Policy, Research, and External Affairs WORKING PAPERS Public Economics Country Economics Department The World Bank March 1991 WPS 615 Socialist Economic Growth and Political Investment Cycles Heng-fu Zou Investment rates in China have often been highest under leftist (hardline) political regimes, not rightist (softline) political re- gimens. I he iP s . R, -. *~. . 1: k :~ . x :crp:retat. )r. . and cornlhr onq are the ,:AI . : * .. :: . ti' .mi, .:.:.rd i% i,r ii nic.rnN r (siltr:so Policy, Research, and External Affairs Public Economics WPS 615 This paper - a product of the Public Economics Division, Country Economics Department - is part of a largereffort in PRE tostudy theentrallyplanned economics in transition. Copies are available irre from the World Bank, 1818 H Street NW, Washington, DC 20433. Please contact Ann Bhalla, room N 10-059, extension 37699 (24 pages). Socialist economic growth in China and Eastem * Why investment hunger is an inevitable Europe has long been characterized by invest- consequence of social planners' rational chioices. ment hunger, drives toward expansion, and cyclical fluctuation of investment rates. * When a drive toward expansion can cause a permanent shortage of consumption goods. For decades, relatively high groA rates - often accompanied by a shortage of consumption Through numerical examples and empirical goods - have typically been achieved at the tests, Zou provides a framework w:thin which to consumers' expense. analyze political investment cycles in a socialist economy. In China, Zou linds that high invest- Treating social planners as self-interested ment rates have often been linked to leftist (or bureaucrats, Zou ofi'rs a positive model to help hardline) political regimes and low or moderate understand the norms of socialist economic investment rates with "rightist" (or so[tline) growth. This model demonstrates: political regimes. H flow rapid capital accumulation tends to serve the social planners' own interests. The lPRE Working Paper Series disseminates the findings of work under way in the lmnks Policy, Research. and Extemal AffairsComplex. An objective ofthe scrics is to get tlce findings out ruickly, even if preseniations are less thaln fully polisled. The findings, interpretations, andl conclusions in thesc (do not nTOCCssdril rcprsent oflficial firink policN. ProduLed h! the PRE D)sscnnmaion aun eCntcr Socialist Economic Growth and ?o)'tical Investment Cycles By Heng-fu Zou Table of Contents I. Introduction 1 II. The Model and Its Justification 4 III. The Dynamics of the Model and the Properties of the Equilibrium 7 IV. A Numerical Example and an Iliustration of Political Investment Cycle 11 V. Historical Evidence 16 VI. Summary 22 References 23 Sociallst Economic Growth and Political Investment Cycles I. introduction In tradltlonal optimal growth models for a centrally planned economy, e. g. Cass (1965) and Koopmans (1965), social planners maximize an lntertemporal soclal welfare function defined on per capita consumption, subject to the dynamic constraint of capital accumulation. The results from these models have become the folklore of modern economics: there exists a unique optimum path converging asymptotically to the unique equilibrium; the optimum capital stock in long-run Is determined by the famous modified golden rule, I.e., marginal productivity of capital is equal to the natural growth rate of population plus the time discount rate of social planners. In these models, social planners all act In the Interest of the society. They do not have any objective function other than the welfare of the people, and their personal Images are only reflected In the time discount rate. Cass (1965) provides a typical picture of the central planners: "The central planning authority's concept of social welfare is related to the ability of the economy to provide consumption goods over time. In particular, welfare at any point of time Is measured by a I am grateful to Bela Balassa, Richard Caves, Janos Kornai, Andrew Newman, Dwight Perkins, Yingyi Qlan, Jeffrey Sachs and Xinsheng Zeng for very helpful discussion and comments. The first version of this paper was presented In Janos Kornai's workshop at Harvard, I thank the seminar participants for their suggestions. 1 utility Index of current consumption per capita .... The central planning authority recognizes that consumption tomorrow is not the same thing as :onsumption today. For thls reason, It takes the politically pragmatic view that its planning obligation is stronger to present and near future generations than to far removed future generations. This view is implemented .n practice by discounting future welfare at a positive rate". This approach to socialist growth suffers .rom serious limitations when compared to socialist reality. First of all, traditional optimal growth models are based on an insufficient understanding or InH ed, a misunderstanding of the nature of the social planners. This point has been emphasized by Janos Kornal In his various studles ( Kornal, 1982, 1986, 1988). With both political power and economic resources in their control, social planners are not constrained or directed to choose the optimum feasible growth path with respect to the only criterion, which Is to maximize social welfare. "Such an unworldly bureaucracy never existed in the past and will never exist In the future. Political bureaucracies have Inner conflicts reflecting the divisions of society and the diverse pressures of varlous social groups. They pursue their own Individual and group Interests, Including the Interests of the particular specialized agency to which they belong. Power creates an irresistable temptation to make use of it. A bureaucrat must be interventionist because that Is his role in the society; It Is dictated by his situation" (Kornai 1986, pp. 1726-27). In practice, social planners are often investment growth rate maximizers (Grosfeld, 1987), and their personal interests are more connected to the persistent expansion of their organizations than to the Increase In people's consumption. In their investment trategies, "the highest priority is placed on Industry, and within industry on heavy industry, and within heavy Industry on the part related to 2 the military. ... Anztong the neglected, non-priority sectors, one typically finds agriculture, and even more so all the branches of the tertiary or service sector, such as transport and telecommunication, housing, other communal services, domestic trade, and health. This diversion of resources from consumption to Investment takes place not provisionally for two or three years, but for decades, for twenty, thirty, or forty years". (Kornal, 1988, p.244). In this paper, we intend to offer a simple alternative model to capture certain essential aspects of socialist economic growth. The most imnortant ieature of the model Is In defining the social planners' objective function in both per capita consumption and per capita capital stock. The model Is Justified and set up ln Section II. In Its abstract form, thls modelling was presented by Mordecal Kurz In 1968. That paper has long been neglected in the economics profession partly because, we guess, Kurz has not offered any Justification for the so-called wealth effects model. In this paper, we are able to find a realistic setting for the Kurz model In socialist economic growth. In Section III, we demonstrate that this simple model provides good framework for the understanding of "Investment hunger" and "expansion-drive" studied by Kornal (1980). Section IV derives a theory of polltical Investment cycle frcm our basic model. It Is shown that the investment rates (or accumulation rates In the terminology of socialist economics) are related to different political regimes In socialist countries. In Section V, we will look at the empirical data on lnvestment rates from 1952 to 1985 In China. The variations on Investment rates throughout those years can be substantially explained by the change of political power at the top level of government, evidence supporting 3 the theory of the political investment cyc:'e. II. The Model and Its Justification We define the Instantaneous utility function of social planners at a given time t as the summation of two parts: u(c t) + iv(k t). ct Is consumption per capita and kt Is capital stock per capita at time t. Social planners derive positive utility from both consumption enjoyed by che people and capital stock owned by the state, so the first order derivatives of functions u(.) and v(.) are positive. The Greek letter x Is a positive constant that measures the importance of capital accumulation from the point of view of the social planners. In later sections, we will allow x to take different values, and its effects on capital accumulation, the Investment rate and const..Iption will be studied. Furthermore, for technical reason, we assume that the second order derivatives of u(.) and v(.) are negative, and that: lim u'Cc ) =X ct -->0 which guarantee the sufficiency of the first order condltions for optimization, and exclude the corner solution of zero consumption. In modelling the social planners' preference, we maintain that social planners do care about people's consumption, and the improvement in the living standard of the people seems to Justify their manipulation of political power and economic resources In a sociallst economy. But it Is more important to note that social planners' own Interest lies more directly In the expansion of the firm and public organization of which they are In charge. Social planners are not Just a group of persons In the central planning bureaU, they consist of all persons involved in formulating the plan, from the managers at the bottom 4 to the minlsters at the top. According to Janos Kornal (1981, 1386, 1988). the first and the most important otivation for accelerated capital accumulation Is the Identification of social planners with their own Jobs. An expansion of the firm or organization under tl'Ir direct control is always a source of satisfaction. The second motivation is prestige. "A la -er organization brings more prestige, and also more power" (Kornal, 1981). "Tha simple urge to exert power over people, and to exercise some discretion over the allocatlon of physical resources can also make managers strive for higher Investment levels for their firm" (Kornal, 1988, p.264). So "it Is lmportant to note that Investment hunger and expansion drive characterize not only the behavior of the top manager and his subordlnates In a particular firm, but also the attitude of economic agents at all levels of the bureaucratic hlRrarchy in a socialist system.... the general ideology of the system favors expansion, and no claimant's application for funds is ever regarded as unreasonable or unethical by anyone in the hierarchy. On the contrary, everyone considers such a request as the natural and normal behavior within the system." (Kornal, 1988, pp.264-265, Italic added.) This assessment of socialist planners Is essentially th. ;-ne as the one used In the analysis of bureaucrats in western democracies. For example, Orzechowski (1977) defines the bureaucrat's utility function directly on the output produced by his bureau and the capltal stock or labor In his control. And the striving for more budget revenue In western public sectors resembles the Investment hunger and expansion drive In rocialist economies. With these discussions, we might call w which appeared in the social planners' utility function as the measure of the degree of expansion drive. A large value of x means that the social planners are highly expansion oriented; and a S zero value of n brlngs us back to Ramsay-Cass-Koopman's mathematical utopia of socialism. (Phelps (1961) presents the golden rule of capital accumulation In "a fable for growthmen". In reality, where we can f'nd the King of the Kingdom f Solovia?) To proceed with our model, we assume that the social planners maximlze the following lntertemporal *tility with discounting (for notation convenlence, we omit the time subscript t of all variables from not on): X[u(c) + nv(k)le dt p > . (1) 0 where p Is the social planners' subjective rate of discount. There Is a standard neoclassical production function f(k) In the economy with f'(k) > 0, and f"(k) < 0. Capital Is subject to a depreciation rate 6. The population growth rate Is exogenously given as n. So capital accumulation in per capita term follows the dynamic equation: k = f(k) - c - (n + 8)k (2) Social planners maximize (1) subject to the dynamic constraint (2). The current value Hamiltonian H is defined by: H = u(c) + iv(k) + AMf(k) - c - (n + M)kI (3) The optimal paths for consumption and investment are: 1 c = -- rv'(k) + u'(c)(f'(k) - n - 6 - p)] (4) - u'(c) k - f(k) - c - (n + 5)k (5) 6 lim eptAk -O (6) t-->m We are going to make a detailed analysis of above dynamics In the next sf:ct Ion. III. The Dynamics of the Model and the Properties u the Equilibrium As noted by Kurz (1968), the dynamic systems (4) and (5) may easily result In multiple equilibria, and some equilibrium points are saddle point stable, while some are totally unstable. To see this, denote the equilibrium values of consumption and capital as c and k , and linearize the systems around these values: ..v"(k ) + u'(c )f"(k ) c (n+8+p) - f'(k) c - c (7) k -1 f'(k) - n - a k - k Denote the 2x2 matrix as M. The trace of the matrix: Trace of M = p > 0 . (8) As the t-ace Is the sum of the two char-cteristic roots of the systems, at least one of the roots is positive. Therefore we cannot have a stable equilibrium point. Next the determinant of the matrix Is: irv"(k )+u'(c )f"(k ) A - (rA+8+p-f'(k )l[f'(k )-n-1] - (9) u"(c 7 The seoond term on the right hand side of (9) Is negative; the first term in positive or negative depending on wihether the capital stock is smaller or larger than the golden rule capital as pointed out by Kurz (1968). If the steady state cap!tal stock is equal to or larger than the golden rule capital, f'(k) is equal to or less than n + 6; the flrst term on the right hand side cf (9) is also negative because [n+6+p-4"(k I is positive as shown below in proposition one. In this case, A Is negative. For A is the product of two characteristic roots, negative A Implies that one root Is pos'tive and one negative. If A Is positive, then both roots will bw positive as the existence of two negative roots contradicts (8). For this section, we will focus ..n the case that A is negative, that is to say, there exists a unique optimal path in the neighborhood of the equilibrium. Furthermure, we assume that there exists only one equilibrium for the systems. A numerical example is presented in the next section before we go on to discuss the political Investment cycle. Of course, If the time discount rate Is very small, the first term on the right hand side of (9) Is negative; so Is A. The properties of the unique saddle point equilibrium follow In order: Property One: The equilibrium capital stock Is larger than the modified golden rule one. To show this, note that, in a steady state, we have: * {nv'(k ) + u'(c )[f'(k ) .,-8-p]D = 0 (10) -u' (c f(k ) - c -(n + 8)k = 0 (11) 8 Earn (10):. iTv' (k ) f'(k j = n + + p - . < n + 6 + p x f(kmg) (12) u' (c ) where k m denotes the modified golden rule amount of capital. From (12), it Is clear that k > kmg as f"(.) Is negative. The explanation Is simple. Since :ocial planners benefit dlrtctly from the expansion of the eco Jmic organizations and since the welfare of consumers over the lnfinlte horizon is not the only criterion for planning, the short-riin consumption will be partly sacrified for the expanslon drive. It is quite possible that, as shown In next numerical example, consumotion Is permanently sacrificed In this kind of models: equilibrium consumption Is lower than the golden rule one and capital is over-accumulated. Property Two: The higher the value of n, the higher the steady state capital. Totally differentiating equations (10) and (l1J, we have: s v'(k) (13) It is simple to show that: dk 1 v'(k) = _ . (14) dir a U (c which Is positive as the economy is on the unique optimal convergent path. As for the steady state consumption, the sign Is ambiguous depending on whether the equilibrium capital Is higher or lower than the golden rule capital. The effects of n on Investment and consumption on the unique optimal path can 9 ~lsu be analyzed. From (7), the solutions of the linearized systems for the :.et;ivlor of the capital stock and consumption are: kt = k (k k )eet (15) 0 kt = -O(k - kt ) (16) * *0 Ct = c + (f(k ) - n - - EON)(kt - k (17) where 9 Is the negative root of the dynamic system: 1 2 0 = - p - VI p _4A (18) 2 From (16) and (17), It is clear that, through Its positive effect on steady state capital, k , the high value of n leads to high investment and low consumption on the optimal path for all kt less than k . But we should note that t may also affect e and c . If the increase In i tends to lower e, lr. other words, e becomes more negative, then the Investment will be unambiguously high as a :esult of t being high. Property Three: The higher the value of n, the higher the steady state investment rate (cs- saving rate). In the steady state, Investment Is Just (n + 8)k . Let the investment rate (or saving rate) be s, then: (n+8)k s = . (19) f(k ) ds (n+6) * * dk -- [ tf(k )-f'(k )k- (20) dx (f(k)2 dx 10 h is positive since dkk/dn is positive and If(k )-f'(k )k I ive f-or any concave funct ion. ' of pr(>tert es *,t.i Above reve:al how social planners' pref t. i t e growth pattern lin sE30lallst economles. In the Cass model, ", ..w .. the form of social welfare functions does not enter Into the !etermination of equilibrium capital stock. Even if we interpret the social welfare function as the social planners' own preference, the equilib[ lum capital and consumption are still Independent of the social planners' preference as long as their preference is defined only on consumption. Recali from Cass (1965), that In equilibrium: f'(k g) = n + a + p (21) f(k g) - c - (n + S)kmg = O (22) so the social welfare function itself plays no role In the determination of k mg in Cass model. (Please compare (21) and (22) with equilibrium conditions (10) and (11).) The invention by Kurz (1968) provides us a rich picture for she link between preference and economic growth. Of course, as a positive ipproach, the Kurz model with proper Justification is much more realistic thuin .he Cass model when applied to a socialist economy. IV. A Numerical Example and an Illustration of Political Investment Cycle .'ven though the modified Kurz model gives us Interesting results, the e>istence of multiple equilibria brings about complicated dynamics even with simple preferences and technology. Here we show that, If preference is the .-c,ular lcgarithm functions cf consumption and capital and if technology is 11 standard Cobb-Douglas, there r its a unique equilibrium and a unique optimal path. Now the social planners maximize: -Pt I [logc + nlogk]e dt (23) 0 s.t. k k - c - nk (24) where 0 < a < 1, and we have set 8 equal to zero for simplicity. The corresponding optimal conditions are: c c u- rc - ak - (n+p)k] (25) Ic k = k nk -c (26) Set the time derivatives of c and k equal to zero in (25) and (26), the unique optimal equilibrium point Is: 1 kc = n + a - (27) nn+n+p * c =k -nk (28) The determinant of the corresponding matrix M Is: C *O-1 2 Oa-1 a S Wak a - n] + [a k - (n.p) (29) k Upon substitution: (n+p-na)(wn+n+p)[n(a-i)-w(1-a)I S .-2 < 0 (30) 12 15 [: < a < I. So there is one negative characteristic root and one positive )t the equilibrium is saddle point stable. It is straightforward to check that dk . p+(l-an = kil-)k 2 > 0 (31) du (nn+nnp) ds (c)dk - = n(l-a)k ( > 0 (32) di dx Next we are going to see under which circumstances a high degree of expanslon drive leads to dynamic inefficiency. In Phelps (1961), the golden rule capital stock at which consumption is maximized is given as (for Cobb-Douglass technology): kg = - -a (33) n For k Is larger than kg, it is required that: it + aX a (34) nn + n +p n which is the same as require that: ap > ~~~~~~~~(35) 1 - an For a = 0.25, p = 0.05, and n = 0.01, i should be larger than 0.0125. which is not a strict requirement. So, In this case, the people's consumption is not only sacrificed on the dynamic path converging to the steady state, but Is also sacrlficed in the steady state. 13 :' .use ther-e are two groups of social planners in the economy. Fvlh.,wg n, .nt ion, we may ..:aII cne group "softliners" or the "right", and Aho ' : ir',ers' or the ''left" Threy ilternat ively control the pr esn, . . ,an I Ils kriown thtalt in siocIalIst countries sUCIh as H'ng.iry, * ? ts wa,r,d;s (rc<.;umpt Ion rather than investmenrt alw:.ys come ar , * .'t ,f" "softl.ine' rule; the "hardliners" or the "left' are always *'xpcinsion oriented. (see Kornal, 1988, pp. 283-284). In our model, if we J r,.t e IT as the expanrsio n desIre of the "left" and n as the expains;:A-.:; I r ^f the "right" and let I1 > ir > ap/(1-an), then tnie "left" maximizes: f[logc + nl1logkI]eP dt (36) 0 s.t. k= k- c -nk (24) the "right" maxlmizes: Jr[logc r+ Tr rlogk rle Pdt (37) 0 s.t. k = ka - c - nk (24) r r r r :'he initial capital stock is the same for both g.oups: ko = k. To avoid t'.e problem of time Inconsistency, we assume that the "left" and the "right" both -cmmit to the optimal programs they calculate at time zero, and make no -hanges later on. From the calculations above, It Is easy to obtain that In steady state k1 > kr c < c (38) From (32), the steady state investment rate for the "left' is always larger than the one for the "right". If the "left" Is In power, the economy 14 experiences higher investment and lower consumption; If the "right" Is in :&"er, consumption is relatively high and Investment relatively lower. The y:l-ical change in consumption, investment and the investment rate is shcwn iiagramatlcally below: c k Figure I where E1 and Er are equilibrium points for the "left" and the "right" respectively. If the economy is currently In El, the power shift from the "left" to the "right" results In an immmediate upward Jump In consumption and in a reduction of investment; the new long-run equilibrium Is Er where capital stock Is lower than, but consumption Is higher than, the equilibrium levels at E l The Investment rates fluctuate following the political power shifts. This is a demonstration of political Investment cycles at steady states. Investment cycles can also happen on the paths converging to the steady states. In Figure II, Pr and P are the optimal convergent pathes for the 'right" and the "left" respectively. c Figure II 15 .f the economy is Initially on the path for the "right" P , a change in Dlitical regime from the "right' to the "left" leads to an immediate Jowrnward 'Jmp from the path P to the path P Throughout time, the economy follows a ,-mpr I . i ,zag path, and Investment rates fluctuate accordingly on the path. V. Historical Evidence In this 3ection, we present a preliminary empirical study of the effects of political change on the Investment rate In China. The labels for different wings of the communist party, the "right" and the "left", are well known in China. The "left" consists of strong, dogmatic adherents of socialism; they advocate the centralization of economic activities, the rapid abolishment of private ownership In the industrial sector, and the rapid transition of the agricultural sector from prlvate ownership to collective ownershlp and then to state ownership. Chairman Mao Is the symbol of the "left". Those on the"right" are more often associated with economic policies with a capitalist flavor, such as relying on market mechanisms and material Incentives In the planned sectors and allowing private plots and contract systems In agricutural production. The prominent members of this group are Liu Shaoqi and Deng Xiaoping. They were known as the capitalist representatives In the communist party during the Cultural Revolution. The power struggles between the "left" and the "right" have shaped the hlstory of China In the past four decades, and their effects can be seen in every aspect of Chinese society. Our present focus Is on the effects of these struggles on the Investment rates. The following table contains relevent data for our analysis. 16 investment consumption productive power regimes year rate as % rate as % Investment as % of as a dummy of GNP of GNP total investment variable 1952 21.4 78.6 50.8 0 195J 23.1 76.9 49.4 0 1954 25.5 74.5 50.3 0 1955 22.9 77.1 51.4 0 1956 24.4 75.6 71.0 0 1957 24.9 75.1 58.8 0 1958 33.9 66.1. 82.3 1 1959 43.8 56.2 86.9 1 1960 39.6 60.4 97.4 1 1961 19.2 80.8 78.5 0 1962 10.4 89.6 63.6 0 1963 17.5 82.5 63.9 0 1964 22.2 77.8 60.8 0 1965 27.1 72.9 70.7 0 1966 30.6 69.4 68.9 1 1967 21.3 78.7 82.2 1 1968 21.1 78.9 78.5 1 1969 23.2 76.8 76.2 1 1970 32.9 67.1 71.8 1 1971 34.1 65.9 76.2 1 1972 31.6 68.4 78.7 1 1973 32,9 67.1 73.7 1 1974 32.3 67.7 75.4 1 1975 33.9 66.1 73.4 1 1976 30.9 69.1 79.3 1 1977 32.3 67.7 70.9 1 1978 36.5 63.5 71.8 1 1979 34.6 65.4 64.1 0 1980 31.5 68.5 54.5 0 1981 28.3 71.7 46.8 0 1982 28.8 71.2 46.4 0 1983 29.7 70.3 52.5 0 1984 31.2 68.8 58.6 0 1985 33.7 66.3 57.7 0 Source.---Statistical Year Book of China, 1986 In the table, the power over economic planning Is represented by a dummy variable; a value of zero means that the "right" controls the planning board, while a value of one means that the "left" controls the planning. The term "productive Investment" is special to Marxist and socialist economics, and needs some explanation. It refers to Investment that directly serves material production or meets the needs of materlal production. Its counterpart Is 17 non-productive Investment, which includes Investment on public utilities, nousing, pub'ic health, social welfare and education. Since non-productive investment Is more or less related to people's consumption, especially durable and public consumption, the percentage of productive investm- t in total Investment outlay is a more accurate measure of accumulation. The fluctuations in inves.ment rate and the productive investment rate are depicted In Figure III below. toot so - 70 00 20 ;2534 5567 51Dw6l 16265 466467 6a D 7177273 7,67S 76r7 77a79808 Si 82jS 85 Y R Figure III During the period 1952-57, economic decision-makings was more under the control of the "right" as Mao did not totally dominate the planning processes and political life was more democratic In the communist party than lt subsequently became. The Investment rates were in the range of 21.4 percent and 25.5 percent. The average share of productive investment In total Investment was 55.3%. Both production and consumption went up rapidly in those six years, and those Investment rates were later regarded as the optimal or 18 normal ones. The year 1957 was a turning point in the political ^llmate of China. The anti-`rightist" movement launched by Mao had a fundamental effect on the political and economic life of China. With the beginning of the "Great Leap Forward" In 1958 and of the movement of people's communes some time later, economic planning was dominated by the ideology of the "left". The Investment rate jumped up to 33.9 peicent, 43.8 percent and 39.6 percent In 1958, 1C59 and 1960 respectively. The average share of the productive Investment for those three years was up to 88.3 percent. High Investment rates and natural calamities during this period caused poverty, hunger and death all over China. Facing economic disaster, Mao retreated from economic planring and even admitted to having made a mistake in 1962. The power over planning shifted back Into the hands of the "right", and the Chinese economy entered a period of adjustments. From 1963 to 1965, the average investment rate was set at 22.7 percent and productive investme S only accounted for 64 percent of total Investment. President Liu Shaoqi even introduced many programs In agricultural production wl'ich later under Deng Xiaoping became important Ingredients of economic reforms. The reign of the "right" was short-lived. The next ten years, 1966-76, were those of the "Great Cultural Revolution", and Mao and the "left" were In absolute control of economic planning. Except for the years 1967-69 when the economy was almost paralyzed by destructive political turmoil, the Investment rate on the average was above 31 percent, and 75 percent of which was for productlve purposes. After Mao's death, his chosen successor. Hua Guofang, 19 * antinued the expansion drive of the "left", and even started a "Foreign Leap -Krward" from 1977 to 1979, importing large amounts of foreign technology The ivt!rage investment rate was above 34 percent. r,, 1979, politlcal power began to shift back to the "right", and Deng Xiaoping and the "reformers" came to the forefront, though the Ideology of the "left" still deeply affected planning and the effects of the "Foreign Leap Forward" stlll kept the Investment rate at a high level of 34.6 percent. But in that year, the proportion of productive investment In total Investment began to decrease. From 1981 to 1985, the average Investment rate went down to 30.8 percent, and the average share of productive investment was at a historical low - 52.4 percent. That is to say, a large proportion of Investment was diverted to the Improvement of residentlal conditions, service sectors, public health and education. So we can see that the Investment rates and political changes are closely related In China. It Is convenient to test how much fluctuations In investment rates and productive Investment rates can be explained by the political changes In China's socialist history. Here we report results of a few regression equations: It = 11.23 + 4.081)D + O.55ItI (39) (2.84) (2.05) (3.36) 2 R = 0.50, DW = 1.13 PIt = 31.25 + 12.36Dt + 0.45PI (40) (4.47) (5.17) (4.35i ( R2 = 0.73, DW = 1.93 where It= Investment rate at time t, Dt = dummy variable of political change (a value of one refers to the "left" regime and a value of zero the "right" 20 regime), and PI.= share of productive Investment In total investment. E:quations (.9) and (40) both show that political changes have substantial effects on the investment rate and the productive investment rate. 1he positive coefficients say that a "left" regime causes high rates, and a 'right" regime leads to low rates. As Invest.ient projects ofteni last for a few years, the lagged varlables also help to explain the rates. If we exclude the politically abnormal years 1967-69, then political changes alone can explain about half of the variations In the Investment rates: It = 25.36 + 8.9Dt (41) (18.17) (5.19) R = 0.43, DW - 0.74 PI = 58.32 + 19.12D (42) (28.14) (6.51)t R2 = 0.57 DW = 0.93 Two points should be added to our analysis of political Investment cycles in China. First, the "right" and the "left" are both expansionists by definition because they are both social planners, the difference being only a matter of degree.Throughout time, there Is a tendency for social planners to Increase the investment rates; this can be seen from regressing the investment rates against a time variable: It = 5.07 + 4.22Dt + 0.48It 1 + 0.114TIME (43) (1.02) (2.21) (2.56) (1.135) 2 R = 0.52 DW -1.13 21 Second, political factors as an exogenous variable cannot fully exolain all fluctuations in Investment rates; a theory espoused by T. Bauer (1978) and Tanos Kornal (1980, 1988), which we may call as model of endogenous investment .,ycles, has developed to explain investment cycles under the same political regime. The focus of this theory Is to relate the Investment rate to the intensity of shortage In the economy. Social planners will reduce the Investment rate when shortage intensity Is high, and raise the Investment rate when shortage intensity Is low. For a model developed In this line, see Zou (1990). These two theories of investment cycles should be taken as complementary, and "they can be usefully and effectively placed side by side and, taken together, they do a good Job not only of explaining the regular pattern of the cycle, but also of explalning Its Irregularities" (Kornai, 1988, P.284). VI. SUMMARY In this paper, we have offered a positive growth model that sheds considerable light on the "norms" of socialist growth, such -s investment hunger, expansion drive, chronic shortage and Investment cycles. This model also provides an analytical framework within which to study the relationship between Investment fluctuations and political changes In socialist countries. Preliminary empirical work on China has provided strong support for this approach. 22 References Alesina, Alberto: "Macroeconomics and Politics", In Stanley Fischer (ed.): NEER Macroeconomics Annual 1988, MIT Press, 1988. Alesina, Alberto and Jeffrey Sachs: "Political Parties, and the Business Cycle In "he United States, 1948-1984", Journal of Money, Credit and Banking, Vol.20, February 1988, pp.63-82. BaJt. Alexander: "Investment Cycles In European Socialist Economies: A Review Article", Journal of 'conomic Literature, Vol.9, 1971, 1:53-63. Bauer, Tamas: "Investment Cycles In Planned Economies", Acta Oeconomica, 1978, pp.243-60. Cass, David: "Optimum Growth In an Aggregate Model of Capital Accumulation", Review of Economic Studies, Vol.32, July 1965, pp. 229-43. Grosfeld, Irena: "Endogeneous Planners and the Investment Cycle In the Centrally Planned Economies", Comparatlve Economic Studies, Spring 1986, pp. 42-53. Grosfeld, Irena: "Modelling Planners' Investment Behavior: Poland, 1956 - 1981", Journal of Comparative Economics, Vol.11, 1987, pp.180-191. Ickes, Barry: "Cyclical Fluctuations In Centrally Planned Economies: A Critique of the Literature", Soviet Studies, January 1986, pp.36-52. Koopmans, I.C.: "On the Concept of Optimal Economlc Growth", In The Econometric Approach to Development Planning. Amsterdam: North-Holland, 1965. Kornai, Janos: "Rush versus Harmonic Growth", Amsterdam, North-Holland, 1972. "Econoazlcs of Shortage", Amsterdam, North-Holland, 1980. "Some Properties of the Eastern European Growth Pattern", World Development, Vol.9, Nos.9/10 1981. "Growth, Shortage and Efficiency", University of California Press, 1982. 23 "The Hungarian Reform Process: Visions, Hopes, and Reality", Journal of ;.conomic Literature, Vol.26, 1986, pp.1687-1743. ''Lecture Notes on Socialist Political Economy", Harvard University, unpubished, 1987, 1988. ..rz, M.:"Optimal Economic Growth and Wealth Effects", International Economic Review, Vol.9, October 1968, pp.348-57. McRae, R.: "A Political Model of Business Cycle", Journal of Political economy, April 1977, pp.239-64. 4Jordhaus, William: "The Political Business Cycle," Review of Economic Studies, April 1975, pp.169-90. Orzechowski, W.: "Economic Models of Bureaucracy: Survey, Extension, and Evidence", in T.E. Borcherding (ed.): Budgets and Bureaucrats, Durham, Duke University Press, 1977. Phelps, Edmund: "The Golden Rule of Accumulation: A Fable for Growthmen", American Economic Review, Vol.51, September 1961, pp.638-43. ---: "Golden Rules of Economic Growth," W.W. Norton & Company Inc. New York, 1966. Ramsey, Frank: "A Mathiematical Model of Savil.g", Economic Journal, Vol.38, December 1928, pp.543-59. Roland, Gerald: "Investment Growth Fluctuations In the Soviet Union: An Econometric Analysis", Journal of Comparative Economics, Vol.11, 1987, pp.192-206. Statistical Year Book of China, State Statistics Bureau, Beijing, China, 1986. Zou, Heng-fu: "A Model of Bauer-Kornai Investment Cycle Theory", mimeo, the World Bank, 1990. 24 PRE Working Paper Series Contact II& Author La or 2agee WPS596 The Mexican Sugar Industry Brent Borrell February 1991 P Kokila Problems and Prospecis 33716 WPS597 Rent Sharing in the Multi-Fibre Refik Erzan February 1991 G. liogon Arrangement: Theory and Evidence Kala Krishna 33732 from U.S. Apparel Imports from Ling Hui Tan Hong Kong WPS598 Africa Region Population Projections: Patience W. Stephens February 1991 0. Nadora 1990-91 Edition Eduard Bos 31091 My T. Vu Rodolfo A. Bulatao WPS599 Asia Region Population Projections: Eduard Bos Februiary 1991 0. Nadora 1990-91 Edition Patience W. Stephens 31091 My T. Vu Rodolfo A. Bulatao WPS600 Latin America and the Caribbean My T. Vu February 1991 0. Nadora Region Population Projections: Eduard Bos 31091 1990-91 Edition Patience W. Stephens Rodolfo A. Bulatao WPS601 Europe, Middle East, and North Eduard Bos February 1991 0. Nadora Africa Region Population Projections: Patience W. Stephens 31091 1990-91 Edition My T. Vu Rodolfo A. Bulatao WPS602 Firm Output Adjustment to Trade Mark A. Dutz February 1991 S. Fallon Liberalization: Theory with 38009 Application to the Morozcan Experience WPS603 The Role of Officially Supported Ash L)emirguc-Kunt February 1991 G. llogon Export Credits in Sub-Saharan Refik Erzan 33732 Africa's External Financing WPS604 Foreign Trade and Its Relation to Faezeh Foroutan February 1991 S. Fallon Competition and Productivity in 37942 Turkish Industry WPS605 Overview of Contractual Savings Dimitri Vittas March 1991 W. Pitayato- Savings Institutions Michael Skully nakarn 37666 WPS606 Adjustment Policies and Investment Luis Serven March 1991 E. Khine Performance in Developing Andres Solimano 39361 Countries: Theory, Country Experiences, and Policy Implications PRE Working Paper Series Contact El: Author f tor 1Der WPS60, Aboo h'g Green Rates The Effects Donald F Larson March 1991 D Gustafson on Cerei s. Sug:ir. and Oilseeds Simon Glance 33714 In W-st Gurminay Brent Borrell Merlirda Ingco Jonathan Coleman WPS608 Cross-Country Studies of Growth Ross Levine March 1991 CECMG and Policy: Methodological, David Renelt 39175 Conceptual, and Statistical Problems WPS609 A Sens t˘vity Analysis of Cross- Ross Levine March 1991 CECMG Country Growth Regressions David Renelt 39175 WPS610 Can Preshipment Inspection Offset Alexander J. Yeats March 1991 J. Jacobson Noncompetitive Pricing of Developing 33710 Countries' Imports? The Evidence from Madagascar WPS611 Tariff-based Commodity Price Jonathan R. Coleman March 1991 S. Lipscomb Stabilization Schemes in Venezuela Donald F. Larson 33718 WPS612 Education and Productivity in Lawrence J. Lau March 1991 WDR Office Developing Countries: An Aggregate Dean T. Jamison 31393 Production Function Approach Frederic F. Louat WPS613 Price-Wage Dynamics and the Simon Commander March 1991 0. Del Cid T ransmission of Inflation in Socialist Fabrizio Coricelli 39050 Economies Emprical Models for Hungary and Poland WPS614 Accountability in Public Services' Samuel Paul March 1991 E. Madrona Exit, Voice, and Capture 37496 WPS615 Socialist Economic Growth and Heng-fu Zou March 1991 A. Bhalla Political Investment Cycles 37699