WPS8052 Policy Research Working Paper 8052 Industrial Policies vs Public Goods under Asymmetric Information Constantino Hevia Norman Loayza Claudia Meza-Cuadra Development Research Group Macroeconomics and Growth Team May 2017 Policy Research Working Paper 8052 Abstract This paper presents an analytical framework that captures the restricting information on firm productivity to be private informational problems and trade-offs that policy makers to the firm. The paper develops an optimal contract (which face when choosing between public goods (for example, replicates the first best), consisting of a tax-based mechanism infrastructure) and industrial policies (for example, firm- or that induces firms to reveal their true productivity. As this sector-specific subsidies). After a discussion of the literature, contract requires high government capacity, other, simpler the paper sets up the model economy, consisting of a govern- policies are considered. The paper concludes that providing ment and a set of heterogeneous firms. It then presents the public goods is likely to dominate industrial policies under first-best allocation (under full information) of government most scenarios, especially when government capacity is low. resources among firms. Next, uncertainty is introduced by This paper is a product of the Macroeconomics and Growth Team, Development Research Group. It is part of a larger effort by the World Bank to provide open access to its research and make a contribution to development policy discussions around the world. Policy Research Working Papers are also posted on the Web at http://econ.worldbank.org. The authors may be contacted at nloayza@worldbank.org. The Policy Research Working Paper Series disseminates the findings of work in progress to encourage the exchange of ideas about development issues. An objective of the series is to get the findings out quickly, even if the presentations are less than fully polished. The papers carry the names of the authors and should be cited accordingly. The findings, interpretations, and conclusions expressed in this paper are entirely those of the authors. They do not necessarily represent the views of the International Bank for Reconstruction and Development/World Bank and its affiliated organizations, or those of the Executive Directors of the World Bank or the governments they represent. Produced by the Research Support Team Industrial Policies vs Public Goods under Asymmetric Information Constantino Hevia Norman Loayza Claudia Meza-Cuadra Universidad Torcuato di Tella The World Bank The World Bank Keywords: Industrial Policy; Public Goods; Uncertainty; Private Information; Firm Subsi- dies and Taxes JEL Classi…cation: H2, H4, O1, O2 We thank Aart Kraay, Claudio Raddatz, and Luis Serven for insightful comments and discussion. We gratefully acknowledge funding from the Knowledge for Change Program (KCP) of the World Bank. The …ndings and conclusions expressed in this paper are entirely those of the authors, and do not necessarily represent the views of the institutions to which they are a¢ liated. 1 Introduction In the last decade there has been a revival of interest in industrial policies among policy makers around the world (Warwick and Nolan, 2014; Stiglitz, Lin, and Monga, 2013; Pellegrin et al., 2015). This resurgence of interest strengthened particularly in the aftermath of the economic crisis of 2008–2009, as governments looked for ways to increase their economies’productivity in the context of severely constrained …nance (Warwick, 2013). However, since their heydays from the 1940s to 1960s, industrial policies have been the subject of heated discussion and debate. The main theoretical justi…cation for the use of industrial policy is the need to address market imperfections. In an environment with full information and strong governance, optimal design of industrial policy is in principle a simple matter. Policy makers should eliminate rela- tive distortions across sectors and resolve or take advantage of the externalities and spillovers that some sectors could have relative to others. An optimal policy would then equalize the social marginal value of allocating resources across sectors. In practice, however, the pub- lic sector faces two key issues that hinder the implementation of industrial policy (Rodrik, 2004): its imperfect knowledge of existing constraints, incentives, and opportunities across the economy; and its vulnerability to corruption, manipulation, and rent seeking. Developing countries, which tend to have weaker institutions and lower capacity to implement complex policies, may thus face greater risks when pursuing industrial policies. The aim of this paper is to present an analytical framework that captures the informational problems and trade-o¤s that policy makers face when choosing either public goods (e.g., public information, infrastructure, and law and order) or industrial policies (e.g., …rm or sector- speci…c subsidies, grants, and tax breaks). The model attempts to capture the possibility that private entrepreneurs may have incentives to misrepresent information about the social value of their …rms or industry in order to obtain special treatment from the government. We explore an optimal industrial and tax policy that is robust to uncertainty about …rm-speci…c productivity. It requires, however, substantial government capacity, as the planner must be able to set …rm-speci…c taxes that are a function of …rms’claimed productivity. Through this tax system, the government induces …rms to reveal their true productivity, thus being able to implement the …rst best allocation despite asymmetric information. Finally, the model explores less optimal but simpler policies, more appropriate when the planner does not have the ability to set up an elaborate tax and compliance system. In this, possibly more realistic context, the model …nds that providing public goods tends to be preferable to industry or …rm-speci…c industrial policies. To motivate the model, the next section presents a brief overview of experiences and issues in the implementation of industrial policies. We then present and solve the model. We do it under full information to serve as a benchmark. We then solve it under private information, where we examine optimal policies. 2 A brief overview of selected industrial policy experiences Industrial policies consist of selective government interventions to promote certain economic sectors with the aim of increasing their productivity and spread positive externalities through- out the economy (Pack, 2000; Aiginger and Sieber, 2006; Weiss, 2013). Industrial policies can vary in a range that goes from “vertical” policies that favor speci…c …rms or narrow sectors 2 to “horizontal” policies that target broad sectors by improving their business environment (Rodrik, 2008, and Warwick, 2013). The more horizontal these policies are, the more they approach public goods. Countries around the world have implemented industrial policies with varying degrees of success. An analysis of Chilean industrial policy, for instance, describes the use of several horizontal and vertical policy instruments, though with a growing emphasis on the latter in recent years (Inter-American Development Bank, 2014). Horizontal industrial policies used in Chile include guarantees for loans to small enterprises, subsidies to new ex- ports, and a program to foster innovation; while vertical industrial policies feature the creation of a semi-public entrepreneurial institution (Fundacion Chile) and a program to attract FDI in technology. As is often the case with vertical industrial policies, Fundacion Chile has had many failed projects, for instance the cultivation of the southern hake, but also a few huge successes, including the development of the salmon and blueberry industries (Inter-American Development Bank, 2014). Choosing between industrial policy instruments is often complicated due to uncertainty and the existence of information asymmetries between the public and private sectors. In order to manage these challenges, Rodrik (2004) proposes that industrial policy be viewed as a discovery process, whereby public and private sector collaborate to identify underlying costs and opportunities. In this vein, Fernández-Arias et al. (2016) describe several instances of successful collaborations across Latin America, including in the sugarcane industry in Argentina, the tourism industry in Costa Rica, and shipbuilding programs in Uruguay. This type of public-private collaborative approach, however, is hampered by the risk that the private sector might exploit its informational advantage to derive unproductive rents from industrial policies through capture. IADB (2014) illustrates this risk using the case of the rice industry in Costa Rica. It describes how private involvement in the institution in charge of managing policies for the rice sector have resulted in excessive support to rice producers and a decrease in agricultural productivity. An analysis of industrial policies in the Middle East and North Africa region also highlights the risk of state capture in the use of both vertical and horizontal instruments (Jaud and Freund, 2015). The report notes that, in Tunisia, for example, …rms highly connected to former president Ben Ali were found to be most present in protected sectors including telecoms, automobiles, and tourism. Given the risks and costs associated with the implementation of industrial policies, poli- cymakers, particularly in developing countries, should consider the best match between their capacity and the type of policy to be implemented (Chang, 2011; IADB, 2014). Vertical policies require a greater capability to control capture by the private sector, and thus higher administrative costs, than do horizontal policies. For example, while tax incentives have been widely used to attract new investment and spur economic growth, including in Singapore and Korea, the cost of implementing and enforcing these policies can be particularly high (Tanzi and Shome, 1992). Further, these implementation costs generally increase with the complexity of subsidies and taxes involved in industrial policies, especially under low gov- ernment capacity (Chang, 2011). A committee reviewing the use of tax incentive policies in Papua New Guinea, for instance, highlighted concern about the challenges of implementing or e¤ectively monitoring R&D and infrastructure incentives in the face of scarce administrative or technical capacity (PNG Tax Committee, 2014; Chang, 2011). 3 3 A model of industrial policy under private information The objective of our model is to capture, in a simple way, the trade-o¤s that a benevolent planner faces when deciding whether to provide industry- or …rm-speci…c subsidies or a public good (i.e. infrastructure) to maximize total income in a context of private information. The main assumptions are that the government does not observe the productivity of the …rms and that there are …nancial constraints that prevent …rms from increasing their sizes by borrowing or issuing equity in …nancial markets. To simplify the exposition, we consider a static model with two …rms— we can think of two industries or two …rms within the same industry— whose productivities are private information and that are constrained in their initial capital k0 . Productivity can take on two values, high or low, represented by zH > zL > 1, respectively. Let H = Pr (zj = zH ) and L =1 H denote the probabilities that …rm j = 1; 2 draws a high and a low productivity type, respectively. The assumption zL > 1 ensures that, when there is perfect information, the planner prefers subsidies or investing in the public good to running a …scal surplus. We normalize the marginal cost of production to zero and the goods prices to one. Im- portantly, …rms choose their size, and are constrained in their initial capital k0 , which we assume, without loss of generality, to be the same for both …rms.1 That is, if kj is the size of …rm j; the …rm can produce zj kj goods at a marginal and total cost of zero. Crucially, we assume that …rms cannot use …nancial markets to increase their sizes, either because they are underdeveloped, or due to some other …nancial frictions like complete lack of commitment to repay their debts. Yet, the government can provide a subsidy so that …rms can increase their sizes (vertical industrial policy) or a public good that increases simultaneously the productivity of all …rms (horizontal industrial policy). In particular, the government has a budget of T which can be allocated to provide a public good, denoted by g , or to provide a subsidy to …rm j = 1; 2, denoted by sj . The government budget constraint is thus g + s1 + s2 T; and we denote the government surplus by d=T g s1 s2 0: The value of output of …rm j = 1; 2 that receives a subsidy sj and has a productivity draw of zj is given by vj = zj (k0 + sj + g ) : We assume that < 1, which means that the public good could increase the productivity of each …rm, but by less than a direct subsidy. While we consider a static model, a single period is composed of di¤erent sub-periods. The timing of events within the period is as follows: 1. Nature draws the …rms’productivities. Firms observe their productivities but the gov- ernment does not. 1 The assumption is without loss of generality because the technology is linear in the stock of capital. 4 2. Firms report their productivities to the government (as argued below, by the Revelation Principle this assumption is without loss of generality). 3. Contingent on the …rms’reports, the government provides the public good and subsidies to the …rms. 4. Firms produce. 3.1 First best allocation We begin by considering the …rst best allocation assuming that the government is able to observe the productivities of both …rms. The objective of the government is to maximize the total value of output plus the government surplus, WFB = max z1 (k0 + s1 + g ) + z2 (k0 + s2 + g ) + d s1 ;s2 ;f1 ;f2 ;g;d subject to d = T (g + s1 + s2 ) d 0; g 0; sj 0 for j = 1; 2: Let the government policy be a vector G = (s1 ; s2 ; g ) ; which includes the subsidies and the provision of the public good.2 To solve this problem we …rst note that, in the …rst best solution, the government surplus d is zero. Since zL > 1, the marginal bene…t of allocating a dollar to a subsidy is always greater than the marginal bene…t of keeping that dollar to increase the government surplus FB @W F B ( @W @sj = zj > 1 = @d for j = 1; 2). Thus, d > 0 cannot be optimal. Therefore, the …rst best problem is reduced to W F B = max z1 (k0 + s1 + g ) + z2 (k0 + s1 + g ) s1 ;s2 ;g subject to T = g + s1 + s2 ; s1 0; s2 0; and g 0: Since this is a linear programming problem, the solution is at a vertex of the feasible set. The policies to consider are the following, G = (s1 ; s2 ; g ) = (T; 0; 0) G = (s1 ; s2 ; g ) = (0; T; 0) G = (s1 ; s2 ; g ) = (0; 0; T ) : 2 We assume that the planner cannot transfer capital from one …rm to the other. If such a policy were feasible, the planner would expropriate all capital from a low productivity …rm (either directly or through taxes) and give it to a high productivity …rm. We do not allow for such expropriatory policies. The government may tax …rms, but those taxes cannot be used to transfer resources across …rms. In this case, taxes are a transfer from the …rms to the government and do not a¤ect aggregate welfare. Therefore, we set those taxes to zero. 5 Let z = (z1 ; z2 ) denote the vector of realized productivities. We have the following cases to consider: 1. Suppose that z = (zH ; zL ) : If G = (T; 0; 0) ) W = zH (k0 + T ) + zL k0 = (zH + zL ) k0 + zH T: If G = (0; T; 0) ) W = zH k0 + zL (k0 + T ) = (zH + zL ) k0 + zL T: If G = (0; 0; T ) ) W = zH (k0 + T ) + zL (k0 + T ) = (zH + zL ) k0 + (zH + zL ) T: To make the problem interesting, we assume that zH < (1) zH + zL for otherwise the optimal subsidy is zero and it is always optimal to provide the public good. With this assumption, the optimal policy is GF B (zH ; zL ) = (T; 0; 0) : 2. Suppose that z = (zL ; zH ) : This is the symmetric case, therefore, GF B (zL ; zH ) = (0; T; 0) : 3. Suppose that z = (zj ; zj ) for j = H; L : If G = (T; 0; 0) ) W = zj (k0 + T ) + zj k0 = 2zj k0 + zj T: If G = (0; T; 0) ) W = zj k0 + zj (k0 + T ) = 2zj k0 + zj T: If G = (0; 0; T ) ) W = zj (k0 + T ) + zj (k0 + T ) = 2zj k0 + 2zj T: Then, if 2zj T > zj T , or > 1=2, it is optimal to invest in public infrastructure. On the other hand, if < 1=2, it is optimal to provide the subsidy since public investment is always dominated.3 To have a meaningful trade-o¤ between vertical and horizontal industrial policy, we assume from now on that > 1=2: (2) The optimal policy is thus GF B (zj ; zj ) = (0; 0; T ) for j = H; L: Conditions (1) and (2) give, respectively, and upper and lower bound on the productivity of public infrastructure for this problem to have a non-trivial solution. Summarizing, the …rst best solution when productivity is observable is characterized by the policy 8 < (T; 0; 0) if (z1 ; z2 ) = (zH ; zL ) FB G (z1 ; z2 ) = (0; T; 0) if (z1 ; z2 ) = (zL ; zH ) : (0; 0; T ) if (z1 ; z2 ) = (zL ; zL ) or (z1 ; z2 ) = (zH ; zH ) 3 Of course, if providing subsidies dominates investing in the public good, the distribution of subsidies between the two equally productive …rms is irrelevant. 6 The associated …rst best welfare is 8 < (zH + zL ) k0 + zH T if (z1 ; z2 ) = (zH ; zL ) or (z1 ; z2 ) = (zL ; zH ) FB W (z1 ; z2 ) = 2zH (k0 + T ) if (z1 ; z2 ) = (zH ; zH ) (3) : 2zL (k0 + T ) if (z1 ; z2 ) = (zL ; zL ) The ex-ante expected …rst best welfare is thus E WFB = Pr (z1 = zH ; z2 = zH ) 2zH (k0 + T ) + Pr (z1 = zH ; z2 = zL ) [(zH + zL ) k0 + zH T ] + Pr (z1 = zL ; z2 = zH ) [(zH + zL ) k0 + zH T ] + Pr (z1 = zL ; z2 = zL ) 2zL (k0 + T ) or E WFB = 2 2 H zH (k0 + T ) + 2 H L [(zH + zL ) k0 + zH T ] + 2 2 L zL (k0 + T): (4) 3.2 Optimal contract with private information In this section we develop a simple direct mechanism that is able to implement the …rst best allocation when the government does not observe the …rms’productivities. By the Revelation Principle, without loss of generality, we can focus on direct mechanisms where …rms report their productivities to the planner. Since talk is cheap, if convenient, …rms will have an incentive to misrepresent their types to receive the subsidy. For example, suppose that the government policy is such that everyone who claims to be high productivity receives a subsidy and whatever remains in the budget is allocated to the public good. In such a case, a low productivity …rm will report high productivity because the marginal pro…t of a subsidy is always greater than the marginal pro…t of the public good, since < 1. Without additional instruments, if the government provides the subsidy to …rms depending on their reported productivities, some or all …rms will lie about their type. Therefore, we assume that the government is able to impose a …rm-speci…c tax fj to …rm j = 1; 2 that will be a function of the …rm’ s reported productivity type. Those taxes will help provide the right incentives for the …rms to reveal their true (unobserved) productivity, and the proposed mechanism will be able to implement the …rst best allocation. Let i 2 fH; Lg denote …rm i’ s type and mi ( i ) 2 fH; Lg denote the message space of …rm i. That is, …rm i can report that it is a high productivity or a low productivity …rm. Let m = (m1 ; m2 ) be the vector of reports of the two …rms. The mechanism is a mapping from the reported types to a vector of policies fm1 ; m2 g ) (s1 ; s2 ; g; f1 ; f2 ) which speci…es the subsidies, taxes, and the provision of the public good as a function of the reports. The mechanism is as follows. Consider an arbitrary …rm i = 1; 2: If …rm i reports low productivity, mi ( i ) = L; then …rm i does not receive a subsidy and does not pay any tax. This is independent of the report of the other …rm. If both …rms report low productivity, the government provides the public good. If …rm i reports high productivity, mi ( i ) = H , and …rm j 6= i reports low productivity, mj ( j ) = L, …rm i pays a tax f and receives a subsidy si = T . If both …rms report high productivity, m1 ( 1 ) = m2 ( 2 ) = H , then both …rms pay a tax f~, the individual subsidies are zero, si = 0 for i = 1; 2, and the government provides the public good, g = T . 7 In other words, the mechanism is represented by the policy 8 > > (s1 ; s2 ; g; f1 ; f2 ) = (T; 0; 0; f; 0) if m = (H; L) > < (s1 ; s2 ; g; f1 ; f2 ) = (0; T; 0; 0; f ) if m = (L; H ) G (m) = ~ ~; f > > (s ; s ; g; f1 ; f2 ) = 0; 0; T; f if m = (H; H ) > 1 2 : (s1 ; s2 ; g; f1 ; f2 ) = (0; 0; T; 0; 0) if m = (L; L) We next show that by appropriately choosing the taxes f and f ~, the government can induce truth-telling and achieve the …rst best level of utility. The di¤erence with the …rst best prob- lem, however, is that part of the overall welfare will take the form of government surplus, d > 0, rather than just the …rms’ pro…ts. Since the model is symmetric, we focus on the decision problem of …rm 1: Suppose that 1 = H . The pro…t of …rm 1 as a function of all possible messages is 8 > > (m1 ; m2 ) = (H; L) then v1 = zH (k0 + T ) f < ~ (m1 ; m2 ) = (H; H ) then v1 = zH (k0 + T ) f 1 = H ) if : > > (m1 ; m2 ) = (L; L) then v1 = zH (k0 + T ) : (m1 ; m2 ) = (L; H ) then v1 = zH k0 We now construct taxes f and f ~ that make truth-telling optimal for …rm 1. If …rm 2 reports high productivity, m2 = H , truth telling is optimal if zH (k0 + T ) ~ f zH k0 or ~ f zH T: (5) If …rm 2 reports low productivity, m2 = L, truth telling is optimal if zH (k0 + T ) f zH (k0 + T ) or f (1 ) zH T: (6) ~ such that a high productivity Conditions (5) and (6) are upper bounds on the taxes f and f …rm does not want misrepresent its type. For example, setting f = f ~ = 0 works. The problem remains, however, that a low productivity …rm will always claim to be high productivity to receive the subsidy. We now turn to this case. Suppose now that 1 = L. Then, 8 > > (m1 ; m2 ) = (H; L) then v1 = zL (k0 + T ) f < ~ (m1 ; m2 ) = (H; H ) then v1 = zL (k0 + T ) f 1 = L ) if : > > (m1 ; m2 ) = (L; L) then v1 = zL (k0 + T ) : (m1 ; m2 ) = (L; H ) then v1 = zL k0 We now look for conditions such that truth-telling is optimal for a low productivity …rm. If …rm 2 reports high productivity, m2 = H , truth-telling is optimal for …rm 1 if zL k0 zL (k0 + T ) ~ f 8 or ~ f zL T: (7) If …rm 2 reports low productivity, m2 = L, truth-telling is optimal if zL (k0 + T ) zL (k0 + T ) f or f (1 ) zL T: (8) Conditions (7) and (8) are lower bounds on the taxes so that a low productivity …rm will not claim that it is a high productivity …rm. Summarizing, we have found that it is optimal for all …rms to report their true produc- ~ satisfy tivities as long as the taxes f and f (1 ) zL T f (1 ) zH T; (9) zL T ~ f zH T: (10) The mechanism always taxes …rms that report high productivity. The taxes are such that low productivity …rms do not …nd it optimal to claim to be of high productivity. But the taxes cannot be so high that a high productivity …rm would want to claim to have low productivity. This mechanism implements the …rst best allocation since the subsidy is only given whenever it is productive to do so. The level of welfare is the same as that in the …rst best solution, the di¤erence being that part of that welfare is derived from government surplus d = f1 + f2 rather than just by the …rms’pro…ts. The only case in which the government does not raise any surplus is when both …rms claim to be low productivity and the government provides the public good. 3.3 Simple policies Even though the mechanism that we described above is fairly simple, it may be argued that it still requires some degree of sophistication that may not be available or feasible in less developed countries. For that reason here we compare two simple (but sub-optimal) policies that do not involve taxes at all and that set the government surplus to zero. Simple policy 1: provide only the public good Simple policy 2: provide a subsidy to whomever claims to be of high productivity. If both …rms report high productivity, set the subsidy to s1 = s2 = T =2.4 3.3.1 Simple Policy 1 If the government provides the public good and sets subsidies to zero, the welfare conditional on productivities z1 and z2 is z1 (k0 + T ) + z2 (k0 + T ) = (z1 + z2 ) (k0 + T ) 4 A third Simple policy that randomizes between the two …rms and gives the subsidy accordingly gives the same ex-ante welfare as Simple policy 2. 9 It then follows that the expected welfare under Simple Policy 1 is E W SP 1 = Pr (z1 = zH ; z2 = zH ) (zH + zH ) (k0 + T ) + Pr (z1 = zH ; z2 = zL ) (zH + zL ) (k0 + T ) + Pr (z1 = zL ; z2 = zH ) (zL + zH ) (k0 + T ) + Pr (z1 = zL ; z2 = zL ) (zL + zL ) (k0 + T ) or E W SP 1 = 2 H zH + H L (zH + zL ) + 2 L zL 2 (k0 + T ) (11) It is simple to show that the expected …rst best welfare can be written as E W F B = E W SP 1 + 2 H LT [zH (zH + zL ) ] : Then, by assumption (1), E W F B > E W SP 1 ; so that the optimal policy strictly domi- nates Simple Policy 1. Indeed, assumption (1) guarantees that there are cases in which it is optimal to provide the subsidy and hence the suboptimality of the proposed simple policy. 3.3.2 Simple Policy 2 The second simple policy consists of subsidizing any …rm that claims to be high productivity. Since < 1, all …rms will report high productivity and receive a subsidy si = T =2. The expected welfare under Simple Policy 2 is thus E W SP 2 = Pr (z1 = zH ; z2 = zH ) (zH + zH ) (k0 + T =2) + Pr (z1 = zH ; z2 = zL ) (zH + zL ) (k0 + T =2) + Pr (z1 = zL ; z2 = zH ) (zL + zH ) (k0 + T =2) + Pr (z1 = zL ; z2 = zL ) (zL + zL ) (k0 + T =2) or E W SP 2 = 2 H zH + H L (zH + zL ) + 2 L zL 2 (k0 + T =2) : (12) Note that assumption (2) ( > 1=2) implies that Simple Policy 1 always dominates Simple Policy 2. 4 Discussion The paper derives two main results. The …rst is that industrial policies in the form of …rm subsidies can attain the …rst-best allocation of government resources if accompanied by an appropriate mix of taxes, even in the context of private information. Implementing this tax- and-subsidy mechanism, however, requires a certain degree of government capability. The second result is that when this capability is lacking and productivity information is not pub- licly observed, the provision of public goods always dominates the granting of …rm subsidies (evenly, randomly, or to whomever claims to be of high productivity). These are strong results. They follow from the condition that public goods be su¢ ciently productive, in the sense that there be a meaningful trade-o¤ between public goods and …rm subsidies under both perfect and private information. Finally, note that the …rst result relies on the linearity of the pro- duction function. In a neoclassical production function, where, say, public infrastructure and private capital are both factors of production, the optimal policy is likely to involve providing a mixture of public goods and …rm subsidies, instead of …rm subsidies alone. Other possible extensions include allowing for costly state veri…cation or imperfect monitoring, which we leave for future research. 10 5 Bibliography Aiginger, Karl. 2007. "Industrial Policy: A Dying Breed or A Re-Emerging Phoenix." Journal of Industry, Competition and Trade 7 (3): 297-323. doi:10.1007/s10842-007-0025-7. Aiginger, Karl, and Susanne Sieber. 2006. "The Matrix Approach to Industrial Policy." International Review of Applied Economics 20 (5): 573-601. doi:10.1080/02692170601005507. Inter-American Development Bank (IADB). 2014. "Rethinking Productive Development: Sound Policies and Institutions for Economic Transformation." Edited by Ernesto H. Stein, Gustavo Crespi, and Eduardo Fernández-Arias. New York: Palgrave MacMillan. Chang, Ha-Joon. 2011. "Industrial Policy: Can We Go Beyond an Unproductive Con- frontation?" In Justin Lin and Boris Pleskovic, (eds.), Annual World Bank Conference on Development Economics 2010, Global: Lessons from East Asia and the Global Financial Cri- sis. Washington, DC: World Bank. Fernández-Arias, Eduardo, Charles Sabel, Ernesto H. Stein, and Alberto Trejos. 2016. "Two to Tango: Public-Private Collaboration for Productive Development Policies." Wash- ington, DC: Inter-American Development Bank. IMF, OECD, UN, and World Bank. 2015. "Options for Low income Countries’E¤ective And E¢ cient Use of Tax Incentives for Investment." A report to the G-20 Development Working Group. Jaud, Mélise, and Caroline Freund. 2015. "Champions Wanted: Promoting Exports in the Middle East and North Africa." Directions in Development. Washington, DC: World Bank. doi: 10.1596/978-1-4648-0460-1. Pack, Howard. 2000. "Industrial Policy : Growth Elixir or Poison?". The World Bank Research Observer 15 (1): 47-67. http://documents.worldbank.org/curated/en/2607214683 31838470/Industrial-policy-growth-elixir-or-poison. Pellegrin, Julie, Maria Letizia Giorgetti, Camilla Jensen, and Alberto Bolognini. 2015. "EU Industrial Policy: Assessment of Recent Developments and Recommendations for Future Policies. Study for the ITRE Committee." European Parlament. PNG Tax Committee. 2014. "Papua New Guinea Taxation Review (2013-2015), Issue Paper No.5: An Examination of the Advantages and Disadvantages of Tax Incentives." Papua New Guinea. Rodrik, Dani. 2004. "Industrial Policy for the Twenty-First Century." KGS Faculty Research Working Paper Series. No. RWP04-047. http://dx.doi.org/10.2139/ssrn.617544. Rodrik, Dani. 2008. "Industrial Policy: Don’ t Ask Why, Ask How." Middle East Devel- opment Journal 1 (01): 1-29. http://dx.doi.org/10.2139/ssrn.617544. Stiglitz, Joseph E., Justin Yifu Lin, and Célestin Monga. 2013. "Introduction: The Reju- venation of Industrial Policy." In The Industrial Policy Revolution I: The Role of Government Beyond Ideology, edited by Joseph E. Stiglitz and Justin Yifu Lin, 1âe“15. London: Palgrave Macmillan UK. doi:10.1057/9781137335173_1. Tanzi, Vito, and Parthasarathi Shome. 1992."The Role of Taxation in the Development of East Asian Economies." In The Political Economy of Tax Reform, NBER-EASE Vol 1, edited by Takatoshi Ito and Anne O. Krueger. University of Chicago Press. 11 Wade, Robert H. 2016. "Industrial Policy in Response to the Middle-Income Trap and the Third Wave of the Digital Revolution." Global Policy 7 (4): 469-80. doi:10.1111/1758- 5899.12364. Warwick, Ken. 2013. "Beyond Industrial Policy: Emerging Issues and New Trends." OECD Science, Technology and Industry Policy Papers 2. Paris: OECD Publishing. Warwick, Ken, and Alistair Nolan. 2014. "Evaluation of Industrial Policy: Methodological Issues and Policy Lessons." OECD Science, Technology and Industry Policy Papers 16. Paris: OECD Publishing. http://dx.doi.org/10.1787/5jz181jh0j5k-en. Weiss, John. 2013. "Industrial Policy in the Twenty-First Century: Challenges for the Future." In Pathways to Industrialization in the Twenty-First Century: New Challenges And Emerging Paradigms, edited by Adam Szirmai, Wim Naudé, and Ludovico Alcorta. Oxford: Oxford University Press. doi:10.1093/acprof:oso/9780199667857.003.0015. 12