4/28/2004 1 The Rise and Fall of Township and Village Enterprises and their Private Rebirth Lecture 7 4/28/2004 2 China’s Privatization Movement and the Rest of the World ? Maybe 4 times more than the ROW combined … ? Almost unreported … ? Lots of misunderstandings: – Failure – Corruption – Top down … Policy driven 0 0.5 1 1.5 2 Millio n s o f Firm s China ROW ? 4/28/2004 3 Main Points ? Not much new … continuation of long string of changes in China’s rural firms … ? Driven by economic necessity … decentralized … policy played little role ? Current privatization wave: – Example of Innovative Institutional Change … almost all insider privatization … difficult problem to solve … – Determined by Economics – Mostly successful – Little impact on local finances 4/28/2004 4 Outline ? Rise of TVEs (Walder) ? Evolution into contracting (Township Enterprises) ? A twist in contracting in Village Enterprises ? Determinants of privatization ? Insider privatization 4/28/2004 5 The Rise of TVEs ? Local State Owned Firms … ? Millions of them … ? Engine of China’s Economic Growth in the 1980s … 4/28/2004 6 Walder (1995): Most Concise / Believable Explanation ? Combination of factors made TVEs the most optimal form of enterprise, despite their shortcomings – Sudden wealth from HRS + other reforms (demand by consumer) – Reform rural enterprises first (SOEs not allowed to deviate from plan) – High Profits – Fiscal Reform (can keep profits) – Lack of markets ? Inputs ? Commodity Markets – Ideological baggage of Socialism ? Formula for success: – Build a factory; run it yourself (leaders had connections); – Build another; Hire a manager (profits are so high, do not need very good incentives); – Etcetera (build a lot of factories; get rich; get promoted; protected by position) 4/28/2004 7 Puzzles and Misconceptions ? Lots of economists try to explain their success – Fuzzy Property Rights – Corrupted House of Cards – Communism Works ? In fact, 10 years after their appearance, economists were still trying to explain how firms could function so well with poor incentives ? What is problem with this work? 4/28/2004 8 Incentives were not so poor by late 1980s 0 10 20 30 40 50 60 70 80 90 100 1985 1988 1991 1994 Leader-run Profit-share Fixed-lease Manager-run firm … after payment of fixed fee, manager keeps residual profits Traditional, Leader-run firm … incentive problems Profit-sharing … leaders and managers split profits Chen and Rozelle, JDE, 1999 … based on survey of 120 firms in ZJ/JS/JX/Hub 4/28/2004 9 Changing Environment (Naughton, 1995) ? Initially high profits ? Entry of firms ? FDI; new technologies ? SOE manager reforms ? High profits fall ? Search for efficiency 0 2 4 6 8 10 12 14 16 18 20 198 5 198 8 199 1 199 5 Millio n s o f f i rm s ; Percen t Pro f it s Profits Firms 4/28/2004 10 Explanation – Why Shift / Who Shifted First ? Fixed Lease Firms: – Labor intensive – Areas with good markets – Busy leader (lots of firms) – Educated manager (educated leader ran firm himself) 0 10 20 30 40 50 60 70 80 90 100 K-intense. L-intense. Poor mkts Good mkts Fixed Lease Leader Run 4/28/2004 11 Explanation: property rights were pretty good – in fixed lease firms, manager already had good: income rights / good control rights 0 10 20 30 40 50 60 70 80 90 100 Leader Run Profit share Fixed lease Hire/fire Distribute profits Sell Assets Types of Firms 4/28/2004 12 Village Enterprise Puzzle ? Chen and Rozelle, JDE, 1999 ? Fixed Lease Firms: – Labor intensive – Areas with good markets – Busy leader (lots of firms) – Educated manager ? Mohapatra, Goodhue and Rozelle, 2003 ? Fixed Lease Firms: – Nothing is significant – Performance is not enhanced by fixed lease contract – Why? Towns Villages 4/28/2004 13 Results –descriptive evidence (tables) Table 1. Profit sharing contracts All Regions Contract type 1988 1995 Fixed wage 26 16 Share 24 25 Fixed rent 50 59 Puzzle: Why did village enterprises behave so differently? -- no change in contract -- no systematic determinants of firm types 4/28/2004 14 Results –descriptive evidence(contd) Table 2. Investment Bonding contracts All firms Fixed rent firms Non-fixed rent firms % of firms with Investment bonding 35 74 26 Investment amounts-all sources (as a percentage of total investment) Managers 24 35.30 8.66 Village funds 26 20.12 34.75 Bank loans 22 14.81 31.07 Individual loans 12 11.01 13.73 Firm’s own funds 12 13.10 10.90 Village lenders 4 5.59 0.81 4/28/2004 15 Performance of Village-run Firms, 1988 to 1995 ? No impact of fixed leasing (unlike TEs) ? Performance of all firms the same ? Except when managers have current incentives + long run incentives 0 1 2 3 4 5 6 7 P erc ent per ye ar gr ow t h , Pr o f i t s Leader-run Fixed Lease Bonded Lease + Bonding 4/28/2004 16 Summary – Rural Industry on the Eve of the Privatization Wave ? Most firms leased to their managers – Income rights – Most of control rights ? Increasing use of investment bonding – Make manager “invest” in part of firm ? Mid-1990s: very competitive; low profit rates; need for new investments in newer technologies 4/28/2004 17 Li and Rozelle (2004) paper … we began our study just as the privatization (gaizhi) movement was starting ? What are the trends of the movement? ? What is the determinant of why firms were privatized? ? How did they do it? [The process] ? What was the impact on performance? 4/28/2004 18 Background of our Study ? International privatization movement: – US / Western Europe / Australia and New Zealand – But especially: Central and Eastern Europe / CIS ? The experiences colored our view of things … provided a series of maintained hypotheses … some of which were hard to “un-maintain” 4/28/2004 19 Insider Privatization ? Insider Privatization: the sale of a government- owned firm from the “government official” (seller) to its “manager or employees” (buyer) ? Practice – very common in: – Russia (Boycko, Shleifer, and Vishny, 1995; Blanchard and Aghion, 1996; Carlin and Aghion, 1996; Earle and Estrin, 1996) – CEECs (Carlin and Aghion, 1996; Frydman et al., 1999) – Asian countries: Mongolia (Anderson et al., 1999) 4/28/2004 20 Record—Rest of World ? No substantial improvement in performance in most privatization movements (Earle, 1998; Anderson et al. 1999) ? Privatized firms under-perform compared with other private firms (Frydman et al., 1999; Barberis, Shleifer et al. 1996) ? Insider privatization: least successful ? Two major reasons (Frydman and Rapaczynski, 1994; Blanchard and Aghion, 1996; Barberis et al., 1996; Black et al., 2000) – Lack of capital – Lack of human capital 4/28/2004 21 One of biggest constraints to insider privatization: Valuation ? Asymmetric information between official (seller) and manager (buyer) (Frydman and Rapaczynski, 1994; Putterman, 1997) ? Managers possess insider knowledge and have incentives to understate the value of the firm ? There are no good CPAs that can provide objective valuations ? The official’s problem: has to accept low price or not to privatize at all 4/28/2004 22 Data ? Primary data collection: focused on TEs ? Stratified random sampling by income – 59 townships in two provinces: Jiangsu and Zhejiang ? Township level survey – Township Census: all TEs existed in 1994 (670 firms) – Township government official survey ? Firm-level survey (3 firms per township, 168 firms in total, 94-97) … reinterviewed in 1999/2000 – Include 33 private firms – Manager interview on privatization, governance, finance – Firm accounting information: cash flow and balance ? Bank Survey: bank behavior and balance sheet data 4/28/2004 23 The Case of Rural China ? 2 million government-owned firms privatized spontaneously in the past decade ? Most firms are sold to insiders, especially former managers (Kung, 1999; Li and Rozelle, 2003) ? Many potential reasons about why privatize ? Information is also asymmetric: valuation problem (Putterman, 1997; Kung, 1999; Lin and Kung 2000) ? Not all privatized firms have succeeded – Lack of capital and human capital cannot explain all 4/28/2004 24 Privatization Trends ? Not all the same way – three types of privatization – Share-shifting (SS) any shift of shares – Controlling interest Share-shifting (CI-SS) > 50% – Complete Privatization (CP) 100% private shares ? There were also fully private firms emerging during this time … we will analyze these firms later 4/28/2004 25 Rates of Privatization of TE Firms, 1993 to 1999 0 20 40 60 80 100 SS CI-SS Full TE Privatized 4/28/2004 26 Differences among Townships on Rate of Privatization, 1993-1999 0 5 10 15 20 25 30 Percen t o f Firm s Priv a t ized 0-20 21-40 41-60 61-80 81-100 4/28/2004 27 Privatization, 2004 ? Accelerated over mid- to late 1990s! ? Lots (most?) townships have privatized 100% ? SOE privatization is only thing of interest now (TVEs are all done!) … ? Local state still involved in many firms (formal / informal … explicit / implicit) … more complicated relationship … 4/28/2004 28 The Case of Rural China ? 2 million government-owned firms privatized spontaneously in the past decade ? Most firms are sold to insiders, especially former managers (Kung, 1999; Li and Rozelle, 2003) ? Many potential reasons about why privatize ? Information is also asymmetric: valuation problem (Putterman, 1997; Kung, 1999; Lin and Kung 2000) ? Not all privatized firms have succeeded – Lack of capital and human capital cannot explain all 4/28/2004 29 Inside Job ? Little choice: “of course, insider” ? Almost 100% former managers ? On average, manager managed the firm 5 years and worked in the firm for 12 years ? Even if not “insider” there was close connection Cadre Party Member Soldier / Formal Position Friend/ Relative of Leader Other 4/28/2004 30 The Case of Rural China ? 2 million government-owned firms privatized spontaneously in the past decade ? Most firms are sold to insiders, especially former managers (Kung, 1999; Li and Rozelle, 2003) ? Many potential reasons about why privatize (DETERMINANTS) ? Information is also asymmetric: valuation problem (Putterman, 1997; Kung, 1999; Lin and Kung 2000) ? Not all privatized firms have succeeded – Lack of capital and human capital cannot explain all 4/28/2004 31 Hypothesis 1: Policy ? Privatization is purely policy driven ? Policy measures – County policy indicator (1 if there is a county level privatization policy) – Policy intensity: the deadline (number of months) to privatize all firms ? If privatization is purely policy driven, then other variables should have no effect 4/28/2004 32 Hypothesis 2: Human Capital ? Comparative advantage hypothesis: privatization may occur when a manager has a comparative advantage in running the firm relative to an official ? Official and manager co-run TVEs in 1980s – Officials have connections – Managers may not have more management experience than officials ? In the 1990s – Connections became less important with the development of markets – Managers have accumulated enough human capital – Chen and Rozelle (1999) find that income and control rights have shifted to managers even in the mid-1980s ? Measures of human capital: education and age of official and manager 4/28/2004 33 Hypothesis 3: Competition ? Privatization is more likely when the market becomes competitive ? Ownership structures of TVEs is a result of – Government’s monopolistic power (Chang and Wang, 1994) – Government regulations in market transactions (Li, 1996) ? When the monopolistic power is gone and there are fewer regulations, government ownership becomes less preferable due to its large agency cost (this means local leaders just don’t care as much about profits for their firms / are too busy to care) ? Measure of market competitiveness: an barriers- to-entry index developed by Yang (1998) 4/28/2004 34 Hypothesis 4: Hardening Budget ? Privatization is more likely when the government faces a hard budget constraint from the bank ? TVE expansion in the 1980s and early 1990s was accompanied by credit expansion in rural China – Soft credits came from ABC and RCCs, which were guaranteed by township governments – Local governments got further financing even if projects were unsuccessful from the beginning (Oi, 1999) ? Bank reform in 1994 hardened the budget and led to privatization – Lending authorities were centralized to upper-level banks (Park and Shen, 2003) – Township officials do not have advantage in getting loans – They care more about firm profitability ? Measurement of budget hardness: an indicator – 1 if officials cannot persuade banks to give extensions of overdue loans to TEs before liquidating them 4/28/2004 35 Other Determinants ? Firm quality – Profitability – Exporting firms ?Size – Smaller firms are more likely to be privatized 4/28/2004 36 Summary of Hypotheses ? Privatization is policy driven ? Privatization is an institutional evolution; and is more likely when – Managers have comparative advantage – Budgets are hard – Markets are more competitive – Firms are smaller – [profitability? could go either way …] 4/28/2004 37 Findings—Determinants Analysis ? Privatization is more likely when the managers are more educated or older ? Privatization is more likely in more competitive industries (controlling firm size) ? Privatization is more likely when governments face hard budget constraint ? Pressure from upper-level governments cannot explain privatization of the sample ? Smaller firms are more likely to be privatized ? [profitability? Some studies more profitable, more likely to privatize / others privatize the “losers”] 4/28/2004 38 Endogeneity: Budget Hardness ? The hardness of the budget could be endogenous – Reverse causality: governments may privatize to harden the budget – Omitted variable bias: both privatization and hard budget may be determined by other unobserved township characteristics ? Two methods to deal with endogeneity – IV method (the hardness of budget in neighboring townships and county indicators as IVs) – Excluding firms privatized in 1994 ? Results are similar to previous findings 4/28/2004 39 Conclusions ? Privatization is prevalent in rural China regardless of the definition ? Privatization – Not central policy directed or outright corruption – A response to changing environments ? The rise of TVEs was a response to the imperfect conditions of the market and state ? The privatization of TVEs is a response to the improving conditions / competitive pressures 4/28/2004 40 Summary ? Massive privatization ? Sell to insiders ? Mainly spontaneous ? NOW – question for China: – How privatize? – Did they succeed? When insider privatization in other countries failed 4/28/2004 41 Research Questions ? How can local officials overcome information asymmetries and get a good deal from privatization? And do they? ? Is there any contractual mechanism that facilitates insider privatization and helps increase the performance of privatized firms? ? Do privatized firms in rural China actually perform better after privatization (e.g., how do they do compared to private firms)? Is performance uniform across insider privatized firms? 4/28/2004 42 Big Problem: Managers know firm prospects / Leaders do not [Asymmetric Information] ? Each township owned 12 firms in 1994 ? Each firm sold to 4 different destinations ? officials have other duties ? The manager managed the firm 5 years and worked in the firm for 12 years Question: how can (insider) privatization move forward? 4/28/2004 43 The Screening Theory (1) ? Follow Laffont and Tirole (1986) ? Official offers a menu of contracts – Low buyout price, government shares future profits (leaving a “tail” with the government) – High buyout price, government does not share future profits ? Manager chooses one contract, pays V and becomes the new owner ? Manager put effort e to manage the firm ? Profit is realized and divided between the two parties according to the contract chosen by manager earlier 4/28/2004 44 A Screening Theory (2) ? Good managers (or managers of high quality firms) choose high buyout price contract, while poor managers (or managers of low quality firms) choose low buyout price contract ? This occurs because – Good managers know that they will earn profits in the future and will choose the contract that let them keep most or all future profits -- hence they are willing to pay for this – Poor managers, on the other hand, do not expect to make large future profits, and so choose the contract in which they do not have to make a large up front buyout price 4/28/2004 45 A Screening Theory (3) ? Consequences – The size of the buyout price will be positively related to the share of future profits kept by the manger – Managers that pay low buyout prices, only receive part of future profits, and will under- perform – Managers that pay high buyout prices, face good incentives, and will perform well in the future 4/28/2004 46 Privatization Process ? Evaluation and the buyout price base value = assets - debts buyout price = base value + premium (discount) ? The normalized buyout price Premium (discount) = buyout price – book value of equity Premium (discount) rate = premium/assets ? The tail – Postprivatization sharing rule – Screening contract indicator 4/28/2004 47 Example of Establishing Buyout Price / Premium or Discount -20 0 20 40 60 80 100 Assets Debt Book Val. By CPA + Gaizhi committee Book value = Assets - Debt Premium Discount Premium / discount established thru negotiations: leader & manager Buyout Price (2 examples) 1 2 4/28/2004 48 (Buyout Price to Base Value Ratio) 0 5 10 15 20 25 N u mb e r of s a mp l e fi r m s 0-20 21-50 51-75 76-100 >100 NEGATIVE Premium / Discount Level Pay a discount Pay a premium Base Value = Book Value Assets minus Book Value Debts (+ BP) / (- BV) 4/28/2004 49 The Premium/Discount and the Tail 0 10 20 30 40 50 60 70 80 90 100 Firms in which Manager Paid a Premium Without Tail With Tail Percent Good incentives: ALL of profits to manager Poorer incentives: on average, 40% go to township gov’t 4/28/2004 50 Hypotheses ? H1: the tail (share of gov’t) is negatively correlated with the buyout price ? H2: the postprivatization performance of a firm increases with the buyout price ? H3: the postprivatization performance of a firm decreases with the tail (the profit share of the government) ? H4: the buyout price decreases with the degree of information asymmetry (more the entrepreneur knows relative to gov’t, lower the buyout price) 4/28/2004 51 Measuring Performance ? 3 performance (effort) measures – Manager’s work time: number of hours per week – Accounts receivable management =1-accounts receivable/asset value – Value added per worker (in log) 4/28/2004 52 Summary of Empirical Findings Results support all four hypotheses ? The tail is negatively correlated with the buyout price ? Performance increases with the buyout price – Both OLS and fixed-effect estimations ? Performance and the tail – Performance decreases with the size of the tail – Performance is lower when governments use the screening contract ? The buyout price decreases with the degree of information asymmetries 4/28/2004 53 Conclusions ? Insider privatization prevalent in rural China ? Privatization – Not central policy directed – A response to changing environments – Institutional flexibility allows officials and managers to solve a difficult economic problem, and have successful insider privatization ? China’s rural industry – Performance will improve over time – Private firms will be important