|
on Regulation |
| By: | Bergman, Aaron (Resources for the Future); Krupnick, Alan (Resources for the Future) |
| Abstract: | Clean hydrogen can be a key component to decarbonization, particularly in the industrial sector. Beyond its current use in chemicals and refining, hydrogen has potential new and expanded uses in, for instance, process heat, iron and steel, electricity generation and transportation. However, current hydrogen production technologies yield significant carbon emissions, and little economic incentive has existed to expand the use of hydrogen to new areas. But that has begun to change, with the US Congress recently placing large bets on a future hydrogen economy. Last year’s Infrastructure, Investment and Jobs Act (IIJA) contains $9.5 billion funding for hydrogen, including $8 billion for hydrogen hubs. And this year’s Inflation Reduction Act (IRA) contains two provisions that will subsidize clean hydrogen production. The first is a new tax credit (section 45V of the tax code) where the value of the credit is based on life cycle emissions. The second is a substantial increase in the value of the existing tax credit for carbon sequestration (section 45Q of the tax code), which is used to make “blue†hydrogen.Each of these tax credits can reduce the price difference between clean hydrogen and more carbon-intensive alternatives. To better understand the cost-effectiveness of these policies, this price difference can be converted into an implicit carbon price. As we will see, the values so obtained are significantly higher than many estimates of the social cost of carbon and may thus appear uneconomic. However, the goal of these tax credits is not solely to correct for the lack of a price on carbon but also to aid the deployment of nascent hydrogen technologies. Such deployment can have spillover effects, disseminating knowledge and potentially lowering costs in the future, an additional externality that, although difficult to quantify, may justify the higher implicit carbon prices.These tax credits have different impacts depending on the form of hydrogen production. Fossil-fuel based production generally uses natural gas (although it can use other fuels as we discuss later). This process produces greenhouse gas emissions from the carbon dioxide released as the hydrogen is extracted from the natural gas (or other hydrocarbon). To be clean hydrogen, these emissions must be captured. On the other hand, electrolysis, the production of hydrogen from water using electricity, produces no direct greenhouse gas emissions. However, electrolysis consumes large amounts of electricity that can lead to both high costs and high lifecycle emissions if the electricity is purchased on the wholesale market. Costs could be lowered, however, by using lower cost electricity, either through a direct connection to a generator or by only producing hydrogen when the price of electricity is low.Both forms of hydrogen production are potentially eligible for the 45V tax credit, but they must demonstrate low life cycle emissions to do so, with the magnitude of the credit depending on the level of emissions. In the case of fossil-fuel based production, beyond the direct emissions, the largest component of the life cycle emissions is upstream methane leakage; for electrolysis, it is the emissions associated with electricity production. The Treasury Department will have to issue a regulation on how to calculate these life cycle emissions, which will have a major impact on the subsidies available to hydrogen producers and the competitiveness of various forms of hydrogen production.In contrast to the 45V tax credit, only hydrogen producers using carbon capture, utilization and storage (CCUS) are eligible for the 45Q tax credit. This tax credit is available irrespective of the life cycle emissions and, as we will see, can be more valuable than the 45V tax credit. Producers are not allowed to take both tax credits.We will analyze the impacts of these tax credits on the costs of hydrogen production using a set of hydrogen production models from the National Renewable Energy Laboratory (NREL). We will see that the 45Q tax credit is sufficient to make some forms of fossil fuel–based hydrogen production competitive with current high-emission production on a levelized cost basis. The high cost of grid electricity and the associated emissions, on the other hand, make it hard for electrolysis to compete. However, electrolyzers that source cleaner and cheaper electricity can qualify for high levels of the 45V tax credit and compete with fossil-fuel based hydrogen production. In the long run, costs for electrolyzers are expected to decrease, and the grid should be less carbon intensive, making electrolyzers competitive more broadly.In this report, we will review various forms of hydrogen production and the changes to the tax law made by the Inflation Reduction Act. We will calculate the implicit carbon prices and discuss the calculation of life cycle emissions. Next, using the NREL models, we will discuss the impacts of the tax credits on the levelized and marginal costs of hydrogen production and see how they depend on upstream methane leakage rates and the carbon intensity of electricity production. We will also examine in detail how the relative competitiveness of the various forms of production depends on natural gas and electricity prices. We conclude with a discussion of the broader hydrogen economy. |
| Date: | 2022–11–09 |
| URL: | http://d.repec.org/n?u=RePEc:rff:dpaper:dp-22-13&r=reg |
| By: | Axel Gautier (HEC Liège); Julien Jacqmin (NEOMA - Neoma Business School); Jean-Christophe Poudou (MRE - Montpellier Recherche en Economie - UM - Université de Montpellier) |
| Abstract: | Renewable energy communities involve various agents who decide to jointly invest in renewable production units and storage. This paper examines how these communities interact with the energy system and can decrease its overall cost. First, we show that a renewable energy community can contribute positively to welfare if the electricity produced by the investment is consumed close to its place of production, i.e. if the community has a high degree of self-consumption. Second, our analysis identifies the condition on prices and grid tariffs to align the community's interest with welfare maximization. We also show that some of these grid tariffs do not have a negative impact on non-members of the community and could therefore limit potential distributional issues. Third, we argue that various internal organization of the renewable energy communities are feasible. The internal organization impacts the distribution of benefits among members but not the global efficiency of the community. |
| Keywords: | energy communities, decentralized production unites, energy transition, grid regulation |
| Date: | 2023–02–08 |
| URL: | http://d.repec.org/n?u=RePEc:hal:wpaper:hal-04032253&r=reg |
| By: | Dirk Bergemann; Alessandro Bonatti |
| Abstract: | We analyze digital markets where a monopolist platform uses data to match multiproduct sellers with heterogeneous consumers who can purchase both on and off the platform. The platform sells targeted ads to sellers that recommend their products to consumers and reveals information to consumers about their values. The revenue-optimal mechanism is a managed advertising campaign that matches products and preferences efficiently. In equilibrium, sellers offer higher qualities at lower unit prices on than off the platform. Privacy-respecting data-governance rules such as organic search results or federated learning can lead to welfare gains for consumers. |
| Date: | 2023–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2304.07653&r=reg |
| By: | Markus Brueckner; Haidi Hong; Joaquin Vespignani |
| Abstract: | This paper presents estimates of the effects that government regulation of diesel and petrol prices has on GDP growth. Theory suggests that when supply curves are convex, a decrease in the regulatory price has a larger effect on output than a tantamount increase. Motivated by this theoretical insight, we specify VAR models with asymmetric effects of positive and negative changes in the regulatory prices of diesel and petrol. We estimate the VAR models on quarterly data from China’s national accounts during the period Q1 1998 to Q4 2018. Our main findings are that: (i) negative growth rates of regulatory diesel and petrol prices significantly reduce GDP growth; (ii) positive growth rates of regulatory diesel and petrol prices have a positive, but quantitatively small and statistically insignificant effect on GDP growth. |
| Keywords: | GDP growth; energy price regulation |
| JEL: | E60 |
| Date: | 2023–04 |
| URL: | http://d.repec.org/n?u=RePEc:acb:cbeeco:2023-690&r=reg |
| By: | Vivek Bhattacharya; Gastón Illanes; David Stillerman |
| Abstract: | We document the effects of a comprehensive set of US retail mergers. On average, prices increase by 1.5% and quantities decrease by 2.3%, with significant heterogeneity in outcomes across mergers. Price changes correlate with the screens codified in the Horizontal Merger Guidelines. Through a model of enforcement, we find that agencies challenge mergers they expect would increase average prices more than 8–9%. Modest increases in stringency reduce prices and the prevalence of approved anti-competitive mergers, with minimal impacts on blocked pro-competitive mergers, at a significantly greater agency burden. Our findings inform the debate over whether antitrust enforcement has been lax. |
| JEL: | D43 K21 L13 L41 |
| Date: | 2023–04 |
| URL: | http://d.repec.org/n?u=RePEc:nbr:nberwo:31123&r=reg |
| By: | Brown, David B.; Muehlenbachs, Lucija (Resources for the Future) |
| Abstract: | To avoid electric-infrastructure-induced wildfires, millions of Californians have had their power cut for hours to days at a time. We show that rooftop solar-plus-battery-storage systems increased in zip codes with the longest power outages. Rooftop solar panels alone will not help a household avert outages, but a solar-plus-battery-storage system will. Using this fact, we obtain a revealed-preference estimate of the willingness to pay for electricity reliability, the Value of Lost Load, a key parameter for electricity market design. Our estimate, of around $4, 300/MWh, suggests California's wildfires-prevention outages resulted in losses from foregone consumption of $322 million to residential electricity consumers. |
| Date: | 2023–04–18 |
| URL: | http://d.repec.org/n?u=RePEc:rff:dpaper:dp-23-10&r=reg |
| By: | Dirk Bergemann; Alessandro Bonatti; Nicholas Wu |
| Abstract: | We develop an auction model for digital advertising. A monopoly platform has access to data on the value of the match between advertisers and consumers. The platform support bidding with additional information and increase the feasible surplus for on-platform matches. Advertisers jointly determine their pricing strategy both on and off the platform, as well as their bidding for digital advertising on the platform. We compare a data-augmented second-price auction and a managed campaign mechanism. In the data-augmented auction, the bids by the advertisers are informed by the data of the platform regarding the value of the match. This results in a socially efficient allocation on the platform, but the advertisers increase their product prices off the platform to be more competitive on the platform. In consequence, the allocation off the platform is inefficient due to excessively high product prices. The managed campaign mechanism allows advertisers to submit budgets that are then transformed into matches and prices through an autobidding algorithm. Compared to the data-augmented second-price auction, the optimal managed campaign mechanism increases the revenue of the digital platform. The product prices off the platform increase and the consumer surplus decreases. |
| Date: | 2023–04 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2304.08432&r=reg |
| By: | Foroogh Nazzari Chamaki (Department of Banking and Finance, Eastern Mediterranean University); Glenn Jenkins (Queen's University); Majid Hashemipour (Faculty of Engineering, Cyprus International University, North Cyprus) |
| Abstract: | One of the most effective strategies to mitigate water shortages worldwide is to reuse the treated wastewater for freshwater production employing reverse osmosis (RO) technology. This strategy is appropriate in urban areas of arid or semi-arid regions as it can provide a sustainable and reliable water source close to the consumers. One of the drawbacks of RO is the high variability of production costs due to the electricity intensity. In addition, depending on the electricity source, it can also result in substantial environmental costs.This study showed that upgrading pumping and RO membrane systems of a wastewater reuse plant in Cyprus can significantly alleviate these drawbacks in terms cost, water recovery rate, and air pollution. The water recovery rate of the upgraded RO plant increased from 43.2 to 75 percent which results in a substantial net financial benefit due to less quantity of wastewater to be purchased and more potable water to be produced. The upgraded system also reduced the electricity requirement from 3.63 kWh/m3 to 1.92 kWh/m3. Pollution emissions decreased substantially because of the reduction in electricity requirements. The beneficiaries of these lower emissions costs are the residents of Cyprus and global society. Overall, the benefit of upgrading the plant is highly attractive with more than 65 percent of annual real internal rates of re-turn in financial and economic terms. Positive net present values are realized for all the scenarios considered. |
| Keywords: | circular economy, reused wastewater, reverse osmosis, levelized cost, economic cost, membrane technologies, emission cost, environmental externalities, distributive analysis, energy saving |
| JEL: | I38 L95 H43 Q25 |
| Date: | 2023–04 |
| URL: | http://d.repec.org/n?u=RePEc:qed:wpaper:1503&r=reg |
| By: | Thierry Blayac (CEE-M, Univ Montpellier, CNRS, INRAE, Institut Agro, Montpellier, France); Patrice Bougette (Université Côte d'Azur; GREDEG, CNRS, France) |
| Abstract: | This study estimates the competitive effects of horizontal mergers in the French long-distance bus industry. We examine the two mergers that followed the 2015 Deregulation Act (the Macron Law); we use an exclusive and exhaustive dataset that covers eight consecutive quarters. We analyze the merger effects by comparing bus links that were affected by mergers with those that were unaffected; we use difference-in-differences estimations. We find that the two mergers are associated with price increases of about 13.5% immediately that then moderate to 5.3%; and with the frequency decreases from -21.5% to -25.7%; we observe no effects on load factors. These findings show evidence of short-run anticompetitive effects, while the mergers under study were not scrutinized by the French competition agency, as they were below the notification thresholds. |
| Keywords: | Long-distance bus industry, Mergers and acquisitions, Deregulated industry, Consolidation, Intramodal competition, Difference-in-differences estimation |
| JEL: | K21 L12 L40 L42 |
| Date: | 2022–09 |
| URL: | http://d.repec.org/n?u=RePEc:gre:wpaper:2022-34&r=reg |
| By: | Shubhdeep Deb (UPF Barcelona); Jan Eeckhout (Institute for Fiscal Studies); Aseem Patel (University of Essex); Lawrence Warren (US Census Bureau) |
| Date: | 2022–09–26 |
| URL: | http://d.repec.org/n?u=RePEc:ifs:ifsewp:22/40&r=reg |
| By: | Merino Troncoso, Carlos |
| Abstract: | The book is intended to be a reference book of Competition Economics for economists, consultants and/or practitioners. It is a modern review of demand and supply estimation, market structure, merger analysis, damage estimation, welfare loss, abuse of dominance, network effects, and a math and statistics review. Faced with potential multibillion fines, and thousands of damage claims firms are hiring and paying high fees to comply, defend or claim in antitrust cases. Complex economic and statistical issues appear in most cases and all the parties involved in cases are expected to have a good knowledge of them. This book tries to cover a demand of practitioners for a compact introductory level book on this field. |
| Keywords: | antitrust, competition policy, merger simulation, demand estimation |
| JEL: | L10 L11 L13 L16 L4 L40 |
| Date: | 2023–01–15 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:115999&r=reg |
| By: | Maurizio Trapanese (Banca d'Italia); Michele Lanotte (Banca d'Italia) |
| Abstract: | Technological progress in finance has been accelerating over the last decade. In the future, it is likely that financial intermediaries may undergo significant challenges as regards their traditional business model and functions, since an increasing share of payments may be settled without banks’ deposits and capital markets may increasingly provide direct credit to the economy. This paper aims to outline the theoretical and regulatory implications stemming from digital financial markets, with a particular focus on the growing importance of BigTech and FinTech firms. We study the importance of information and communication in financial intermediation, and outline the impact of technological progress on the core functions traditionally performed by banks and other financial institutions, and on payment systems. In this context, we discuss the role of public policies, and the main issues for regulation, supervision, competition, and consumer protection. |
| Keywords: | firm behaviour, international financial markets, financial institutions, financial policy and regulation, risk management JEL Classification: D21, G15, G20, G28, G32 |
| Date: | 2023–04 |
| URL: | http://d.repec.org/n?u=RePEc:bdi:opques:qef_758_23&r=reg |
| By: | Panova, Elena; Garrett, Daniel F. |
| Abstract: | Large-scale infrastructure investments are often carried out in set- tings where their eventual usefulness or importance is diffi cult to pre- dict. This paper studies optimal incentives for investment when the agent undertaking the investment has superior information on two dimensions: the cost of investment and the likelihood it is useful or beneficial to the principal. Usefulness eventually becomes public, but punishments are limited as the regulator aims at ensuring the agent earns non-negative profits in each period. We characterize the opti- mal incentive scheme and show it involves either: (i) investments by the agent even though he knows they are useless and rents to only cost-effi cient types, or (ii) rents to all types. The possibility that rent is left to all types contrasts with the usual prediction in static (and also dynamic) mechanism design and arises though the agent's preferences are stable over time. |
| Keywords: | Monopoly regulation; Real options; Multidimensionl asymmetric information |
| JEL: | D81 D82 L51 |
| Date: | 2023–04–26 |
| URL: | http://d.repec.org/n?u=RePEc:tse:wpaper:128054&r=reg |
| By: | Simon Cowan |
| Abstract: | The conditions for monopolization to be good for social welfare are examined. Social welfare can be higher when a monopoly sells to a monopoly, with double margins, than when a competitive industry sells to a downstream Cournot oligopoly with differing efficiency levels. This requires inverse demand to be sufficiently concave, and cannot hold when demand is convex. When there are no vertical issues an efficient monopoly can yield higher social welfare than an asymmetric Cournot duopoly as long as demand is logconcave. In general greater demand concavity increases the relative importance of the benefit of redistributing output to the efficient firm. |
| Date: | 2023–03–31 |
| URL: | http://d.repec.org/n?u=RePEc:oxf:wpaper:1006&r=reg |
| By: | Ginger Zhe Jin; Mario Leccese; Liad Wagman |
| Abstract: | We examine how public firms listed in North American stock exchanges acquire technology companies during 2010-2020. Combining data from S&P, Refinitiv, Compustat, and CRSP, and utilizing a unique S&P taxonomy that classifies tech M&As by tech categories and business verticals, we show that 13.1% of public firms engage in any tech M&A in the S&P data, while only 6.75% of public firms make any (tech or non-tech) M&A in Refinitiv. In both datasets, the acquisitions are widespread across sectors of the economy, but tech acquirers in the S&P data are on average younger, more investment efficient, and more likely to engage in international acquisitions than general acquirers in Refinitiv. Within the S&P data, deals in each M&A-active tech category tend to be led by acquirers from a specific sector; the majority of target companies in tech M&As fall outside the acquirer’s core area of business; and firms are, in part, driven to acquire tech companies because they face increased competition in their core areas. |
| JEL: | D04 D22 L1 |
| Date: | 2023–04 |
| URL: | http://d.repec.org/n?u=RePEc:nbr:nberwo:31126&r=reg |
| By: | Gebrezgabher, Solomie (International Water Management Institute); Kodua, T.; Mateo-Sagasta, Javier (International Water Management Institute) |
| Keywords: | Water reuse; Projects; Cost recovery; Economic analysis; Cost benefit analysis; Wastewater treatment plants; Agriculture; Landscaping; Investment; Potable water; Prices; Aquifers; Groundwater recharge |
| Date: | 2022 |
| URL: | http://d.repec.org/n?u=RePEc:iwt:bosers:h051739&r=reg |
| By: | Tito Boeri; Andrea Garnero; Lorenzo G. Luisetto |
| Abstract: | Non-compete clauses (NCCs) limiting the mobility of workers have been found to be rather widespread in the US, a flexible labour market with large turnover rates and a limited coverage of collective bargaining. This paper explores the presence of such arrangements in a rigid labour market, with strict employment protection regulations by OECD standards and where all employees are, at least on paper, subject to collective bargaining. Based on a representative survey of employees in the private sector, an exam of collective agreements and case law, we find that in Italy i) collective agreements play no role in regulating the use of NCCs while the law specifies only the formal requirements, ii) about 16% of private sector employees are currently bound by a NCC, iii) NCCs are relatively frequent among low educated employees in manual and elementary low paid occupations having no access to any type of confidential information, and iv) in addition to NCCs, a number of other arrangements limit the post-employment activity of workers. Many of the NCCs do not comply with the minimum requirements established by law and yet workers do not consider them as unenforceable and appear to behave as they were effective. Even when NCCs are unenforceable they appear to negatively affect wages when they are introduced without changing the tasks of the workers involved. Normative implications are discussed in the last section of the paper. |
| Keywords: | non-compete clauses, monopsony, labour market concentration, employment, wages |
| Date: | 2023–04–03 |
| URL: | http://d.repec.org/n?u=RePEc:cep:cepdps:dp1912&r=reg |
| By: | - |
| Abstract: | En este documento se analiza la caracterización de la pobreza energética en el Uruguay, un problema multidimensional —social, económico y cultural— y estructural que afecta directamente a los hogares y su entorno. Se ha definido, más específicamente, como la imposibilidad de garantizar el derecho de las personas a satisfacer sus necesidades energéticas mediante servicios seguros, asequibles y de calidad. En los últimos años, el país ha presentado indicadores positivos de acceso y abastecimiento, así como una elevada eficiencia del sector energético. Además, registra altos rendimientos macro- y microeconómicos respecto de la distribución del ingreso y el acceso al financiamiento. Pese a ello, el Estado uruguayo ha establecido que, dependiendo de la metodología y las variables de medición utilizadas, entre el 28% y el 38% de los hogares (principalmente pobres, monoparentales, con jefas de hogar, de tramos etarios jóvenes y afrodescendientes) están bajo la línea de pobreza energética, especialmente respecto del acceso a fuentes energéticas seguras, artefactos modernos y calidad de la vivienda. Un segundo objetivo de este trabajo, por lo tanto, es identificar políticas públicas que promuevan el bienestar de los hogares mediante el acceso equitativo a servicios energéticos de calidad. |
| Keywords: | RECURSOS ENERGETICOS, POLITICA ENERGETICA, PLANIFICACION DE LA ENERGIA, SERVICIOS ENERGETICOS, ENERGIA ELECTRICA, ABASTECIMIENTO DE AGUA, VIVIENDA, ESTADISTICAS DE ENERGIA, ENERGY RESOURCES, ENERGY POLICY, ENERGY PLANNING, ENERGY SERVICES, ELECTRIC POWER, WATER SUPPLY, HOUSING, ENERGY STATISTICS |
| Date: | 2023–03–22 |
| URL: | http://d.repec.org/n?u=RePEc:ecr:col022:48767&r=reg |
| By: | Giuseppe Attanasi (Université Côte d'Azur, France; GREDEG CNRS); Michela Chessa (Université Côte d'Azur, France; GREDEG CNRS); Sara Gil Gallen (Università degli studi di Bari "Aldo Moro", Italy); Elena Manzoni (University of Bergamo) |
| Abstract: | We investigate theoretically and experimentally the performance of a bargaining-overstrategies protocol with confirmed proposals, with either symmetric (i.e., alternating among the two players) or asymmetric power of confirmation (i.e., assigned to one of the two players alone). We apply it both to a Bertrand duopoly market, in which competition is on prices, and to a Cournot duopoly market, in which players compete on the amount of output they produce. We characterize the set of subgame perfect equilibrium outcomes of the bargaining game with confirmed proposals for each of the four combinations of confirmation power and market game, and formulate theory-driven hypotheses based on these different characterizations. Our experimental results show that bargaining over strategies of price- or quantity-setting (i) acts as a communication device in competitive environments, (ii) increases the level of collusion, and (iii) reduces the bargaining length. In particular, we report experimental evidence of overall better performance of a Bertrand duopoly market in reaching an equitable, welfare-maximizing, and Pareto-efficient agreement. However, competing in price rather than in quantity setting reduces the bargaining length only under asymmetric power of confirmation, while under symmetric power price-setting only has a second-order effect on reducing the bargaining length. We complement the data analysis with a qualitative analysis of the sequence of players' declared strategies as communication devices. |
| Keywords: | Bargaining, Tacit collusion, Experiments, Bertrand duopoly, Cournot duopoly |
| JEL: | C72 C91 |
| Date: | 2022–07 |
| URL: | http://d.repec.org/n?u=RePEc:gre:wpaper:2022-18&r=reg |
| By: | Schmitz, Patrick W. |
| Abstract: | In the property rights approach to the theory of the firm, ownership matters if parties have to make partly relationship-specific investments, but ownership would be irrelevant if the investments were completely relationship-specific. We show that if negotiations after the investment stage require transaction costs to be paid, then ownership matters even when investments are completely relationship-specific. While in the standard model without transaction costs there are underinvestments compared to the first-best benchmark, in our setting a party may overinvest in order to induce the other party to incur the transaction costs that are necessary to enter the negotiation stage. |
| Keywords: | incomplete contracts; investment incentives; ownership rights; relationship specificity; transaction costs |
| JEL: | D23 D86 |
| Date: | 2023 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:117065&r=reg |
| By: | Daniel McFadden |
| Abstract: | The elegant economic picture of rational consumers achieving Pareto optimality through trade in decentralized self-organized markets is blurred by market imperfections and choices inconsistent with consumer self-interest. Behavioral economics has documented these errors in choice, and considered interventions that can mitigate harm to individuals from bad choices. This essay considers more broadly market distortions that arise when sellers design and market products to exploit consumer errors, and suggests market management policies to mitigate these distortions. I give examples in which consumers’ lack of attention and diligence in collecting, filtering, and processing information has an outsize effect on market outcomes. I discuss identification and estimation of a two-stage model of decision-making, with attention determined in the first stage, and choice among alternatives described in the second stage. The stages in this model are linked by unobservable factors and identified by available information specific to each. I use this model to quantify the effects of inattention in a market for health insurance. |
| JEL: | D01 D04 D15 D18 D47 D83 D91 L14 L15 |
| Date: | 2023–04 |
| URL: | http://d.repec.org/n?u=RePEc:nbr:nberwo:31165&r=reg |