein and Narayanamurti 2018). A plausible interpretation is that they combine practical invention with scientific discovery on the model of use-inspired basic research made famous by Pasteur. As we will see next, commercial constraints and the penalty defaults imposed by EU environment law have made use-inspired research on similar lines central to the regulation of the water quality in Ireland and its dairy industry generally.
IRISH DAIRY FARMING
Regulation, and especially environmental regulation, differs from ARPA-E's contractual governance of research in two ways. First, agreements between the agency and award recipients are fully consensual (i.e., candidates compete for awards). Many addressees of regulation prefer no public constraints on their behavior; some even actively resist the imposition of rules. Penalty defaults therefore play an important role in inducing cooperation with the regulator, but none in the formation of award agreements. Second, ARPA-E faces the uncertainty that arises from manifest limits of our knowledge of science and technology: the “white spaces” mark the places where we do not know the laws of nature that apply to a particular problem. Environmental regulation encounters such frontier uncertainty as well, only it is often challenged instead or in addition by uncertainties arising from the singularities of place: the way known factors—familiar pollutant streams, types of subsoil and geology, for instance—combine in particular contexts to produce unforeseeable results.
“White spaces” get filled in once and for all. Once we learn the electrochemistry of cutting energy storage costs by a certain amount, that problem is solved. But environmental problems typically have to be redefined and addressed place by place: they are more often white dots rather than white spaces, and filling in one is of limited or no help in filling in an adjacent one. In this regard, environmental regulation strongly resembles and can serve as a partial model for regulation of the “good jobs” externality. In both cases, a central task of governance is creating an information-exchange regime that induces the local actors to cooperate to contextualize solutions while enabling them to benefit from the pooled experience of others, and vice versa.Within environmental regulation, nonpoint source pollution is the paradigmatic case of contextual uncertainty. The regular emissions of large polluters, such as power plants or sewage treatment facilities, are (relatively) easy to detect and control. Intermittent emissions from diffuse sources, such as the runoff from sporadic detergent use in scattered households, are not. Agricultural runoff is especially refractory because of the great variation in the pitch and absorptive capacity from field to field, the stark seasonal variations in weather and the rapid changes in the level and nature of productive activity induced by cycles of cultivation. We look to advances in the regulation of water pollution in agriculture to refine ideas about the governance of contextualization of the good jobs strategy, and to Ireland in particular, where pressures to reconcile demanding legal requirements to limit pollution with the needs of an expanding dairy industry have produced both an especially sharp understanding of the problem of contextual uncertainty and innovative reforms to address it.
The conviction that environmentally sustainable dairying could be a modern engine of growth came late to Ireland. Through much of the twentieth century, Irish dairy farming was dominated by extremely small holdings, with limited export opportunities and relatively low productivity and incomes. Membership in the European Economic Community (the predecessor of the EU) and its Common Agricultural Policy (CAP) together with imposition of EU milk quotas prompted consolidation, yielding a smaller but more efficient and capable cohort of specialized dairy farms that are still small— measured by farm acreage and herd size—in comparison to industrial producers. The Irish co-ops also consolidated and became first-tier suppliers of ingredients to global consumer food firms.5 Ireland—which accounts for less than 1 percent of global milk output (Eurostat 2017; FAO 2018, 5)—supplies
Building a GoodJobs Economy 77 almost 10 percent of the world's infant formula market6 and exports 90 percent of its dairy output.7
Grass is the source of the competitiveness of Irish dairy. The larger representative Irish dairy farm has the lowest cash cost-to-output ratio of the key international milk-producing regions, including the US, New Zealand, and Australia (Thorne et al. 2017, 70). Homegrown, grass-feed is much cheaper than purchased-feed concentrates; its price is relatively stable, sheltering Irish dairy farmers against a substantial risk. Cows that pasture on grass produce milk solids of superior quality; the grazing cow is, for watchful consumers in many parts of the world, the emblem of food production at its most natural.
For all these reasons the Irish dairy sector and its counterparts in various government departments have, since the turn of this century, come to see the national system of grass-based dairying on family farms as a model of production with a bright future and a central role in the overall development of the country—provided it can reconcile increasing efficiency with regulatory and consumer demands for environmental sustainability.8
EU law compelled Ireland to respond, haltingly and reluctantly, to agricultural pollution long before farmers, farm organizations, dairy co-ops, and the state extension service—Teagasc—became active advocates of sustainability. The Nitrates Directive of 1991 sets out precise concentration limits. Farms that fail to comply can be fined or disqualified from the EU single farm payment. Countries that fail to meet national limits must submit a plan for improvement to secure a temporary derogation of requirements or face potential draconian sanctions typical of penalty defaults.
The Water Framework Directive (WFD) of 2000 has, in contrast, extremely broad objectives: “good water” is defined for each type of water body (such as alpine streams or freshwater lakes) as minimal deviation from the chemical values and distribution and quantity of life forms associated with a pristine body of water of that type (Poikane et al. 2014). The basic unit of management is the river basin or catchment: the contiguous territory that drains into the sea at a single river mouth, estuary, or delta. Member states produce a six-year River Basin Management Plan (RBMP) for each basin using a collaborative process in which public officials, experts, and stakeholders specify objectives as well as procedures for translating them into concrete activities. Until 2027, counties9 that fall short can submit a new RBMP at the end of each planning cycle by asserting that the earlier approach proved technically infeasible or disproportionately expensive. Thereafter, as a penalty default, cost and feasibility will not excuse noncompliance.
Implementation of both directives has proved frustratingly difficult. Ad-
herence to “good practices” in agriculture has often failed to produce improvements in nitrate levels; effective, inclusive participation of local actors in the definition and continuing revision of the intentionally open-ended goals has been a major stumbling block in the application of the WFD. RBMPs, however made, have not achieved their objectives. Many member states will fail to meet the 2027 deadline. The directive was to be revised in 2019, but the revision postponed, among other reasons to reset the penalty default.10
In Ireland, compliance failures triggered a series of research programs to improve understanding and control of pollution flows at the catchment and field levels. These programs, linked with similar ones in other member states, have helped generate a web of institutions that is coming to function as an integrated system of local governance of water quality, greatly expanding public participation in environmental decision-making in the process.
In Teagasc's Agricultural Catchments Programme (ACP) of 2008, for example, six catchment areas, differing in soil types, geology, and types of farming, were selected to study the relations among farm management practices, flows of nutrients such as nitrogen and phosphorous, and the resulting changes in water quality. The ACP's key finding is that variations in soil and subsoil types and the underlying geology are in combination so influential in the absorption and drainage of nutrients that general rules of nutrient management, let alone plans based on them, will regularly fail. Poorly drained fields with environmentally innocuous phosphorus values may still pollute because of fast surface runoff while well-drained soils with alarming phosphorus concentrations may not pollute at all (Shortle and Jordan 2017, 17). The policy implication is that a nutrient management plan should be a starting point or provisional guide for joint investigation, by farmers and extension advisers, of environmental risks and how most economically to address them.
This kind of catchment program is part of a larger effort by the Environmental Protection Agency (EPA) and its partner institutions in water quality management to establish a cascading process of national, regional, and local consultation to select areas for intervention and to ensure full and effective participation of the affected local actors in the execution of projects that concern them. The selection process and new governance institutions come together in “local catchment assessments”: field-level examinations by the local actors themselves of the source of pollution in water bodies identified as intervention priorities. This assessment determines the local work plan, specifying and prioritizing projects. Agricultural problems detected by field assessments are referred to specialist sustainability advisers who assist the implicated farmers to improve their nutrient management practices,11 linking Contextualization of pollution mitigation measures on the farm to con- textualization of water management at the catchment or territorial level.
We draw three lessons for the design of the good jobs strategy from the Irish and EU experience with environmental regulation of contextual uncertainty. First, while framework legislation (the WFD) and penalty defaults orient and incentivize the creation of new governance instruments for local adaptation of general policies, making those institutions actually work requires continuing revision of initial plans in light of—frequently disappointing— experience. The recent flurry of institution building in Irish water regulation is the culmination of systematic investigation and hard experience, punctuated by false starts and half measures. There are principles of design for these institutions, but no blueprints.
Chief among these principles—the second lesson—is that contextualiza- tion in the sense of recognition of the need for local solutions to idiosyncratic local problems is a corrective and supplement to higher-level decisionmaking and procedures, but not a substitute for them. Local catchment assessments modify the specifications of targets identified by national and regional review and the order in which they are approached. Local authorities and stakeholders are not free to disregard the national list of priorities. Lower levels correct higher levels and vice versa. We can think of contextu- alization of this kind as a variant of the reciprocal review of collaborators we encountered in contracting for innovation.
Finally, contextualization blurs the distinction between regulation, directed to ensure compliance with rules—order maintenance within a given system—and the creation of new institutional systems. Contextualization induces collaboration between regulators, other public officials, and regulated entities in the development of novel forms of capacity building and public participation in regulatory decision-making. Irish dairy farmers in the catchment projects prepare their nutrient management plans with the support of specialist extension agents, who consult with catchment specialists; farmers with environmental problems collaborate with newly formed catchment assessment teams, connected in turn to a new corps of specialist sustainability advisers. Traditional extension agents propagate consolidated expertise. In codeveloping improvement plans with individual farmers and each other, these new specialists are reconsidering and revising current understandings as much as applying them. Collaborative investigation is necessary precisely because current rules and best practices run out, and establishing what should be done goes hand in hand with understanding and building the capacity needed to do it. When we speak, as we do next, of regulation in relation to the good jobs strategy we mean the term in this enlarged sense of fixing (and revising) requirements and inducing the creation of novel institutions, with all their further spillovers, that enable the addressees of regulation to meet them.