Market Dynamics and Productivity in Developing Countries

This page intentionally left blank

Market Dynamics and Productivity in Developing Countries

Economic Reforms in the Middle East and North Africa

Khalid Sekkat
Editor

Editor
Khalid Sekkat
University of Brussels
1050 Brussels
Belgium
ksekkat@ulb.ac.be

A copublication with the
International Development Research Centre
PO Box 8500
Ottawa, ON, Canada K1G 3H9
info@idrc.ca/www.idrc.ca

e-ISBN (IDRC e-book) 978-1-55250-459-8

ISBN 978-1-4419-1036-3                       e-ISBN 978-1-4419-1037-0
DOI 10.1007/978-1-4419-1037-0
Springer New York Dordrecht Heidelberg London

Library of Congress Control Number: 2009939074

© Springer Science+Business Media, LLC 2010
All rights reserved. This work may not be translated or copied in whole or in part without the written permission of the publisher (Springer Science+Business Media, LLC, 233 Spring Street, New York, NY 10013, USA), except for brief excerpts in connection with reviews or scholarly analysis. Use in connection with any form of information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed is forbidden.
The use in this publication of trade names, trademarks, service marks, and similar terms, even if they are not identified as such, is not to be taken as an expression of opinion as to whether or not they are subject to proprietary rights.

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

Contents

This page intentionally left blank

About the Editor and Contributors

Editor

Khalid Sekkat holds a Ph.D. in Economics from the University of Brussels (ULB). He is Full Professor of Economics at the ULB and a member of the Royal Academy of Sciences and Technologies (Kingdom of Morocco). His fields of interest are industrial organization, international economics, and development economics. He has published numerous articles in international refereed journals (including Journal of Industrial Economics, Journal of International Economics and European Economic Review), and several books, including Trade and Jobs in Europe (co-editor, Oxford, 1999), Vertical Relationships and the Firm in the Global Economy (Elgar, 2006), Competition and Efficiency in the Arab World (editor, Palgrave-McMillan, 2008), and Industrial Policy in Europe, Japan, and the USA (co-author, Palgrave-McMillan, 2009). He was economic adviser at the European Commission and has acted as a consultant to the European Parliament, the European Commission, the World Bank and the OECD.

Contributors

Lahcen Achy is Professor of Economics at the National Institute of Statistics and Applied Economics (INSEA), Morocco. He has a Ph.D. in International Economics and Finance from the Université Libre de Bruxelles (ULB), Belgium. His fields of specialization are international economics and trade policy, competition policy, and manufacturing sector issues. Lahcen Achy has several publications in internationally referred journals such as Review of Development Economics and Applied Economics. He has also contributed to a number of books on finance, competition, globalization, and income distribution in developing countries. Lahcen Achy has been adviser and consultant for a number of international organizations such as UNIDO, WIPO, WHO, ILO, the World Bank and the European Commission. He is also a Research Associate of the DULBEA (Department of Applied Economics at ULB), Research Fellow of the Economic Research Forum (ERF), member of the Global Development Network (GDN), and the academic liaison of Research Alliance for Development (RAD).

Nesreen Barakat is the Managing Director of “To Excel Consulting & Associates” – a consulting firm based in Amman (Jordan) specialized in strategic planning, institutional strengthening, policy advocacy, and economic research. Nesreen Barakat holds an MBA degree from Durham University Business School in the UK and a BSc in Computer Science from the University of Jordan. Previously she managed the Policy Support and Institutional Strengthening Component of the Euro-Jordanian Industrial Modernisation Programme. She also worked as the Director of the Competitiveness Unit (TCU) at the Ministry of Planning and International Cooperation and was assigned as an official counterpart to the World Economic Forum (WEF) to follow up on Jordan’s competitiveness ranking in the Global Competitiveness Report (GCR). Additionally, Nesreen Barakat was a member of the Technical Policy Analyst Team to support the official negotiators during Jordan’s accession to the Jordan-EU Association Agreement.

Riadh Ben Jelili is Ph.D. in Economics and Econometrics. Since September 2005, he is Economic Expert and Coordinator of the Consultancy Committee at Arab Planning Institute. From 1994 to 2005, he was Associate Professor at University of Rennes 1 and University of South Britain. His fields of specialization are performance, competitiveness and investment decision in manufacturing sector, rural–urban migration, international migration, and simulated methods in econometrics. He published in books and international journals (e.g. Palgrave-Macmillan, Journal of Development and Economic Policies).

Mohamed Goaied is Professor of Econometrics and Statistics at the Institute of High Business Studies in Carthage and Polytechnic School of Tunisia. He is Ph.D. in Mathematical Economics and Econometrics from University of Bougogne in Dijon, France (1988). He published many papers in academic journals including: Journal of Productivity Analysis, Econometric Reviews, International Review of Finance and Economie et Prevision. His fields of specialization are productivity analysis, firm efficiency, demand systems, and poverty analysis. He served as a Managing Director of The Institute of High Business Studies in Carthage (2002–2005) and is Director of the Laboratory of Applied Economics and Empirical Finance at the University of 7 November in Carthage, since 2005. He also served as a consultant for national and international institutions. He is a fellow of the ERF.

Teoman Pamukçu is currently Associate Professor at the Department of Science and Technology Policy Studies at the Middle East Technical University, Turkey. He previously worked as teaching and research assistant at the ULB, and was an expert at the Belgian Federal Planning Bureau and Research Fellow at the University of Luxembourg. He holds a BA degree in economics, a MA degree in econometrics and a Ph.D. in economics from the ULB. He worked on issues relating to the emergence and development of information society, on labor market polices and on the determinants of innovation activities of firms in developing countries. His current research interests are foreign direct investment, clusters of enterprises and their impact on innovation and economic development. He has published articles in journals and chapters in books. Teoman Pamukçu is a Research Associate of the ERF, a Research Associate of the DULBEA and member of the International Joseph A. Schumpeter Association.

Ibrahim Saif is an Economist at the Economic Studies Unit at the Center for Strategic Studies at the University of Jordan. He obtained his Ph.D. from the School of Oriental and African Studies; University of London. Ibrahim Saif has taught both at the University of London (UK) and Yale University (USA) where he offered courses on the Economies of the Middle East. He also worked as a consultant to international organization such as the World Bank, the IMF and the ILO. He is a Fellow of the ERF and a member of the GDN. His core research areas include international trade and structural adjustment programs in the developing countries and his regional interest in addition to Jordan includes Middle East region. Currently he is working on regional project on competition and the political economy of Euro-Med Association Agreement. He is a member of the Competition Committee in Jordan.

Erol Taymaz is a Professor of Economics at Middle East Technical University (Ankara, Turkey). He was educated in mechanical engineering at METU. He received his M.S. degree in Economics at METU (1985), and Ph.D. degree in Economics at Case Western Reserve University, USA (1989). He worked as a Research Engineer at Aselsan Electronics Industries (Ankara, 1982–1985), Research Assistant at CWRU (Cleveland, 1985–1989), and Research Fellow at Industrial Research Institute (IUI, Stockholm, 1989–1992). He is at the Department of Economics of the Middle East Technical University since 1992. His current research interests include economics of technology and innovation, industrial and technology policy, industrial dynamics, small business economics, and microsimulation modeling.

This page intentionally left blank

Chapter 1
Reforms, Market Dynamics and Productivity in Developing Countries

Khalid Sekkat

1.1 Introduction

Between the end of World War II and the mid-eighties, development strategy in a number of developing countries was based on the protectionist “Import substitution” (IS) concept. Many of these countries’ governments were of the opinion that maintaining free trade would prevent their economies from industrializing and would therefore render them vulnerable to long-term adverse movements in terms of trade, and impact their growth and welfare. However, during the 1980s, both economists and policy-makers became skeptical about the beneficial impact of the IS strategy. The difference in performance between the outward-oriented Asian and the inward-oriented Latin American economies clearly called for a reconsideration of the strategy. Empirical evidence (Sachs and Warner 1995) also suggests that open economies tend to adjust more rapidly from primary-intensive to manufacture-intensive exports, and to achieve sustained growth. Since the mid-eighties, many LDCs have engaged in a process of economic reform, involving a more outward orientation of their economies, the lowering of trade barriers, privatization of many industries and reform of the foreign-exchange market.

The rationale behind reforms is that increased competition is an important driver of economic performance. Economists agree, in general, that fair competition is beneficial to growth because it induces efficiency gains. Among the latter, productive efficiency is of particular importance. It brings the output-input combination to the optimal production frontier and induces firms to produce at lowest costs. This is achieved by fighting overstaffing, sluggish response to new opportunities and poor management. It forces inefficient firms either to exit the market or to take necessary actions to reduce costs and rationalize production and management processes. The outcome is an increase in productivity, both at micro and macro levels.

K. Sekkat ()
University of Brussels, Belgium
e-mail: ksekkat@ulb.ac.be

K. Sekkat (ed.), Market Dynamics and Productivity in Developing Countries: Economic Reforms in the Middle East and North Africa, DOI 10.1007/978-1-4419-1037-0_1, © Springer Science + Business Media, LLC 2010

Recent analyses (see Hoekman and Winters (2005) for an overview) of the impact of liberalization on efficiency in developing countries (LDCs) lend support to the role of such “natural selection” process among firms. For instance, Wacziarg and Wallack (2004) analyzed a set of 25 liberalization episodes in developing countries and found a strong effect of intra-industry reallocation on economic performance.

While there has been a profusion of theoretical work on entry and exit of firms, there is comparatively little empirical work in the area even for developed countries (Disney et al. 2003a, b). Firm entry and exit is a part of the market selection process by which resources are reallocated within or across industries. The process of entry and exit influences economic performance through firms’ internal restructuring, reallocation of resources among firms, and changes in market shares of incumbents. It also induces the introduction of new technologies, thereby improving economic performance.

This chapter presents a brief review of the knowledge about firms’ entry and exit and economic performance. The objective is to set the stage and motivate the researches presented in the subsequent chapters. Section 2 presents the conceptual framework underlying the determinants and impacts of firms’ entry and exit. Section 3 summarizes the main finding in developing and transition economies. Section 4 concludes.

1.2 Firms’ Entry and Exit: Determinants and Impacts

A series of firm, industry, and country specific factors determines the extent of entry and exit. The process of entry and exit has impacts on economic performance. However, such relationships are not unidirectional, and literature documents various cases where performance affects either firm, industry, or country determinants as well as the process of entry and exit itself (Tirole 1988). For clarity, we do not discuss the latter links in this chapter (Fig 1.1).

Fig. 1.1 Firms’ entry and exit: determinants and impacts

1.2.1 Determinants

1.2.1.1 Firm Determinants

There is a relationship between firm entry and size. Entrants tend to have a size which is lower than the average firm size found in the industry (Boeri and Cramer 1992). Some authors (Audretsch and Mahmood 1995) have uncovered evidence that firms that enter the market with a small size do so because they consider that their chance of success is small. However, if after entry, results seem promising, small sized firms have the option of investing heavily.

A relationship was also found between firm exit and size. Firms exiting the industry have a smaller size than the average size of the industry. Moreover, many small firms exit the industry before reaching their efficient scale of production, which may reflect the intensity of the market selection process.

Another finding is that young firms tend to exit the market after a relatively short period of activity. This may be linked to the fact that newly created firms generally enter the market with a small amount of own funds. As they are not competitive enough during the first years of their existence, they make losses which decrease the level of their own funds. After 2 or 3 years, the latter become insufficient to allow the firm to pursue its activity, and its goes bankrupt. The result may even happen to newly created firms gaining market shares if they have no access to external funding and if the level of their own funds does not allow them to finance their expansion. So, one important element among the determinants of exit is the initial amount of own funds and more generally the financial structure of the company.¹

In terms of firm survival, research shows that the survival of entrants is low, with a large number of entrants failing within the first year (Churchill 1955; Baldwin 1995). The firms that survive need 5–10 years to properly compete with incumbents and their failure rates decrease over time. Firms that survive in the market have a larger size than those that exit and also have a more rapid growth rate which declines with age however.

1.2.1.2 Industry Determinants

Industry’s characteristics include profit margins, concentration ratio, growth rate, capital intensity and specific workers skills. Profit margin determines the attractiveness for new firms to enter into the industry, but it could also be associated with imperfect competition. In the former case, the expected effect on entry is positive, while in the latter, the reverse is expected. The concentration is an indicator of the easiness to enter a market. It is easier to enter perfectly competitive industries in which many small firms produce standard products. The growth rate of the industry is a proxy of its life cycle. New firms prefer to enter rapidly growing industries. Capital intensity captures “natural” barriers to entry. It may discourage entry because if the industry uses capital-intensive technology, the cost of the initial investment could be substantial. Finally, the requirement of specific workers’ skills could also deter entry if not enough is available in the country.

Regarding exit, similar variables are in play, but with an effect opposite to the one on entry. For instance, capital intensity (because of potential sunk costs) delays exit and high industry growth rate allows firms’ survival despite their low performance. Finally, high concentration ratio reduces competition among firms and may reduce exit.

According to Caves (1998), entry and exit rates tend to be positively correlated in industries with steady states of maturity, but varying structural entry barriers. The correlation between the two is negative during the early and late phases of a product’s life cycle. During the expansion phase, industries have both high entry rates and high exit rates. The overall impact on employment tends to be small, given the lower than average size of both entering and exiting firms.

1.2.1.3 Country Determinants

Entry and exit are affected not only by firm and industry characteristics, but also by country characteristics such as macroeconomic shocks. More importantly, researches suggest that a firm’s investment decisions are highly sensitive to the country’s institutions and policies. Such policies, by affecting the business climate, can either promote or deter firms’ willingness to enter or stay in the market. To our knowledge, there are little studies focusing on the impact of policies and institutions on firms’ entry and exit. Indirect insights can be gathered from the studies concerning firms’ investment. Some of them focused on trade and foreign exchange policies and found that the size and openness of a country are important determinants of investments (Lucas 1993; Cushman 1985). Others looked at policies such as grants, subsidies, tax abatement, loan’s guarantees, and interest subsidies. They found that the impacts differ between developing and developed countries (Grubert and Mutti 1991; Loree and Guisinger 1995).

Another strand of empirical literature, inspired by Douglas North’s works, is increasingly dealing with the role of institutions. These refer to a large set of factors including political instability, corruption, investment regulation, democratic accountability, and bureaucratic quality. For instance, Schneider and Frey (1985) found that both economic and political factors are crucial to fostering investment. Mauro (1995) found that corruption depresses domestic investment; a result confirmed by Wei (2000) regarding foreign investment. Brunetti and Weder (1998) used various indicators of the quality of institution (e.g., voice and accountability, government effectiveness, regulatory burden, and rule of law) and found that their deterioration decreases investment. Finally, Henisz (2000) examined the effect of commitment to rules on growth and investment. He focused on the effect of frequent or arbitrary changes in taxation, regulation and other relevant economic policies. He found that commitment to rules has a statistically and economically significant impact on investment. Although indirect, these findings suggest that taking account of institutions and policies is important for the study of firms’ entry and exit.

1.2.2 Entry, Exit and Economic Performance

A large part of the studies on the impact of firms’ entry and exit on economic performance concentrates on productivity growth. The impact of firm entry and exit on aggregate productivity growth is decomposed into three sources. First, the “within effect,” or restructuring effect, refers to factors internal to the firm such as organizational change, the introduction of new technologies, R&D activities or a change in the mix of labor and capital. Second, there is a process of creative destruction by which low productivity firms exit the market and are replaced by new entrants. Among them, the most efficient will survive, while the least efficient will exit the market in subsequent periods. Third, there is a change in market shares among incumbents which will also have an impact on aggregate productivity growth. There are also a number of important interactions between the various sources of aggregate productivity growth.

Entry and exit of firms is generally found to have an important contribution to aggregate productivity growth. Scarpetta et al. (2002) analyzed several OECD countries and found that entry and exit contributed between 20% and 40% of aggregate productivity growth. There were significant differences in the contributions of entry to aggregate productivity growth across sectors however. In high technology sectors, the entry of new firms has a larger than average contribution to total growth, whereas in mature industries, the exit of firms has larger contributions to growth. The results also differ according to whether aggregate productivity is measured by Total Factor Productivity (TFP) or labor productivity. Disney et al. (2003b) found that, in the UK, the contribution of entry and exit to TFP growth was sensitive to the business cycle and was larger in periods of economic expansion. Martin and Jaumandreu (1999) uncovered evidence that entry and exit play an important role in aggregate productivity growth in Spain. The impact was stronger in the period before Spanish integration in the EU.

The contributions of the three sources of growth to aggregate productivity growth varies from one study to another, depending on the method of decomposition used, the measurement of aggregate productivity, the time horizon over which changes occur, the business cycle, as well as on the country and the industry under investigation. The contribution of the “within effect” was found to be an important source for aggregate growth by Foster et al. (1998). Scarpetta et al. (2002) uncovered evidence that the within effect had larger contributions to growth in mature industries. It also has a higher contribution to growth in periods of economic expansions. The contribution of the “within effect” to aggregate productivity growth remains important, but is smaller when productivity is measured by TFP than when it is measured by labor productivity. The results on the contribution of market share reallocation to aggregate productivity growth are mixed. Aw et al. (1997), Hahn (2000) and Griliches and Regev (1995) found that the contribution is small, whereas Baily et al. (1992) and Foster et al. (1998) found that the contribution is important among US industries. According to Scarpetta et al. (2000), the contribution of market share reallocation is positive, but typically small and varies widely from one country to another. It also increases in periods of economic slowdown.²

1.3 Entry, Exit and Economic Performance in LDCs

1.3.1 Entry, Exit and Reforms

To date, a major reference about manufacturing firms in developing countries is the survey by Tybout (2000). He documented entry rates, exit rates, net job creation and net job destruction patterns among the population of plants with at least ten workers. The survey shows that there is more plant and job turnover in a number of developing countries than in the United States and Canada. In Chile, entering plants with at least ten workers captured 15% of the market and in Colombia they captured 20% of the market. In terms of job creation and job destruction, Chile and Colombia average 25 and 27% annual turnover rates, respectively. In Morocco, even more flux in plants and jobs were found. The annual manufacturing job turnover rate was 31%. Finally, in Korea and Taiwan, new entrants captured an average of 33 and 44% of the market, respectively

The reasons for such high turnover are different among countries. In Latin America, high turnover seems to reflect the dramatic business cycles. In Korea and Taiwan, it partly reflects rapid expansion of the manufacturing sector. Policies also seem to matter. Market shares and turnover rates are higher in Korea and Taiwan than they are in Latin America. This can be linked to labor markets regulation. For instance, Taiwan’s labor markets are less regulated than in Latin America, and sunk entry costs are relatively modest, because the business environment makes subcontracting easy.

However, it seems that the turbulence takes place among plants with 10–50 workers. Moreover, the moderately small producers never seriously challenge the larger incumbents. Hence, high turnover rates in LDCs may simply reflect the relative importance of small and medium enterprises and not necessarily imply that large firms’ market shares are more at risk. Studies for Chile and Colombia suggest that all of the market share loss comes at the expense of small producers. Here again, one finds a contrast between Latin America and Asia. Studies for Korea and Taiwan show that large plants suffer important market share losses.

It seems that there are marked contrasts among countries about the profile and the reasons of entry and exit. The contrast is related to countries’ institutions, policies and economic environment. This implies that detailed country studies should be more helpful to provide policy recommendations than aggregate cross-countries analysis. Some recent country studies are presented in what follows.

Fajnzylber et al. (2001) focused on the role of trade reform in Chile and Colombia. It used establishment level data to document patterns of job creation, destruction and turnover. It also assesses the share of employment changes due to within-firm versus entry and exit effects. The paper established that the contribution of entering and exiting firms to employment change is almost as important as that of continuing firms. However, the paper failed to show whether job creation, destruction and turnover are due to trade reform or to business cycle. Campos and Iootty (2005) used Brazilian data for the manufacturing sector (at the 3-digit level) for the period 1996–2002 and found that the share of exports in the sector’s output is one main determinant of entry and exit rates. It also identified a relationship between the impact of export orientation and the business cycle. In years of real per capita GDP decline, entry rates increase with the export propensity, in years of GDP expansion, net entry increases with domestic growth. Masso et al. (2004) focused on Estonia and showed that firm turnover has been rather high during the period 1995–2001, thanks to low institutional entry barriers and the emergence of numerous SMEs. Moreover, new firms achieved higher productivity compared to incumbent and, therefore, exhibited high survival rates. Yang (2004) examined the impact of the transition to the “socialist market economy” in China on firms’ entry, exit and survival in the electrical and engineering industry. The analysis suggested that the competitive selection process is taking place, with entrants contributing substantially to both output growth and productivity growth. Old firms are, however, still important in the economy. Kaya and Ucdogruk (2002) analyzed the case of the Turkish manufacturing industry (at the 4-digit level) over the period 1981–1997. The findings are that entry rates are highly affected by profit margin, capital intensity, concentration ratio and growth rate. The main determinants of exit are growth rate, capital intensity and concentration ratio. The entry rate is highly correlated with the exit rate.

Finally, Klapper et al. (2004) used a comprehensive database of firms in Western and Eastern Europe to study how regulations governing entry drive the creation of new firms. They found that entry regulations hamper entry, especially in industries that naturally should have high entry. Also, value added per employee in naturally “high entry” industries grows more slowly in countries with onerous regulations on entry. Interestingly, regulatory entry barriers have no adverse effect on entry in corrupt countries, but only in less corrupt ones. Taken together, the evidence suggests that bureaucratic entry regulations are neither benign nor welfare improving. However, not all regulations inhibit entry. In particular, regulations that enhance the enforcement of intellectual property rights or those that lead to a better developed financial sector lead to greater entry in industries which do more R&D or need more external finance.

1.3.2 Entry, Exit and Economic Performance

Regarding economic performance, Roberts and Tybout (1996) provides evidence supporting a positive relationship between high turnover and gains in productivity in LDCs. Other studies confirm the relationship in different countries. For Korea, Hahn (2000) found very large effect of entry/exit on aggregate productivity growth. Aw et al. (1997) focused on Taiwan and found that entry and exit processes are important sources of aggregate growth. Pavcnik (2002) finds that exiting plants are substantially less productive than surviving plants in Chile. Griliches and Regev (1995) showed that most of the growth in labor productivity in Israel is due the “within effects.” This fits with Tybout (2003) findings that competition (through openness to trade) induces markets for the most efficient plants to expand, but at the same time large import-competing firms tend to contract. Palangkaraya and Yong (2006) found that Indonesian incumbent plants, are on average, more productive than entrants and exitors in every year between 1990 and 1995. New plants are relatively less productive than the exiting plants in the early years, but they are more productive in the later years. They also exhibit the highest productivity change during the early years. Decomposition in the change of productivity suggests that such high productivity growth is due to a movement toward the frontier. In Brazil, Campos and Iootty (2005) found that exit (and to a lesser extent, entry and net entry) is a very robust determinant of total factor productivity across industrial sectors.

The transition from a centrally planned economy to a more market oriented one offered an important opportunity to further document whether the competitive selection process induces any improvement in productivity. Masso et al. (2004) focused on entry and exit in Estonia and decomposed productivity change into components consisting of resource reallocation across existing firms, firm entry and exit, and productivity growth within continuing firms. The decomposition shows that the high productivity growth has been mostly from within-firm productivity growth (e.g., the adoption of new production technologies and organizational changes), but the reallocation of production factors (especially the exit of low productivity units) has played an important role as well. Brown and Earle (2004) studied productivity enhancing reallocation for Russia and Ukraine. Before reforms in Soviet Russia, the reallocation rates were low and bore little relation to relative labor and multifactor productivity across firms. After reforms, increasing resource flows have contributed to aggregate productivity growth through both increased flows from less productive to more productive continuing firms, and the exits of less productive enterprises. Orazem and Vodopivec (2003) showed that in the Slovenian manufacturing, competitive pressures sorted out the most efficient firms, and the entry of efficient new private firms was the major source of TFP growth. De Loecker and Konings (2003) showed that more than 40% of average productivity growth in Slovenian manufacturing was due to firm entry and exit. Warzynski (2002), focusing on Poland, found that more job reallocation was connected with more productive industries. Finally, Yang (2004) focused on the Chinese electrical and engineering industry. Productivity decomposition suggests that exits contribute to productivity improvement especially within the small firms. In these firms, as well as in the collectively owned ones, the competitive selection process operates like in a private market economy. However, for state owned enterprises the rate of exit is much slower, and compared with new entry, the contribution of exit to productivity growth is trivial.

1.4 Conclusion

The process of firm entry and exit is a part of a “natural selection” by which resources are reallocated within or across industries. It influences economic performance through firms’ internal restructuring, reallocation of resources among them and changes in their market share. The theoretical literature on entry and exit of firms and their impacts on economic performance is rich but its empirical counterpart is very limited especially on developing countries. The latter have adopted important reforms over the past three decades with the objective of improving their economic performance. The process of firm entry and exit can play an important role in this context.

This chapter briefly reviewed the economic rationale behind the process of firm entry and exit and its impact on economic performance. It first explained how a series of firm, industry and country specific factors affect the process. It subsequently documented the relationship between economic performance and firm entry and exit in developing countries.

The analysis showed that there are marked contrasts across countries about the determinants and impacts of entry and exit. Such contrasts are related to countries’ institutions, policies and economic environment. It follows that detailed country studies should be more helpful to provide policy recommendations than aggregate cross-countries analysis. This is the purpose of the following chapters.

1.5 Notes

1. The literature also found that manager turnover in small businesses had a tendency to predict sales or closures of the firm shortly afterwards.

2. There is also an indirect effect via innovation. Firm entry and exit can also affect productivity growth by stimulating innovation (see Aghion et al. 2003).

References

Aghion P, Blundell R, Griffith R, Howitt P, Prantl S (2003) Firm entry and growth: theory and micro evidence. Harvard University, miméo

Audretsch DB, Mahmood T (1995) New firm survival: new results using a hazard function. Rev Econ Stat 77(1):97–103

Aw BY, Chen X, Roberts MJ (1997) Firm-level evidence on productivity differentials, turnover, and exports in Taiwanese manufacturing. NBER working paper 6235, October

Baily MN, Hulten C, Campbell D (1992) Productivity dynamics in manufacturing plants. Brookings Papers on Economic Activity (Microeconomics), pp 187–267

Baldwin J (1995) The dynamics of industrial competition. Cambridge University Press, Cambridge MA

Boeri T, Cramer U (1992) Employment growth, incumbents and entrants: evidence from Germany. Int J Ind Organ 10:545–565

Brown JD, Earle JS (2004) Economic reforms and productivity-enhancing reallocation in the Post-Soviet transition. IZA discussion paper no 1044

Brunetti A, Weder B (1998) Investment and institutional uncertainty: a comparative study of different uncertainty measures. Weltwirtsch Arch 134(3):513–533

Campos NF, Iootty M (2005) Firm entry and exit in Brazil: cross-sectoral evidence from manufacturing industry. Proceedings of the 33th Brazilian economics meeting

Caves RE (1998) Industrial organization and new findings on the turnover and mobility of firms. J Econ Lit 36(8): 1947–1982

Churchill B (1955) Age and life expectancy of business firms. Surv Curr Bus 35(12):15–19

Cushman DO (1985) Real exchange rate risk, expectations, and the level of direct investment. Rev Econ Statist 67(2):297–308

De Loecker J, Konings J (2003) Creative destruction and productivity growth in an emerging economy: evidence from Slovenian manufacturing. IZA discussion paper no 971

Disney R, Haskel J, Heden Y (2003a) Exit entry and establishment survival in UK manufacturing. J Ind Econ 51:93–115

Disney R, Haskel J, Heden Y (2003b) Restructuring and productivity growth in UK manufacturing. Econ J 113(489):666–694

Fajnzylber P, Maloney WF, Ribeiro E (2001) Firm entry and exit, labor demand, and trade reform evidence from Chile and Colombia. World Bank Policy Research Working Paper 2659

Foster L, Haltiwanger J, Krizan C (1998) Aggregate productivity growth: lessons from microeconomic evidence. NBER working paper 6803

Griliches Z, Regev H (1995) Firm productivity in Israeli industry 1979–1988. J Econometrics 65:175–203

Grubert H, Mutti J (1991) Taxes, tariffs and transfer pricing in multinational corporate decision making. Rev Econ Statist 5:285–293

Hahn C-H (2000) Entry, exit, and aggregate productivity growth: micro evidence on Korean manufacturing. Korea Development Institute, Seoul

Henisz WJ (2000) The institutional environment for economic growth. Econ Politics 12(1):1–31

Hoekman B, Winters A (2005) Trade and employment: stylized facts and research findings. ECES working paper no. 102 May 2005

Kaya C, Ucdogruk Y (2002) The dynamic of entry and exit in Turkish manufacturing industry. Middle East Technical University, Ankara, ERC working papers in Economics 02/02

Klapper L, Laeven L, Rajan R (2004) Business environment and firm entry: evidence from international data. NBER working paper no. 10380

Loree DW, Guisinger SE (1995) Policy and non-policy determinants of U.S. equity foreign direct investment. J Int Bus Stud 26(2):281–299

Lucas REB (1993) On the determinants of direct foreign investment: evidence from East and Southeast Asia. World Dev 21(3):391–406

Martin A, Jaumandreu J (1999) Entry, exit, and productivity growth in Spanish manufacturing during the eighties. Documentos de trabajo, Fundacion SEPI, Programa de investgaciones economicas, Madrid

Masso J, Eamets R, Philips K (2004) Creative destruction and transition: the effects of firm entry and exit on productivity growth in Estonia. IZA discussion paper 1243

Mauro P (1995) Corruption and growth. Q J Econ 110(3):681–712

Orazem PF, Vodopivec M (2003) Do market pressures induce economic efficiency: the case of Slovenian manufacturing, 1994–2001. IZA discussion paper no 901

Palangkaraya A, Yong J (2006) Entry, exit and productivity of Indonesian electronics manufacturing plants. Melbourne Institute working paper no. 8/06

Pavcnik N (2002) Trade liberalization, exit and productivity improvements: evidence from Chilean plants. Rev Econ Stud 69(1):245–276

Roberts M, Tybout J (eds) (1996) Industrial evolution in developing countries: micro patterns of turnover, productivity and market structure. Oxford University Press, New York

Sachs J, Warner A (1995) ‘Economic Reform and the Process of Global Integration’, Brookings Papers on Economic Activity 1:1–117

Scarpetta S, Hemmings P, Tressel T, Woo J (2002) The role of policy and institutions for productivity and firm dynamics: evidence from micro and industry data. Economic working paper (2002)15, OECD: Paris

Scarpetta S, Phillip H, Thierry T, Jaejoon W (2002) The role of policy and institutions for productivity and firm dynamics: evidence from micro and industry data, Working paper No. 329, Economics department, OECD

Schneider F, Frey BS (1985) Economic and political determinants of foreign direct investment. World Dev 13(2): 161–175

Tirole J (1988) The theory of industrial organization. M.I.T. Press, Cambridge, MA

Tybout J (2000) Manufacturing firms in developing countries: How well do they do and why? J Econ Lit XXXVIII: 11–44

Tybout J (2003) Plant- and firm-level evidence on the ‘new’ trade theories. In: Kwan Choi E, Harrigan J (eds) Handbook of international trade. Basil-Blackwell, Oxford

Wacziarg R, Wallack JS (2004) Trade liberalization and intersectoral labor movements. J Int Econ 64:411–439

Warzynski F (2002) Do Schumpeterian waves of creative destruction lead to higher productivity? Panel data evidence from Poland. LICOS discussion paper 120/2002

Wei S-J (2000) How taxing is corruption on international investors? Rev Econ Stat 82(1): 1–11

Yang OG (2004) Entry, exit and the dynamics of productivity growth in Chinese manufacturing industry. ESRC Centre for Business Research, University of Cambridge, Working paper no. 284

This page intentionally left blank

Chapter 2
Exit–Entry Dynamics: Case of the Manufacturing Sector in Jordan

Nesreen Barakat and Ibrahim Saif

2.1 Introduction

The manufacturing sector in Jordan is the most important sector in terms of its contribution to the GDP at constant basic prices. Relative to other sectors, the manufacturing sector has been increasing from 18.4 to 21.3% during the period of 2003–2006. This sector has been playing a very important role in boosting growth in Jordan over the last few years and therefore, it is pertinent to take a closer look and analyze the dynamics of entrants, survivors and exitors in the manufacturing sector.

Traditional analysis in this sector focuses on the main indicators. Very few studies went further to investigate the internal dynamics of the sector, especially the dynamics between new entrants, exitors and the survivors as this study intends to do.

This study will examine the structure of the manufacturing sector and its main parameters. The study will cover the following aspects of the manufacturing sector:

• Main salient features,

• Institutional arrangements,

• Characteristics of the entrants, exitors and survivors,

• Productivity and efficiency through adopting accounting and econometrics techniques to measure the main indicators and the determinants for entrants and exitors at a sectoral level,

• Conclusion and recommendations.

N. Barakat
To Excel Consulting & Associates, Amman, Jordan
e-mail: nesreen.barakat@to-excel.com

Ibrahim Saif ()
Centre for Strategic Studies, Jordan
e-mail: ibrasaif@gmail.com

K. Sekkat (ed.), Market Dynamics and Productivity in Developing Countries: Economic Reforms in the Middle East and North Africa, DOI 10.1007/978-1-4419-1037-0_2, © Springer Science + Business Media, LLC 2010

To achieve the above, the study is divided into the following sections:

• An overview of the manufacturing sector which will present a historical view of the manufacturing sector in Jordan. This would provide an insight into the extent of the impact of the entry and exit dynamics.

• The regulatory environment which governs the manufacturing sector in Jordan and an overview of the relevant legislations and regulations. The overview of the regulatory environment would provide insight into the legal aspects that affect the dynamics of entry and exit of manufacturing companies.

• An entry and exit dynamics analysis covering data collected during the period of 1999–2004 and an in-depth study of those firms within the manufacturing sector.

• A summary to provide conclusions stemming from the analysis of the entry and exit dynamics.

2.2 Overview of the Jordanian Economy

Jordan is a small country, with limited natural resources, situated in the heart of the Middle East. Debt, poverty, and unemployment are fundamental challenges for Jordan. Average economic growth during the period of 2000–2006 was 5.9% with the lowest rate of 4.2% registered in the year 2000 and the highest rate of 8.4% in 2004 (Department of Statistics 2004).

Jordan has been facing an accelerated pace of change brought about by globalization and trade liberalization with simultaneous opportunities and risks. Jordan had entered into an Association Agreement with the European Union (1997), signed a free trade agreement with the United States (2001) and successfully joined the World Trade Organization (2002).

Jordan’s economic performance has been impressive. During the period 1999–2001, Jordan’s annual real GDP growth averaged 4.2% compared to an average of 6.4% from 2005 to 2007 (Central Bank of Jordan 1999 – 2007). One of the reasons behind this growth has been the establishment of Qualifying Industrial Zones (QIZs) that allows for privileged access to the USA market. This has resulted in the expansion of exports in the garment and textile industry which amounted to 30.8% of total exports during 2006 (Department of Statistics 2007).

This strong economic performance was also driven by domestic consumer demand and supported by a booming construction and real estate sector. Additionally, high levels of remittances from the Gulf, and growth in the services sector boosted private consumption. Other economic indicators have followed similar patterns to the growth rate. The private sector has also assumed a larger role in the economy, especially with respect to investment, with public sector demand decreasing from about 30% of GDP in the late 1980s, to about 23.2% in 2003. Private nonresidential investment has increased from 4 to 7% of GDP in 1990 to 10% by 2003. At present, however, exports amount only to half the level of imports, and competitiveness is hampered by the limited capacity to export high-quality products.

The Jordanian economy continues to exhibit high levels of dependence on the oil economies of the Gulf. One of the major recent developments in the Jordanian economy is the impressive boom in foreign direct investment which has taken place since 2003. In 2002, FDI in Jordan was JD 52.8 million and in 2003, it increased almost sixfold to JD 309.3 million. FDI increased to JD 461.6 million in 2004, and then doubled in 2005 to JD 1.086 billion. In 2006, FDI doubled again, to JD 2.2 billion and accounted for over 20% of Jordan’s GDP. The proportion of FDI to local investment has also increased over the past few years. During the period 1995 – 2002, FDI in Jordan accounted for less than 1% of the total investment, with the exception of anomalous years, such as 2000, in which proceeds from privatization were recorded as foreign investment. FDI accounted for 3.9% of total investment in Jordan in 2002, 17% in 2003 and 18% in 2004, the most recent year for which gross fixed capital formation figures are available (Saif and DeBartolo 2007).

Remittances from Gulf countries continue to play an important role in the Jordanian economy. Remittances accounted for 17.6% of GDP in 2006, with Jordan ranking sixth in the world in terms of remittances as a proportion of GDP (World Bank 2007). The vast majority of these remittances are transferred from Gulf Cooperation Countries (GCC).

However, it is important to note that despite these reforms and some of the positive results that were realized, the underlying structure of the economy has not changed significantly. An analysis of recent growth trends suggests that Jordan’s service sector continues to play a significant role in the economy, accounting for over 70% of the GDP at basic prices.

Also, after more than 15 years of the economic adjustment process, the main conclusion regarding sectoral growth and output composition of the economy is that it has not changed substantially. The services sector is the fastest growing sector, owing mostly to the growth in banking and financial services.

The industrial sector in Jordan is divided into the manufacturing sector and the mining and quarrying sector. The mining and manufacturing sectors averaged nearly 24–26% during the period throughout 1989–2005. The manufacturing sector has been growing steadily, with its share in the GDP growing from 11% in 1989 to 17.9% in 2005. Recently, the contribution of mining and quarrying to GDP has increased because of high international prices for potash and phosphates, the main Jordanian products in this sector. While there has been growth in the manufacturing sector, boosted over the past few years by the establishment of QIZs, this does not constitute a major economic shift towards industrialization. The main indicators of the industrial sector for the period 2003–2006 are displayed in Table 2.1.

2.3 Manufacturing Sector

This section of the study presents data about the manufacturing sector and specifically those subsectors selected for the study. The data would give an indication on the magnitude of entrants, exitors and survivors.

The importance of the manufacturing sector to GDP at constant basic prices, relative to other sectors over the period 2003–2006 has been increasing, as displayed below, from 18.4% in 2003 to 21.3% in 2006¹. The manufacturing sector is the most important sector in terms of its contribution to GDP (Fig. 2.1).

Table 2.1 Main indicators of the industrial sector 2003–2006

Source: CBJ Annual Report 2006

Fig. 2.1 Relative importance of economic sectors to GDP at constant basic prices, 2003–2006

2.3.1 Labor Force in SMEs in the Manufacturing Sector²

Figures.2.2 and 2.3 shows the number of establishments in the manufacturing sector that had 1–19 employees, during the period 2000–2004 as well as the total number of employees those establishments employed. The number of establishments that had between 1 and 19 employees grew from 17,154 in 2000 to 19,778 in 2004. Consequently, the number of employees in these establishments grew from 61,732 to 68,823 during the period of 2000–2004.

Fig. 2.2 Establishments (1-19 employees)

Fig. 2.3 Establishments (1-19 employees)

2.3.1.1 Employment in the Manufacturing Sector

Figures 2.4 and 2.5 display the total number of employees and establishments for establishments with 20 employees or more. As the figure below indicates, the number of establishments with 20 employees or more is drastically lower than the total number of establishments that employ 1–19 workers. Accordingly, the number of establishments that had more than 20 employees amounted to only 929 in 2004. Furthermore, the number of employees in these establishments grew from 64,498 to 98,148 during the period of 2000–2004.

Fig. 2.4 Establishments (20 or more employees)

Fig. 2.5 Establishments (20 or more employees)

2.3.1.2 Employment by Establishment Size

Accordingly it is deducted that by 2004:

• Almost half of all employees in the manufacturing sector, 41.2%, were found in establishments that had (1–19) employees.

• Establishments that employed between 1 and 19 employees, as a percentage of the total number of establishments amounted to 95.5% of all manufacturing establishments. This figure fluctuated from 96.3% in 2000 to 96.7% in 2002 and to 95.5% in 2004.

• Around 4% of establishments in the manufacturing sector in Jordan employ more than 20 employees.

2.4 Entry Exit Dynamics

2.4.1 Methodology

This section will examine the determinants of firms’ entry and exit and their impacts. To guide the choice of the explanatory variables and the specific hypotheses to test, before going further, a clarification of the classification of firms should be presented. For a given year, if a firm was present in t – 1 but absent in t + 1, it will be classified as an exitor. If a firm was absent in t – 1 but present in t + 1, it will be classified as an entrant. A firm that was absent in t – 1 and t + 1 (i.e., it is only present on t) is both entrant and an exitor. It will be put in a “one-year-live” category. Finally, a firm that belongs to none of the three categories will be classified as a survivor. For comparability across sectors, entry and exit rates are defined with respect to the current year’s stock of establishments:

Overall, experience shows that there are marked contrasts among countries regarding the determinants and the impacts of entry in developing countries and the relationship to trade liberalization. The literature shows that these contrasts are related to the countries’ institutions, policies and the economic environment.

2.5 Determinants of Entry and Exit in Jordan

Over the period throughout 1999–2004, the survival rate among firms that employ 20 employees or more was estimated at nearly 90%. The exit rate reached its highest rate in the year 2000 at 10% and its lowest rate in 2003 at 5%. On the other hand, the highest entry rate was registered in 1999 at 4% and the lowest rate of 1% in 2004. Overall, for this group of firms, the entry rate was lower than the exit rate. Not much can be inferred from the pattern of entry and exit over the period 1999–2004, since exit and entry were not associated with major economic events in the country over the period under investigation.

The dynamics of exit and entry seems to result mainly from intra-sectoral reallocation rather than inter-sectoral reallocation. Those firms that exited the market went out of business more often than they shifted to other sectors where they could survive (Table 2.2).

Table 2.2 No. of Firms, Totals, Survivors, Entrants and Exitors

Source: Industrial Census and Survey, Department of Statistics (DOS), (Jordan)

The survival rate within firms that employ 10–20 employees was also high and averaged 82.5% over the period 1999–2004. The entry rate within this group was higher than in the larger group of firms with 20 employees or more. This might be the case because capital requirements and other institutional barriers are lower for firms to start this type of business. The exit rate was also high in this group and it seems to follow an upward trend averaging 10% and reaching as high as 15% in 2004. Small firms demonstrated more flexibility in entering and exiting the market, in contrast to the larger firms. This might be the case because costs associated with entering and exiting the market is lower for small firms. This also might reflect a lower degree of specialization and professionalism as far as small firms are concerned. Normally, small projects start without proper feasibility studies or a clear business model and hence their failure rate tends to be higher (Table 2.2.).

For the entire manufacturing sector, the survival rate averaged 86%, while the entry rate averaged 5%, and the exit rate registered around 8.5% (Table 2.2.).

Table 2.3a presents the number of firms exiting the manufacturing sector, by the number of their survival years. As can be seen below, 63% of all the companies that exited the manufacturing sector between the years 2000 and 2004 had been established for 10 years or more, depicting a trend towards the erosion of long-time established businesses. Very few firms exited the market after 1 or 2 years, and most of them seemed to have spent several years before exiting the market.

For firms with 20 employees or more, Table 2.3b shows that the average median size of employees for surviving firms was nearly double the size for entering and exiting firms for most of the years under investigation. Taking the mean exitors suggests that their employee size is much lower than for survivors or entrants. The mean size for entrants is similar and sometimes higher than for survivors. This means that in order to be able to compete and later to survive in the market, average employee size must be close to the mean of the survivors.

For smaller firms with 10–19 employees, as Table 2.3c depicts, the average employee size of the three groups of firms is similar, and it revolves around 12 employees. In all the cases, it was always less than 14 employees. This suggests that within small firms, size measured by the number of employees, does not influence the firms’ competitiveness and hence, its survival ability. Management and the field of work is more likely to influence performance, than the mere size of the firms. This also suggests that there are other factors such as capital intensity and access to market and funding that may decide the exit and entry rate within this group, as we illustrated in section one of this study when securing finance emerged as one of the major challenges facing the SMEs in Jordan. One may also consider technical and marketing techniques that these firms need, in order to remain in the market. At present, there is little institutional support in Jordan with the aim of overcoming problems facing these small firms.

Clearly firms that enter the market tend to expand in terms of the number of employees as more time elapses. Average size for firms that entered the market in 1999 has increased from 93 to 131 after 5 years in the market. The same applies to the median size, which increased from 40 to 70 employees. Firms that managed to survive began to expand their sizes, reflecting their increase in market share and their confidence. It also exhibits the fact that economies of scale are important for such firms. Increase in the number of employees was gradual and consistent over the 5 year period we are covering.

On the other hand, an identification of a consistent trend within firms that are exiting the market was not allocated as shown in Table 2.4. It seems that firms, in order to survive, were trying to hold on to their share in the market. There was no trend which showed that firms were dismissing their employees and consolidating their expenses as normally would be expected from troubled firms. However, for some years as in 2002, the number of employees was halved just 1 year before exiting the market, while in other years almost the same number of employees was maintained. This may indicate that there is rigidity in firing labor in Jordan. This is confirmed by the “World Bank 2008 Doing Business Report,” according to which Jordan suffers from a high difficulty of firing index (60) covering workers’ legal protections against dismissal, including the grounds permitted for dismissal and procedures for dismissal which indicates rigid regulation in this area. Moreover, firing costs measured as the cost of advance notice requirements, severance payments and penalties due when terminating a redundant worker amount to an equivalent of a 4-week salary (Table 2.5.).

Table 2.3 (a) Exitors by number of survival years. (b) Mean and median firm (20 employees and more). (c) Mean and median size firm (10–20 employees)

Source: Industrial Census and Survey, Department of Statistics (Jordan)

Table 2.4 Mean and Median Entrants Size (20 employees and more)

Source: Industrial Census and Survey, Department of Statistics (Jordan)

Table 2.5 Mean and Median Exitors Size (20 employees or more)

Source: Industrial Census and Survey, Department of Statistics (Jordan)

At a subsector level, exit, entry and survivor rates vary to some extent. In general, as noted earlier, the survivor rate is high. It went down to 50% in other manufacturing items (390) and manufacturing of rubber products. The survival ratio was also low in subsectors such as manufacturing of leather products ISIC (323) (0.67) and manufacturing of wood and wood products (ISIC 331) (0.70). Notably, the survival rate of the manufacture of wearing apparel, except footwear ISIC (322) went as low as 0.65. These subsectors have witnessed a major shift over the last few years and there were many new entrants that boosted productivity within the subsector. Hence, this is a case where productivity was increasing, while many firms exited the market. This supports the argument that competitive pressure may boost efficiency. The entry rate in this subsector was estimated at 8% which is the second highest ratio within the entire manufacturing sector. The same trend was almost repeated in the manufacturing of wood and wood products with the highest entry rate at 20%.

In the case of the remaining subsectors with high exit rates, the story was different. It seems that low productivity in few instances such as manufacturing of leather products has pushed some industries out of the market. In this particular case, there were certain arrangements between these firms and the government in place regarding securing public procurements for certain companies. Once these arrangements had been terminated, incumbent firms could no longer survive and there was no incentive for new firms to enter the market (Table 2.6.).

As witnessed earlier there is no clear pattern that emerges between entry and exit of firms. In order to determine if there is any significant relation between these two variables, an estimation of the correlation between entry and exit was conducted. Findings suggest that there is no significant correlation between exit and entry as Table 2.7 exhibits. The correlation between entry and exit was found to be negative, suggesting that as more firms exit the market, entrepreneurs would find it less attractive to enter into the same sector. This is a process of learning developed over time. However, since the correlation was not significant, this interpretation should be viewed with some caution.

The same exercise was repeated at a subsector level, as exhibited in Table 2.8. Again, no clear trend emerged concerning the correlation between exiting and entering firms at the subsector level. The sign of the coefficient was sometimes positive and sometimes negative, but not statistically significant. It seems that the decision to exit or enter any market depends on several factors, with the exit rate playing a marginal role in such a decision. This finding is consistent with the earlier conclusion regarding entry and exit by year, when no significant correlation was found between entry and exit for the entire period under investigation.

The impact of some correlation between entry and exit rates may vary across time. It could be negative during the early and late phases of a product’s life cycle but it is positive during the expansion phase of industries. For this reason, interaction terms of explanatory variables should be considered later in this study, since not much can be inferred from this analysis.

2.5.1 Productivity Analysis

The importance of analyzing exit–entry dynamics is particularly important to aspects related to productivity and the manner by which this dynamics helps in boosting productivity and hence competitiveness in certain markets. In the following section, we will investigate issues related to productivity and how it evolves over time.

Table 2.9 presents the average productivity measured as the value added per employee for entering, exiting and surviving companies in the manufacturing sector. As can be seen from the figures, over the period from 1999 to 2004, average productivity fluctuated for entrants with an upward tendency. It has significantly decreased for surviving enterprises. For the entire period from 1999 to 2004, average productivity of exitors amounted to JD 7,621 per employee, which is lower than JD 11,290 per employee for survivors, and JD 8,711 per employee for entrants.

Table 2.6 Survivors, Entry and Exit by Industry

Source: Industrial Census and Survey, Department of Statistics (Jordan)

Table 2.7 Correlations between entry and exit rates by year

Source: Authors’ calculations based on Industrial Census and Survey, Department of Statistics (Jordan)

Table 2.8 Correlations between entry and exit rates by industry

Source: Authors’ calculations, Industrial Census and Survey, Department of Statistics (Jordan)

Table 2.9 Average of productivity Average of productivity Year of entry Entrants Exitors Survivors

Source: Authors’ calculations, Industrial Census and Survey, Department of Statistics (Jordan)

These differences between entrants and survivors on one side and survivors and exitors on the other side indicate an increasing average productivity over the life cycle of a firm in the manufacturing sector, resulting from the advantages of an increasing company size and the competition of other enterprises in the manufacturing sector.

Table 2.10 Average labor productivity Average of productivity Isic2 4d Entrants Exitors Survivors

Source: Authors’ calculations, Industrial Census and Survey, Department of Statistics (Jordan)

However, findings so far suggest that liberalization by forcing the exit of the least efficient producers, contributes to productivity growth since as can be seen from the table average productivity among exitors is lower than that among, survivors and entrants.

Depending on the subsector within the manufacturing sector, Table 2.10 shows that development in the average productivity widely varied over the life cycle of the firms in the respective subsectors. Overall, the sectors of manufacture of Paper and Paper Products (341), Printing and Publishing (342), as well as Iron and Steel (371) exhibited the highest average productivity per employee.

Table 2.11a depicts the productivity changes of companies entering the manufacturing sector from 1999 to 2003 up to the year 2004 according to subsector. As for firms entering in 1999, companies active in the subsectors of manufacture of wearing apparel, excluding Footwear (322), Printing and Publishing (342) and Iron and Steel (371) displayed substantial increases in productivity. These developments are an expression of the advantages gained through an increasing company size in these subsectors, e.g., economies of scale. In addition, technological progress and competitive pressure are factors leading to a boost in subsector productivity. As for the manufacture of paper and paper products, it can be noted that firms in this subsector displayed the highest productivity of all subsectors listed, but faced a stagnant productivity level over the following years. This picture slightly changed for the subsector of paper and paper products in 2000, as firms were able to register small productivity gains over the subsequent years. Furthermore, the subsector of food products (311) showed markedly higher levels of productivity for companies that entered the subsector in 2001 than for companies which entered the subsector in other years. Moreover, in 2001, entering firms in the subsector of the manufacture of wood and wood products, excluding furniture (331), exhibited a very low productivity of JD 1,000 per employee, which remained stagnant during the following years.

Table 2.11 (a) Productivity change of entrants (average). (b) Average labour productivity

Source: Authors’ calculations, Industrial Census and Survey, Department of Statistics (Jordan)

It should also be noted that special treatment was given to the manufacture of wearing apparel under the QIZs arrangements so that they were exempted from income taxes, and the government facilitated their work by ensuring good infrastructure and access to the US market (saif 2005).

Overall, there was some productivity gain for entrants, which varied across sub-sectors, as the table exhibits for the period under investigation, productivity improved over time for new entrants.

Regarding companies that entered the manufacturing sector in the years from 1999 to 2002, it can be seen from Table 2.11a and 2.11b that the average productivity per employee of those companies, showed an overall increase over the following years. This rise can be traced back to the productivity gains received from the advantages of an increasing firm size as well as an upward sloping learning curve. It seems also that competition from other companies forced the manufacturing companies to be more productive and efficient. As more time elapsed and firms managed to pass the critical first few years, their productivity exhibited an upward trend.

In order to complete the analysis, examination of productivity development of exitors was conducted. As can be seen from the below, there was a significant decline in terms of productivity of exitors just 1 year before they exited the market.

The overall trend in productivity was negative for most of the years preceding the firms’ exit from the market. Productivity deterioration was clear just 1 year before exiting the market which suggests that firms spend some years striving to survive. They only exit when they can no longer stay in the market.

On a subsector level, the subsector of food products (311) exhibited the biggest average declines in productivity for firms leaving the subsector throughout 1999–2002. Firms exiting this subsector in 2002 recorded an average productivity loss of JD 67,616 over the past 4 years which accounted for the greatest average decrease of all sectors during the years 1999–2003. None of the subsectors registered an increase in average productivity for more than 1 year during the covered time period. This means that exiting companies did not manage to continuously reach productivity increases on average over their last years of business in any of the covered subsectors, which indicates weak, and even decreasing, competitiveness of exiting companies in those subsectors.

Overall, the highest loss in productivity for the, entire sector occurs in the year that precedes the year of exiting the market. It is also clear that lower productivity is associated with other variables such as profitability, which will be analysed later together with the determinants of exiting in the econometric section.

As expected, firms exiting the manufacturing sector during 2000 the period of 2004 showed a decreasing average productivity per employee in the last year before their exit. Table 2.12a and 2.12b depicts a considerable decrease in average productivity for exiting companies when compared to the productivity level 5 years before their market exit. Overall, firms that exited in 2001 possessed an average productivity of JD 7,600 per employee compared with nearly JD 9,000, 5 years before they exited the market. Moreover, productivity seems to be declining over time.

Table 2.12 (a) Productivity change of exitors (average). (b) Productivity Change of Exitor

Source: Authors’ calculations, Industrial Census and Survey, Department of Statistics (Jordan)

2.5.2 Analysis of Productivity Change

The analysis of the productivity change for each group (entrants, exitors, and survivors) has so far been conducted separately. The picture will become clearer if there is a combination of the three groups with a decomposition of productivity changes in order to determine which group contributes more to productivity change.

Labor productivity will be decomposed into the contribution of “internal restructuring” (i.e., productivity growth within the surviving establishments, or the “within” effect), changes in the market shares of the survivors (i.e., productivity grows further if the shares of higher productivity establishments increase, or the “between” effect) and the contribution of entry and exit.

2.5.3 Productivity Change for Entrants, Survivors and Exitors

As Table 2.13a and 2.13b shows, productivity varied across subsectors, with some subsectors achieving strong growth while others suffered from productivity losses. However, the general findings suggest that over the period 2000–2004 positive productivity growth was registered.

Sectors such as wearing apparel, non metallic industry as well as iron and steel scored high levels of productivity growth. Improvements in these sectors can be attributed to favorable economic conditions. For example, the textile industry performance has been boosted by the QIZs and exports to the US market. In the iron and steel industry, the strong growth can be attributed to the expansion in the construction sector.

Table 2.13 (a) Productivity change for entrants, survivors and exitors. (b) Productivity change for entrants, survivors and exitors

Source: Authors’ calculations Industrial Census and Survey, Department of Statistics (Jordan)

On the other hand, there are few sectors that experienced negative productivity growth, namely, the food industry, leather products, manufacturing metal and pottery. These sectors are dominated by a small and medium scale industry and lack economies of scale. Operational costs for these industries are not high, hence they managed to stay in the market even with low productivity rates.

The main concern in this section is how much each group – that is the new entrants, survivors and exitors – contributes to enhancing or reducing productivity. From the table above and utilizing the accounting methodology described earlier, it can be seen that clearly the incumbent plants are on average more productive than entrants and exitors over the period 2000–2004. This is the “within effect” which refers to factors internal to firms such as organizational change, and the introduction of new technologies.

New entrants play a minor role in enhancing productivity. In some subsectors they are less productive than the exiting plants in the early years, but become more productive in the later years. They also exhibit the highest positive productivity change during the early years.

This was the case in most of the subsectors under investigation. However, the main contribution to the achieved productivity came from survivors (within effect) who seem to move towards the production frontier as they spent more time in the market. At this stage, one cannot determine the reasons behind these dynamics, because institutional arrangements concerning all sectors are similar.

Productivity dynamics also varied over time and across sectors. During some years when there was a high rate of entry into specific subsectors, new entrants seemed to contribute positively to the total growth. In the absence of new entrants or exitors, the total productivity change was attributed to the survivors.

2.5.4 Econometric Analysis

2.5.4.1 Determinants of Growth

The above approach is used only for comparison purposes. Being strictly an accounting approach, it does not measure precisely the impact of entry and exit on productivity growth. On the one hand, the distinction between the 4 effects (“within”, “between” and entry and exit) is not necessarily clear since entry and exit can also induce “internal restructuring” by survivors and reallocation of market shares among firms. On the other hand, “present” competition (i.e., abstracting from entry and exit) can also induce “internal restructuring” and reallocation of market shares. To adequately assess the impact of entry and exit on productivity, the following regression was calculated³:

Δlog (Y_it) = η₀ + η₁ × Δlog (K_it) + η₂ × Δlog (L _it) + η₃ × log (C it) + η₄ × (9) + η₅ × log (EX it) + ξ _it

Aggregated firm level data is used.

It emerges from the results that rate output growth is highly dependent on labor and capital. Both coefficients are significant and they hold the expected positive signs suggesting that growth to some extent is resource based rather than productivity driven. In order to emphasize the impact of entry and exit on productivity changes, several regressions are utilized; the first run using present value of the entry and exit demonstrates that entry rate holds negative sign. This might be so because its influence is more likely to appear later. While exit rate hold negative signs but was not statistically significant, it seems that under this model both exit and entry play marginal role in determining the level of output (Table 2.14.).

High concentration, which resembles the market structure, seems to contribute to an increase in survival rate. This simply means that high concentration rate is associated with high output growth. The same logic applies as far as the tariff rate is concerned. Higher tariff rates that represent, to some extent, the state of competition, seems to be attractive and conducive to business survival. This is consistent with earlier findings (Saif and Barakat 2008) which demonstrate that the manufacturing sector in Jordan is highly concentrated and incumbent industries are keen to maintain their interests by sustaining high level of tariffs and through highly concentrated markets.

Table 2.14 Regression results for the following model: Δlog (Y_it) = η₀ + η₁ × Δlog (K_it) + η₂ × Δlog (L_it) + η₃ × log (Cit) + η₄ × log (EN_it) +(9) η₆ × log (EX_it) + ξ _it

Source: Authors’ calculations, Industrial Census and Survey, Department of Statistics (Jordan) Values in parentheses indicate t -statistic
***, **, and * denote that the coefficient is significant at 1, 5, and 10 respectively
Δlog (Y_it) = change in the natural logarithm of outputs, Δ log (K_it) = change in the natural logarithm of capital, Δ log (L_it) = change in the natural logarithm of labor, log (CON_it) = the natural logarithm of concentration ratio, (Tarif_it) = tariff rate, log (EN_it) = the natural logarithm of entry rate, and log (EX_it) = the natural logarithm of exit rate

Overall, variables explain more than 50% of variations in the dependent variables and the main message through this is that output growth is highly dependent on labor and capital. That growth is resource based more than productivity driven.

Utilizing lag variables has not changed the conclusion; most of the variables held the same signs with the same level of significance, lag variables for entry were used, or whether, variables were entered to capture the nonobservable effects, such as the mark-up.

2.5.4.2 Analysis of Entry

This section aims at analyzing determinants of entry at firm level. In order to do so, the following equation was utilized. It included both observable and nonobservable factors⁴.

To include specific variables about industry characteristics and institutional environment, the above (2.3) can be rewritten as follows:

The regression utilizing the pooled data methods was used in order to identify the main determinants of the entry rate over the period under investigation. Though the correlation coefficient was low, however, most of the independent variables found to be significant and holding the expected sign. The most significant factor that seems to influence the decision to enter any market was profit margin. The more profitable the sector is. the more attractive it is for new comers. Similarly, entry rate was found to be positively correlated with the average values added, which is the measure for productivity (Table 2.15.).

These findings suggest that decision to enter the market is rational on the side of the new comers since it is consistent with the conventional wisdom that places profitability and high productivity as the main drivers of new investments. High productivity implicitly means that the overall business environment is conducive and also the available labor force in general is skillful and adequate.

As far as institutional barriers are concerned, it was revealed that concentration for example found to be negatively correlated with entry rate. In other words, highly concentrated sectors seem to erect barriers to new entry, or it may be the case that these sectors are closed to new entrants due to the nature of some of these sectors. Therefore, even while these sectors are highly profitable and secured, in practice, it is extremely difficult to enter such sectors. In other less concentrated sectors, other unobservable factors such as profitability, exposure to competition and productivity, were found to play a more significant role.

Table 2.15 Regression results for the following model: Entry rate_{i, t} = α₀ + α ₁ + average size of entrants_{i, t} + α ₂ + profit margin_i,t + α ₄ × growth rate_i,t + α ₅ + average productivity_i,t + α ₆ × wage rate_i,t + α ₇ × capital intensity_i,t + α ₈* Tarif_i,t + α ₉ × exit rate_i,t + μ_i,t

Source: Authors’ calculations, Industrial Census and Survey, Department of Statistics (Jordan)
Values in parentheses indicate t -statistic
***, **, and * denote that the coefficient is significant at 1, 5, and 10 respectively

The other factor that negatively influences the decision to enter is the wage rate. Higher wage rates, which mean a higher cost, seems to be an important factor before entering a new market. This is not surprising in a country that relies heavily on imported labor and where there is a serious problem regarding finding the right caliber to fill in new vacancies. This is serious problem in the manufacturing sector which strives to survive an era of trade liberalization at a time when surrounding potential competitors, Syria, Egypt, are enjoying a much lower wage rate. It emerges in the first section that there are few restrictions regarding acquiring foreign laborers.

Along with the wage, higher capital intensity, where capital is mostly imported, holds a negative sign suggesting that higher capital intensity does not encourage new comers. Capital intensive industry means higher sunk costs and hence, decision to enter becomes more difficult as our results have suggested.

Other institutional barriers such as investment regulations, labor regulations and other trade and nontrade barriers (discussed in Sect. 2.1) seem to be less significant in deciding whether to enter the market or not. Tariff rates were utilized as one of the factors that might have an influence on the decision to enter, but the coefficient turns out to be significant. Other institutional variables that have been discussed earlier do not discriminate against any sector and hence, it is expected that their impact across sectors is similar to a great extent.

Overall, most of the variables hold the expected sign and decision to enter a certain market is based on a combination of factors that is consistent with the traditional threats associated with doing business, namely, profitability, cost of labor and capital, and productivity rather than the institutional or the exit rate.

2.5.4.3 Analysis of Exit

The other side of the story is related to firm that exit the market. This section analyse determinant of exit through the following equation:

To include specific variables about industry characteristics and institutional environment, the above equation can be rewritten as follows:

Variables employed in the regression explain nearly 35% of the variation in the dependent variables as indicated by the correlation coefficient. Profit margin seems to be the most important factor in deciding whether to stay or to exit from the market. Coefficient of profit margin negatively and significantly correlates with the exit rate. Interestingly, it also seems that lower productivity also pushes inefficient firms out of the market, though the coefficient of average value added for exitors, which holds the expected sign, was not statistically insignificant (Table 2.16.).

Wage rate, which is a major concern, holds the expected negative sign and was significant. This suggests that higher wage rate, especially when associated with low value added growth, may significantly contribute to push firms out of the market.

This is quite consistent with the findings in the accounting exercise when it was found that survivors are the most productive, while exitors are the least, in terms of labor productivity.

Table 2.16 Regression results for the following model: Exit rate_i,t = β₀ + β₁ × average size of exitors_i,t + β₂ × average age of exitors_i,t + β₃ × profit margin_i,t + β₄ × concentration_i,t + β₅ × growth_i,t + β₆ × average productivity_i,t + β₇ × wage rate_i,t + β₈ × capital intensity_i,t + β₉ × tarif_i,t + β₁₀ × entry rate_i,t + υ_i,t

Source: Authors’ calculations, Industrial Census and Survey, Department of Statistics (Jordan)
Values in parentheses indicate t-statistic
***, **, and * denote that the coefficient is significant at 1, 5, and 10 respectively

It is still the case that there is no significant relationship between exitors and entrants. Clearly, the decision to exit from a certain market has little to do with the number of entrants. Other unobservable factors seem to be more significant, especially those related to direct cost, such as wage and profit.

Average size of the firms does not seem to be significant in deciding to exit the market. Exitors vary in terms of size and could be small firms belonging to the small and medium scale industries or could be large firms that belong to a different category. What matters, is the financial performance and the level of productivity.

Also, capital intensity seems to play a minor role in deciding whether to stay or to exit from the market. What is puzzling though, is the positive and significant relationship with the tariff rate which was insignificant as far as the entry rate is concerned. While concentration rate was insignificant as far as the exit rate is concerned, seems that even protected industries may exit market when other factors are not contributing to the survival of the firms.

Other institutional factors are less significant and it seems that the main factors deciding exit and entry rate are related to firm’s behavior and its management rather than the external factors (business environment) or other political factors concerning the overall stability.

Overall growth, entry and exit in the manufacturing sector seem to follow certain logic and behaved in a predicted manner. Results found to be consistent with the conventional wisdom as far as conducting business is concerned. There were some data limitations concerning firms that employ between 10 and 19 employees. Level data of firms is not available because annual surveys do not include all these companies. In order to estimate entry and exit within this group, a number of assumptions were made.

It is for these reasons, that we limited the regression analysis to firms that employed 20 or more employees, while we utilized data regarding firms 10–19 in the descriptive side of the analysis.

2.6 Conclusion

The manufacturing sector is an important sector in the Jordanian economy, contributing both to creating jobs and generating added value. The structure of the sector is such that it comprises many small firms that contribute little to the overall output and value added.

Reviewing the institutional arrangements and regulations governing the sector reveals that, there are some impediments that hinder the sector’s performance. However, there are barely any measures that distinguish among sectors. Hence, variation in performance is sector-specific and the dynamic of entry and exit relates more to the structure of the sectors more than anything else.

Analyzing the market dynamics of entry and exit revealed some important facts. Entry and exit over the period 1999–2004 was stable and no pattern was inferred since exit and entry were not associated with major economic events in the country over the period under investigation.

Very few firms exited the market after 1 or 2 years. Most of them seem to spend several years before exiting the market.

The mean size, measured by the number of employees, for entrants is similar and sometimes found to be higher than for the survivors. This means that in order to be able to compete and later to survive in the market, average employee size must be close to the mean of the survivors.

Firms, which managed to survive, began to expand their size, reflecting their rising share in the market and their confidence. This also suggests that economies of scale are important for such firms. Increase in the number of employees was gradual and consistent over the 5 year period that was covered. At a subsector level exit, entry and survivor rates vary to some extent.

There is no clear pattern between entry and exit. An estimation of the correlation between entry and exit was implemented in order to determine if there was any significant relation between these two variables. Findings suggest that there were no significant correlation between exit and entry. The correlation between entry and exit was found to be negative suggesting that as more firms exit the market, entrepreneurs would find it less attractive to enter into the same sector.

Regarding productivity issues, average productivity per employee for new entrants increased over the following years. This rise can be traced back to the productivity gains received from the advantages of an increasing firm size as well as an upward sloping learning curve. It seems also that competition from other companies forced the manufacturing companies to be more productive and efficient. As more time elapsed and firms managed to pass the critical first few years, their productivity exhibited an upward trend.

The overall trend in productivity was negative for most of the years preceding the firms’ exit from the market. Productivity deterioration was clear just 1 year before exiting the market which suggests that firms spend some years striving to survive. They only exit when they can no longer stay in the market.

This was the case in most of the subsectors under investigation. However, the main contribution to the achieved productivity came from survivors (within effect) who seemed to move towards the production frontier as they spent more time in the market.

It emerged out of the regression, that growth is related to factors such as capital and labor. Entry rate was found to be positively contributing to growth, while exit was less significant. For entry and exit, it seems that the most decisive factors are those related to profitability and wage level more than to the overall institutional barriers or business environment. Moreover firm size concentration and tariff were found to be less significant.

Regarding entry determinants, it emerges that the decision to enter a certain market depends on a combination of factors that is consistent with the traditional threats associated with doing business, namely, profitability, cost of labor and capital, and productivity, rather than the institutional or the exit rate. The same applies to the dynamics of the exit rate when it was found that institutional factors are less significant, and it seems that the main factors deciding exit and entry rate are related more to the firm’s behavior and its management than the external factors (business environment) or other political factor concerning the overall stability.

Finally, findings suggest that there was no significant relationship between exitors and entrants. Clearly, the decision to exit from a certain market has little to do with the number of entrants. Other unobservable factors seem to be more significant especially those related to direct cost such wage and profit.

2.7 Notes

1. CBJ Annual Report, preliminary estimates for 2004–2006.

2. For the purposes of this study, SMEs are those establishments that have between 1 and 19 employees.

3. The number of observation = 55 (11 industry × 4 years). The sample denotes cross section and time series.

4. The sample is a cross-section and times series. Estimation method is pooled data regression.

References

Central Bank of Jordan (1999–2007) Monthly statistical bulletin, several bulletins

Central Bank of Jordan Annual Report (2006) Accessed http://www.cbj.gov.jo/uploads/chapter2/pdf

Department of Statistics (2004) Press release

Department of Statistics (2007) Statistical yearbook 2006

Saif I, DeBartolo (2007) The impact of displaced Iraqis on the Jordanian economy. Center for Strategic Studies, University of Jordan, Amman

Saif I (2005) The socio economic implications of the qualified industrial zones in Jordan. Center for Strategic Studies, University of Jordan

Saif I, Barakat N (2008) Competition and competition policy and economic policy and economic efficiency in Jordan. In Sekkat Kh (ed) Competition and efficiency in the Arab world, Palgrave Macmillan, New York

World Bank (2007) Resolving Jordan’s labor market paradox of concurrent economic growth and high unemployment

This page intentionally left blank

Chapter 3
Industrial Dynamics and Productivity in Morocco: A Quantitative Assessment

Lahcen Achy and Khalid Sekkat

3.1 Introduction

Since the early 1980s, the Moroccan authorities have decided to switch from a model of a relatively closed import substitution economy with a large public sector to free trade and a market oriented economy. Price liberalization policy has been one of the main components of the Structural Adjustment program (SAP). Most prices of goods and services were freed in order to promote competition and allow market mechanisms to have a greater role in allocating and pricing factors, goods and services. The country speeded up the process of economic reforms during the 1990s. The new strategy is thought to set the economy on a path of higher efficiency as a result of the intense competition, and thus foster growth and development.

Twenty-five years after the starting of the comprehensive program of economic reforms, the economy is, however, still highly specialized in comparison to many other developing countries (LDCs) and this specialization is increasing over time. Few industries (Textile, wearing apparel, food products and chemicals) represent around 70% of total manufacturing employment. In these industries, productivity is decreasing markedly (except for chemicals) and mark-ups are relatively high. The issue is the most worrying, as the lack of competitive pressure is found to be associated with low efficiency and concerns the most important industries in terms of employment. There seems, therefore, to be a clear issue of industry dynamisms in the Moroccan manufacturing sector (Achy and Sekkat 2008).

Comparing these observations to recent empirical evidence across a number of LDCs shows that the persistence of inefficiency, despite reforms, is specific to few of them (including Morocco) and is, therefore, worrying. The analysis of a set of 25 liberalization episodes by Wacziarg and Wallack (2004) shows that while liberalization induced little inter-sectoral labor reallocation (i.e., decrease in specialization), it led to important intra-sectoral reallocations towards more productive firms.¹ Aw et al. (2000) find that exposure to trade forces the exit of the least efficient producers in Korea and Taiwan, while Pavcnik (2002) finds that market share reallocations contributed significantly to productivity growth following trade liberalization in Chile. Hoekman and Winters (2005) report that studies using firm-level data conclude that major impacts of trade reforms are natural selection among firms and reductions in X-inefficiency: less efficient firms in a sector are forced to downsize, improve efficiency or exit, while more productive (efficient) firms expand their market shares.

L. Achy ()
INSEA, Rabat, Morocco
e-mail: lachy@ulb.ac.be

K. Sekkat (ed.), Market Dynamics and Productivity in Developing Countries: Economic Reforms in the Middle East and North Africa, DOI 10.1007/978-1-4419-1037-0_3, © Springer Science + Business Media, LLC 2010

The fact that the process of trade liberalization in Morocco has not induced a similar productivity gain as in other emerging countries, suggests that other reforms such as domestic regulation also play an important role for intra-sectoral reallocation. For instance, Revenga (1997) suggests that the small market responses found in developing countries may reflect restrictive labor market regulation. Harrison and Hanson (1999) argue that imperfect product markets may also be a relevant factor underlying the observed limited impacts of trade liberalization. Borjas and Ramey (1995) suggest that capital or financial market distortions or inefficiencies will affect the ability of firms to expand or to enter. These variables may even be more important than the labor market. Finally, studies on the determinants of investment (e.g., Mauro 1995; Wei 2000) suggest that the institutional framework of a country could also have marked impacts on entry and exit.

The present chapter examines whether the above documented processes by which economic liberalization affects productivity have taken place in Morocco. It addresses three specific questions. First, how much important is the dynamic of firms’ entry and exit in Morocco? Second, what are the economic, policy and institutional factors that affect such a dynamic? Third, does the process of firms’ entry and exit improve the manufacturing sector’s productivity?

The rest of the chapter is divided into five sections. Section 3.2 provides a deep discussion of the main changes in regulations and policies that might affect dynamic of firms’ entry and exit in Morocco. Section 3.3 presents our dataset and its main characteristics. Section 3.4 analyses the determinants of firms’ entry and exit in Morocco while Sect. 3.5 is devoted to the impact of firms’ entry and exit on productivity. Finally, Sect. 3.6 concludes.

3.2 The Institutional Environment for Business

3.2.1 Overview

There is a vast body of economic literature showing that business regulations are essential for investment, trade, and ultimately economic growth. Business regulations cover a whole range of laws, formal and informal rules governing business conduct in the economy. These include rules for entry in and exit from business, labor and tax regulations, degree of contract enforcement, cost and efficiency of dispute settlement mechanisms and so forth.

Over the last two decades, Moroccan authorities have made significant efforts to reform and streamline various business regulations, in order to create an environment conducive to investment and growth. These efforts often took place as a component of foreign investment attractiveness package and more recently under the pressure exerted by “international rankings” such as “doing business” ranking conducted by the World Bank.

The implementation of regulatory reforms faced resistance and has been sometimes politically contentious because of their impact on the initial distribution of power and rents among various administrative departments. Still, many improvements have been recorded, such as the reduction in the number of procedures required to start a business, availability of information on services provided by administrative departments for firms, simplification of customs procedures and the diminution of delays in trading across borders.

Despite these efforts, various indicators reveal that business environment in Morocco continues to represent a serious handicap for investment, both domestic and foreign.

Part of the explanation is related to the restrictive nature of legal and regulatory provisions. The cost of firing an employee, for instance, was found to be too high under the Moroccan labor code compared to international standards (World Bank 2006). Corporate tax, although reduced in 2008 from 35 to 30%, is still above its level in Europe or in other competing countries. The average corporate tax rate in Eastern Europe and Central Asia is 20% (Pricewaterhouse Coopers 2008). Similarly, Turkey, which has a Free Trade Agreement with Morocco, has lowered its corporate tax rate from 30 to 20 since 2006. Marginal taxes on personal income, which reach 42% in Morocco, are also very high and represent according to the World Bank (2006) a severe constraint to employing human capital in Morocco.

The second part of the explanation is related to the degree of enforcement of the existing regulations. Regardless of the quality of business regulations, what really matter for investors are the barriers and constraints they effectively face on the ground. For instance, competition law was expected to enter into force in July 2001; however, so far there has been no effective implementation of its provisions. In the same vein, commercial courts have been created, but judges have not been adequately trained to settle disputes and enforce contracts.

The rest of this section deals with various business regulations relevant to the purpose of entry and exit dynamics in the Moroccan manufacturing sector in Morocco.

3.2.2 Corporate Regulations

Business in Morocco can be conducted under various legal forms. In addition to individual companies, two additional forms are frequently encountered: “Joint stock companies²” and “Limited liability companies (LLC).³”

Joint Stock Companies (SA) must have a minimum of five shareholders and a minimum capital of MAD 300,000 (roughly US$ 37,000 in 2007), if the company is non-listed on the stock market and MAD 3 millions (US$ 370,000) otherwise. The capital must be entirely subscribed and at least 25% paid up. The rest must be paid up within 3 years from the date of its registration. The company can be 100% held by foreign companies or individuals.

Regarding LLC, they can be created even with a single shareholder, and a minimum capital of MAD 10,000 (US$1250). They may be 100% owned by foreign companies or individuals.

A majority of manufacturing firms in Morocco (58%) are LLC. Individual firms represent 24% and joint stock companies 15%. The rest of the firms (3%) have other marginal legal forms such as cooperatives.

The legal structure of an LLC is rather simple and its management easier. It offers the possibility of incorporating an individual LLC which is new under the Moroccan law. This legal status is also flexible, as an LLC can be transformed into a joint stock company if shareholders need a more sophisticated structure. Under the LLC, the manager’s liability is civil and criminal, but penalties are less severe compared to the directors and the president of the board in joint stock companies.

3.2.3 Labor Market Regulations

Restrictive labor market regulations are often highlighted among major impediments to business growth and may lead to a large informal manufacturing sector. The degree of restrictiveness of the labor market regulations is examined by reviewing the main provisions of the Moroccan labor code adopted in 2003 and entered into force in 2004.

The Moroccan labor code stipulates that labor contract may be oral or written. The term of the contract can be fixed or indefinite. However, companies can enter into fixed term contracts for 1 year, renewable only once, after which the contract is converted by law into an indefinite period contract. Companies must register all their employees with the social security department (CNSS) from the beginning of their contract. The Moroccan labor code sets a minimum monthly wage (referred to as SMIG) that is revised regularly to preserve purchasing power of workers. However, minimum wage in Morocco seems to evolve very irregularly, and without any close link with the consumer price index variation or labor market conditions. For instance, minimum wages were frozen from 1996 to July 2000, and increased by as much as 10% in July 2000. This irregular pattern of the minimum wage in Morocco can be attributed to the fact that, in practice, it is highly driven by political cycles, and trade unions’ pressures.

As far as layoff of employees is concerned, it can only occur in the case of serious offense. Firing costs per year are paid on the basis of seniority. Dismissal without well founded cause may result in the payment of damages. The labor code has fixed the indemnity to 1.5 months of salary per year worked with a maximum of 36 months. Finally, employers wishing to close their business must first receive an authorization from the local authority represented by the governor.

3.2.4 Trade Policy and Regulations

Protective trade policy implemented before 1983 created anti-export bias and generated rent-seeking situations in the manufacturing sector. Frequent and unpredictable tariff revisions were mainly driven by government financial and balance of payment distresses. They reflect discretionary power enjoyed by public authorities in setting trade policy. Trade liberalization was a key component of the SAP in which Morocco embarked in 1983. The Foreign Trade Law promulgated in 1992 (law 13/89) consolidated trade liberalization process and adapted Morocco’s legislation to principles and provisions of the General Agreement on Tariffs and Trade (GATT).

The general trend of Morocco’s trade policy is one of liberalization according to its WTO commitments. Morocco has therefore kept on reducing its external tariffs. Yet, the observed level of tariffs remains high and even seems to have increased over the late 1990s. This was the outcome of converting quantitative restrictions into equivalent tariffs in 1996, and the inclusion of the fiscal levy on imports (PFI) into tariff rates in 2000 (WTO 2003).

Morocco grants preferential treatment for imports originating in countries members of regional or bilateral trade agreements to which it takes part, on a reciprocal basis. These are for example, the United Maghreb Arab countries (UMA), the Arab Free-trade Area (GAFTA), the Association Agreement with the EU, the Free-trade Agreement with EFTA, and bilateral free-trade agreements with countries in the region such as Egypt, Jordan and Tunisia. Morocco also grants such treatment under trade and tariff agreements and the Global System of Trade Preferences (GSTP). However, to be eligible for preferential treatment, goods must be covered by a certificate of origin. They also must be directly shipped to Morocco from the country of origin.

Moroccan standards and technical specifications are generally optional. However, they become compulsory for reasons related to health, security, hygiene and environmental protection. In such cases, they apply without discrimination to imported and domestic goods. Imports are subject to a conformity certificate issued by the Ministry of trade and industry. Mandatory standards currently apply to some iron and steel products, gas products, electrical equipment, textile products, domestic appliances, and toys. Morocco has not signed any mutual recognition agreement and does not automatically accept foreign certification; the current regulatory framework, which is under revision, may address this issue.

Table 3.1. reports data on tariff rates computed as simple averages at two digit level. It indicates the average tariff for the whole manufacturing sector declined from roughly 68% in the early nineties to 27.6% in 2005. The trend recorded by the level of protection in Morocco reflects the process of tariff dismantling implemented under multilateral commitments (WTO) as reported rates are granted on an MFN basis. Under regional and bilateral agreements (with the EU, GAFTA members or Agadir members) tariffs are even lower.

Table 3.1 Morocco’s MFN simple average tariffs

Source: Data compiled by the authors from TRAINS Database, UNCTAD 2006

However, the average tariff rate for the manufacturing sector dissimulates large differences among industries. The highest tariffs are recorded in labor intensive industries, such as wearing and apparel (49.8%), footwear (44.53%), textiles (41.1%) or in food and beverages industries (47.36%), which make these industries less exposed to import competition. On the other hand, capital intensive and technology-based industries, such as “medical and other precision equipments industry,” “machinery equipment” tend to benefit from substantially lower rates (less than 10%).

3.2.5 Exchange Rate Policy

Morocco reformed its foreign exchange system in the late 1980s and early 1990s by, gradually, unifying and liberalizing foreign exchange markets. Morocco has established current account convertibility since 1993, but still imposes restrictions on capital account convertibility for residents, and fewer restrictions are imposed on inflows than on outflows. Non residents are allowed to hold accounts in foreign and domestic currencies, but residents’ accounts are subject to more regulation than non-residents’ accounts.

The Central bank law that entered into force in February 2006 contributed to clarify the roles of the Bank Al Maghrib (BAM) in relation to those of the Ministry of Finance (MOF) in the area of exchange rate policy. As a consequence, the MOF and the BAM are holding joint monthly meetings to investigate issues related to the exchange rate regime, the deepening of exchange and financial markets, and foreign exchange regulations.

The main objective of exchange rate policy in Morocco has been to stabilize the Real Effective Exchange Rate (REER) vis-à-vis the main trading partners. This objective is sought, in nominal term, by pegging the MAD to a basket of currencies and, in real terms, via budgetary and monetary policies aiming at curbing inflation. In the past, the exchange rate was determined on the basis of a currency basket including Morocco’s major trading partners, weighted by their importance in Morocco’s foreign trade and the pattern of currencies settlement. The composition was kept secret by the central bank. More recently, more transparency has been introduced in the exchange rate setting mechanism by disclosing the composition of the currency basket. The latter is 80% composed by the euro and 20 by the US dollar.

Our estimates of the real effective exchange rate suggest that the Moroccan Dirham exhibited some level of overvaluation over the nineties, particularly starting from 1994. The level of overvaluation has even increased following the adoption of a single currency in Europe. The Moroccan government faced a critical issue since the Moroccan Dirham appeared to be undervalued against the dollar, but overvalued compared to the Euro. In April 2001, the currency composition of the basket was revised by attributing more weight to the Euro. This revision has not entirely eliminated the overvaluation but reduced its magnitude as shown in the Table 3.2.

3.2.6 Competition Policy

Morocco adopted a competition law in July 2001. The law applies to all natural persons or corporations whether their headquarters are established or not in Morocco, provided, that their operations or behaviors have an effect on competition in the Moroccan market or in a substantial part of it. The law covers all activities of production, distribution and services. It also applies to public entities, which engage in commercial activities as economic agents, but does not apply to the sovereign acts of the state itself. The competition law defines provisions that govern freedom of prices and rules to protect competition. It determines the rules to deal with anti-competitive practices, abuse of dominant position and economic concentrations. However, there are, exemptions when anti-competitive practices are the result of implementing legal or regulatory provisions, or in case their authors can justify that their effect on economic progress are sufficiently high to compensate for restrictions on competition, and that they allow consumers to get a fair share of profit, without eliminating competition in a substantial part of the market.

Table 3.2 Bilateral and reel effective exchange rate

Source: Authors’ computation from exchange rate data provided by IMF and trade data provided by Foreign exchange office

There are two major authorities in the field of competition: The Prime Minister and the Competition Council. The Prime Minister, through the “Department of General Affairs,” is the Administrative Authority in charge of competition policy implementation. However, so far there has been no effective implementation of the law’s provisions. The department in charge of competition has received a number of complaints related to anti-competitive practices, but no concrete actions have been taken already.

3.3 Data and Descriptive Statistics

3.3.1 Data Source

The database used originates from the yearly survey conducted by the Ministry of Industry and Trade. This survey covers all manufacturing firms with at least ten employees or with an annual turnover that exceeds 100,000 MAD (Between US $ 9000 and 12,000 depending on the exchange rate). It collects firm level data on a set of variables such as turnover, output, value added, exports, investment, gross labor cost, and the number of permanent and temporary employees.

Table 3.3 Share of firms with more than ten employees and their contribution in the manufacturing sector in Morocco

Source: Authors calculations from manufacturing survey. Data on production, exports and investments are expressed in billion MAD

For the purpose of this study, only firms that provide 10 or more jobs over their survival period are included in our database. Three main reasons have motivated this choice. First, to ensure comparability of our results with those of the existing empirical literature that uses the same criterion. The second, turnover among very small firms tends to be very high, which would have distorted our results. Third, as shown in the Table 3.3, although firms with less than ten employees represented 33% of firms in 1996, 38% in 2000 and 42% in 2004, their contribution to the manufacturing sector tends to be extremely low. By using firms that provide ten or more jobs over their survival period, our study covers 95 of employment and production of the manufacturing sector and almost the total of its exports.

3.3.2 Descriptive Analysis

For a given year t, if a firm was present in t – 1 but absent in t + 1, it will be classified as an exitor. If a firm was absent in t – 1 but present in t, it will be classified as an entrant. A firm that was absent in t – 1 and t + 1 (i.e., it is only present on t) is both entrant and an exitor. Finally, a firm that belongs to none of the three categories will be classified as a survivor. For comparability across sectors, we define entry and exit rates with respect to the current year’s stock of establishments:

3.3.2.1 Entry and Exit Rates

Firm entry and exit is a part of the market selection process by which resources are reallocated within or across industries. Over the period 1996–2004, entry rates of firms in the manufactured sector in Morocco have varied between a minimum value of 3.1% in 2001 and a maximum value of 6.7% recorded in 1999 and 2000. The average entry rate over the period of time covered amounted roughly to 5% (Fig. 3.1).

Regarding the exit rate, its lowest value was 3.5% in 1999 and its highest value was 7.9% in 2000. Measured on the basis of standard deviation, entry rate has been relatively less volatile compared to exit rates with values of 1.2 and 1.5 respectively.

Over the last 5 years covered by the study, the net entry rate has been systematically negative, which probably reflects the impact of trade liberalization that places firms in a more competitive environment and forces the vulnerable ones to exit from the market.

3.3.2.2 Process of Firms’ Exit

On the basis of the distribution of exiting firms by the number of their survival years, it appears that less than 25% of firms exit from the market before their fifth anniversary. In addition, more than half of the exiting firms do survive for more than 9 years, but fail to remain in the market for longer periods. This finding seems curious compared to the literature according which the survival of entrants is low, with a large number of entrants failing within the first year (e.g., Churchill 1955; Baldwin 1995). Market in Morocco seems to have tolerated inefficient firms for a longer period of time compared to other countries (Fig. 3.2).

Fig. 3.1 Entry, exit and net entry rates in the manufacturing sector. Source: Authors’ computation on the basis of annual manufacturing survey data

Fig. 3.2 Cumulative exitors (in %) by number of survival years. Source: Authors’ computation on the basis of annual manufacturing survey data

3.3.2.3 Entry and Exit Effects on Firms’ Size

There is a relationship between firm turnover and size. Entrants tend to have a size which is lower than the average firm size found in the industry (e.g., Boeri and Cramer 1992). On the other hand, firms exiting the industry tend to have a smaller size than the average size of the industry.

For the case of Morocco, data on the manufacturing firms reveal that the average size of surviving firms is higher than the average size of firms that decide to enter the market. The difference between these two averages is statistically significant for every single year over the period 1996–2004. This finding seems to be consistent with the existing empirical evidence reported in the literature. Our computations show that while the average size of surviving firms tends to increase over the years, the average size of entrants has recorded a certain decline, particularly in 2003 and 2004.

The average size of exitors appears to be highly volatile as it ranges between 49.5 in 2004 to 83.2 in 2000. On the other hand, the difference between the average size of surviving and existing firms is only significant in four cases out of eight.

Regardless of the type of the firm (entrant, surviving or exiting), the median size is significantly lower than the average size. This result indicates that the distribution of firms is skewed to the left, in other words, small firms tend to dominate the distribution. The median among surviving firms has been very robust over the period 1996–2004 and amounted on average to 34 employees per firm. The median size among entrants has been more volatile as it ranged from 18 employees per firm recorded in 1997 to 36.5 in 1998.

Fig. 3.3 Mean entrant’s size over years of survival. Source: Authors’ computation on the basis of annual manufacturing survey data

Overall, firms start business with a relatively small size and then grow progressively over the years. Figure 3.3 shows the itinerary of three cohorts of firms in terms of their average size. The figure reveals different growth patterns.

Firms that entered the market in 1996 had on average, a larger size at their birth. However, they struggled to grow over the three first years of their lives. Their average size increased significantly during their fourth and fifth year, and roughly stabilizes over the rest of the covered years. On the other hand, firms that started their business in 1997 were smaller at their birth. But their average size increased faster and become larger by their seventh birthday when compared to firms that entered in 1996.

Figure 3.4 reports the average size of three cohorts of firms over the years preceding their death. In all cases, the average size tends to decline some years before the complete exit of the firm from the market. In other words, firms facing difficulties attempt first to adjust through workers’ layoff. Achy (2006) has shown that 48% of jobs destroyed in the manufacturing sector in Morocco over the period 1990–2002 were due to surviving firms. When firm’s situation worsens, the decision to exit is then taken.

Entry and exit rates presented earlier are computed as averages for the whole manufacturing sector. These averages dissimulate substantial variation among industries. Figure 3.5 shows that in 2004, for instance, entry rates varied between 0 and 13% with an average of 4.6% for the whole manufacturing sector. Exit rates varied within the same range, and their average amounted to 7.9%. The same finding applies to every single year within the period covered by the study. Therefore, the behavior of entry and exit rate tends to be sector specific, which is in line with the previous empirical literature that reports a strong variation of entry and exit rates across industries (Siegfried and Evans 1994; Geroski 1995; Fotopoulos and Spence 1998; Carree and Thurik 1999; Hölzl and Sögner 2004).

Fig. 3.4 Mean firm’s size over the last years of survival. Source: Authors’ computation on the basis of annual manufacturing survey data

Fig. 3.5 Entry, exit and net entry rates by industry in 2004. Source: Authors’ computation on the basis of annual manufacturing survey data

Achy and Sekkat (2008) has shown that irrespective of the indicators under consideration, textiles, wearing apparel (except footwear) and food products (Coded 321, 322 and 311 respectively in Fig. 3.5) emerge as the most important industries in the manufacturing sector in Morocco. Together, they represent more than 50% of the total manufacturing employment and exports and around 30% of its value added. Hence, instead of looking at each one of the 25 manufacturing industries, the rest of this section focuses on the behavior of entry and exit rates within these key industries over the period 1996–2004.

Fig. 3.6 Entry, exit and net entry rates in the sub-sector of food product. Source: Authors’ computation on the basis of annual manufacturing survey data

Fig. 3.7 Entry, exit and net entry rates in the sub-sector of textiles. Source: Authors’ computation on the basis of annual manufacturing survey data

Entry rates in the food products varied between 1.9 and 8.55%. These rates tended to be relatively more volatile compared to exit rates, which varied between a minimum value of 3.5 and a maximum value of 7.1%. On the basis of the net entry rate, three clear sub periods can be distinguished. From 1996 to 1998 and from 2001 to 2004 exit rates surpassed entry rates. Conversely, for the period 1999–2000, entry rates were higher than exit rates (Fig. 3.6).

Regarding textiles, the net entry rate over the period 1996–2004 has been negative. At the exception of 1999, the flow of firms exiting has always been stronger than the flow of firms entering the textile industry. The gap between entry and exit has even widened over the last years and reaches 7.6 by the end of 2004. In addition to the East European competition, textile industry also suffered from the expected effect of dismantling of the Multi-Fiber Agreement (MFA) that occurred in 2005 (Fig. 3.7).

For the sub-sector of wearing apparel except footwear, two contrasting sub periods emerge from analyzing the behavior of entry and exit rates. Over the first sub period that lasted 4 years (1996–1999), entry rates have always been superior to exit rates. The net entry rate reached a maximum of 12.6% in 1999 compared to an exit rate of roughly 4%. However, since 2000, the net flow of firms into the sub-sector of wearing apparel except footwear has been negative. The net entry rate recorded an extremely worrying value of –8.5% in 2004 with the closure of more than 100 firms (Fig. 3.8).

Correlations between entry and exit rates among various industries have been computed for each year over the period 1996–2004 and reported in the Table 3.4 below. These correlations do not exhibit any specific pattern and tend to be globally low.

Regarding correlations between entry and exit rates for each of 25 industries covered by our study, their values are presented in Table 3.5. Theoretically, if entry and exit rates at the industry level are mostly driven by industry specific shocks, then the cross-sectional correlation should be negative (Bartelsman et al. 2004). Alternatively, if entry and exit rates at the sectoral level are driven by a process of creative destruction within the industry, it is expected that entry and exit will exhibit both high entry and high exit rates. In this case, the cross-sectional correlation of entry and exit rates should be positive. According to Caves’ (1998), entry and exit rates tend to be positively correlated in industries with steady states of maturity, but varying structural entry barriers. The correlation between the two is negative during the early and late phases of a product’s life cycle. During the expansion phase, industries have both high entry rates and high exit rates.

Fig. 3.8 Entry, exit and net entry rates in the sub-sector of wearing apparel except footwear.
Source: Authors’ computation on the basis of annual manufacturing survey data

Table 3.4 Correlations between entry and exit rates among industries by year

Source: Authors’ computation on the basis of annual manufacturing survey data

In 5 manufacturing industries out of 25, entry and exit rates are positively correlated. This is particularly the case of food products, rubber and plastic products, and other chemical products. These correlations corroborate the predominance of within industries reallocations over sectoral shocks. On the other hand, entry and exit rates are negatively correlated in two industries (including wearing apparel). This finding confirms the fact that these industries are hit by industry specific shocks that lead simultaneously to high exit rates and low entry rates. Finally, in the rest of the industries, entry and exit rates are not significantly correlated.

3.4 The Determinants of Firm Entry and Exit

A series of firm, industry and country specific factors determine the intensity of entry and exit. We investigate the determinants of entry and exit in Morocco using a methodology inspired by Disney et al. (2003a) and Klapper et al. (2004) and distinguishing between the three categories of factors. The econometric model for the entry is:

where i stands for industry, t stands for time, entry rate is defined in (3.1), exit rate is given by (3.2).

Firm’s characteristics include average size of entrant in terms of employment or output.

Industry’s characteristics include profit margin, concentration ratio, growth rate, average productivity, average wage rate and capital intensity. Profit margin determines the attractiveness for new firms to enter into the industry. It is measured using the markups computed by Achy and Sekkat (2008). The concentration is an indicator of the easiness to enter a market. It is easier to enter perfectly competitive industries in which many small firms produce standard products. The C4 or Herfindahl will be used. The change in the growth rate (acceleration) of the industry is a proxy of its life cycle. New firms prefer to enter rapidly growing industries. The other variables aimed at capturing some “natural” barriers to entry. Capital intensity may discourage entry because if the industry uses capital-intensive technology, the cost of the initial investment could be substantial. It can be computed as the ratio of capital to the number of employees. The average labor productivity reflects the performance of existing firms. It can also be associated with investment. In both cases, high labor productivity should discourage entrants either because investment requirements are indivisible and massive or because of the risks of severe post-entry competition. The average wage rate in an industry can be negatively correlated to the entry rate, if it reflects the demand for industry-specific skills. Finally, since entry and exit rates tend to be correlated as shown in Table 3.5 and reported in (Caves 1998), exit rate is included in (3).

Table 3.5 Correlations between entry and exit rates by industry

Source: Authors’ computation on the basis of annual manufacturing survey data

Country’s variables include trade barriers, exchange rate, investment and labor market regulations as well as indicators of governance (e.g., political stability, corruption, democratic accountability, bureaucratic quality). Some of the indicators are published by international institutions (e.g., ICRG, WB, and TI) with a time dimension and can be fruitfully introduced in the regression. They are complemented by information on regulations and policies presented in Sect. 3.2.

Note that the impact of some explanatory variables may depend on the value of others. For instance, the correlation between entry and exit rates is negative during the early and late phases of a product’s life cycle, but it is positive during the expansion phase industries. For this reason, lags and interaction terms of explanatory variables are considered.

The model for exit is similar to entry:

where i, t, industry’s characteristics, institutional environment indicators and entry and exit rates are defined as above. Average size (age) of entrant is the average size (age) of exitors between t and t + 1. The following signs are expected for the coefficients. Negative for profit margin (losses stimulate the decision to exit), capital intensity (sunk cots delay exit), growth rate (firms can survive in rapidly growing industries) and concentration ratio (high concentration reduces competition among firms).

Estimations of (3.3) and (3.4) were conducted using various combinations of the above explanatory variables, of interactions terms and of lags. To save on space, Tables 3.6 and 3.7 concern regressions giving the most robust results; i.e., the less sensitive to the estimation methods, presence of other control variables and missing values. Given the fact that the dependent variable can not take values below zero (there is no negative entry or exit rate), econometric theory implies that OLS estimation gives biased results and should use instead the Tobit method. However, if the sample contains no (or a very limited number of) zeros, OLS is preferred by applied econometricians. In this case, the bias is almost inexistent and OLS gives easily interpretable coefficients. Since in our sample the proportion of zeros is non negligible, we present the results with both estimation methods. Moreover, the analysis in Sect. 3.3 suggests that there was a change in the behavior of the entry and exit rates in the early 2000s. This may be related to an expected change in labor regulations (see Sect. 3.2) or to the adoption of competition law (see Sect. 3.2). Whatever the reason, this suggests that conducting estimation on the whole sample on the one hand and on a sub-sample covering only the 2000s may be instructive.

Tables 3.6 and 3.7 show that the results are slightly better for the exit than for the entry equation. The quality of fit is markedly higher when estimation is conducted over the 2000s only. The results are almost not affected by the estimation method. Focusing on the Tobit results and the 2000s sub-sample, entry is affected significantly and negatively by the Real Effective Exchange Rate (REER). An increase in the REER means an appreciation of the dirham and, hence, a loss of competitiveness of Moroccan firms with respect to foreigners both on the domestic and international markets. This dissuades firms from entering. The coefficient of the lagged entry rate is positive and significant. This implies that if entry is feasible into a given market (less barrier to entry or expanding demand), there will be more and more entry. Interestingly, the coefficient is almost twice higher during the 2000s than over the whole period irrespective of the estimation method. One possible explanation is that the competition law enacted in 2000, although not completely enforced, facilitated entry of new firms.

Table 3.6 Determinants of entry rate (dependent variable = log ((entrants/survivors) × 100+1))