A Vision of the Growth Process Amnold C, Harberger ‘The American Economic Review, Vol. 88, No, 1, (Mar., 1998), pp. 1-32. Stable URL: http links jstor-orp/sici sici=0002-8282%28 199803 %2988 G24 L%ACLIAAVOTGPSIED.0.CO%SIB2-5 ‘The American Economic Review is currently published by American Economie Association, ‘Your use of the ISTOR archive indicates your acceptance of JSTOR’s Terms and Conditions of Use, available at huff www, jtororglabouthermsshtml, ISTOR's Terms and Conditions of Use provides, in pat, chat unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and ‘you may tise content in the JSTOR archive only for your personal, non-commercial Use, Please contact the publisher regarding any further use ofthis work, Publisher conta information may be obtained at Hp jstororpouinalsfaes hn Bach copy of any patt ofa JSTOR transmission must contain the same copyright notice that appears on the sereen or printed page of such transmission, ISTOR isan independent not-for-profit organization dedicated to creating and preserving a digital archive of scholarly journals. For more information regarding ISTOR, please contact suppon @jstor org. bupifevwo jstor org! ‘Thu Nov 12:14:37 2006A Vision of the Growth Process' By Axnotp C. HaRBERGER* One of the great pleasures of belonging to samy generation of economists is that we were “Presidents Address delivered at the one-hundred veo meeting af the American Economic Associate, Tua 6, 1998, Chicago, Department of Eesnomics, University of Califor, ‘Los Angeles, CA 90095. This baper i But one ste in sequence af waitlags and ater presentations on the pro ‘ceseot growth Inthis work [ave beon greatly helped hy {Comber of retearch ansstnts Luts Alvarado and Haale Beyer ithe early sages, Alfonso Guera and Eagat Robles in the eile pase and Enrique Flees and [Lennard Tote ithe final tage, a this per was being ‘prepared. Renders wil roe tha be pager also cram on Pu. desertion worksdoneby Beyer, Robles, ad Tore Inthe course of the evolution st refeied to, Thave made Pesenations af and received taluable comments fom, fue of diferantforuns— seminars at Clerc Uni eqs, Sanfoed University, Texas A&M Lisversty, and UCLA inthe United States, plus several braad-—the Cathie University of Cle he Uaivesty of Chile and the Ceno de Estudios Piblico (CEP) ix Santiago, he CCenee for Atgentine Macroeeanomle Studies (CEMA). the Inwenuta Toreasto Dk Tella, the Univesided de San “Ants, andthe nstinto Superior de Becnomistas del Go= bier (SEG) en Argentina, as wel asthe Instuto Tez folgeica Aulénemo ce Mexica (ITAM) ia Mexico. Ia ‘dion, tere were conference pesenations argarizat by Comeit Uniecsy, te Bas West Cente (Unverty of Rawal) he Association of Asian Bconoruists (Kuala Tumpur, 1986), te Wesern Beanomic Asrocation (Se ale, (997), the Eeonometic Society (Lane. America ‘meetings, Saniaga, (997), ané the Argentine Poiscl eonomy Associaton (Bahia Basce, 1997). Tanto wake special noe of conference on esta oneaton and eeanoniie pYOWES rpsnized by Mca! eskin and held at the Hlaover nation Oct 1997 ‘This conlecence was wien by « vetiuble glany of ee riots sceas of growth, 2 fat which motvaled me tO ‘wea very lang paper (Harberper, 1998) pring isues of mathocology in the measurement and analysis of srowh, a well as tbe mote substnlive waters ema ‘Sandia the reser pape. [ake this opportu 1 ele teacerst that paper (lerhcoming in 2 volume edited by ‘Boskin for more dealed trearment of methoéotopieal is sues also want give special thanks to Zvi Glicbes, Dale W. Jorgenson, and Paul Rammer as well as © my ‘collespucs Seasian Edwards, Jean Laweet Rosen, and Carles Vegh, each of whom gave ths pape a ery ‘izful realing td protided me with exaemely uel comments, Finally, T want co give special thanks co Conaine Grams. who has performed maces wansaing Ihandwriten serauls fed fem Tour contents ilo he paper you See before you, able to witness the birth and the subsequent evolution of the modera approach to the analysis of economic growth. The center- piece of that approach is probably growth accounting, but we should never forget that growth accounting is firmly rooted in eco- ‘nomic theory ‘My way of telling the story goes like this: “Many, maybe even most, economists expected that increments of output would be explained by increments of inputs, but when we took our best shot we found that traditional inputs typ- ically fell Far short of explaining the observed output growth, Our best shot consisted in at- Icibuting to each factor a inanginal product ‘measured by its economic reward. Thus: () pay = WAL + G+ HAK+R. Here: ‘Ay = change in output (GDP); ‘AL.= change in labor inputs nti general price levels tial teal waa, initial real rate of return to capital; 6 = rate of real depreciation of capital; AK = change in capital stock, and R= “ihe residual” of growth unexplained by increases in traditional inputs. Many economists are probably more famil- iar with a variant of (1) (14) (Ayty) = (BLABY) ALIL), + (B+ 8) Kipy | (AK/K) + (Rly) = se OLIL) + se AKIK) + (RIY) In whichever fom, the measured residual typically accounted for an important fraction of the observed output growth, quite often half or more.‘This result came as 2 surprise to the profes- sion, though perhaps less’ so to thas who reached it, or something very like it, by an al- temative route. They were the people who came at the problem out ofa tradition of mea- susing labor productivity, and at some point complemented output per worker with a mea- sure of output per unit of capital, and finally “Joined the wo fo create a measure of total fac- Torro). The ides oft ste ‘productivity increasiag through time was less 2 shock to these people than the ‘growth re- sidual” was to diose wito approached its mea- ‘surement along the lines of equation (1) or (L'}. See Moses: Abramovitz, (1952, 1956) and Solomon Fabricant (1954) In amy case, asthe newly discovered residual Joomed large in our professional thinking, our discussion centered on two potential explana- tions: “human capital"” and “technical ad- vvanee.”* (See Robert M, Solow, 1957.) These ‘can be thought of a5 complementary explana- tions, at least up to a point, with technical ad- ‘vane: representing truly new ways of doing things, and the accumulation of “human eapi- tal” representing increases in the “quality” of the typical human agent, H was not long before attempts were made to quantify the contibution ‘of improved labor quality. These came. as part of a general move toward disaggregation of the twa factors, which can be represented by: (2) Pay = Em, AL, +20, + DAK, +R. Here the index i can vary over all sorts of ed- tucation and skill groups as well as categories like gender, age, occupation, region, etc. All these are items that may signal a diferent mar- ket wage, In a similar vein, the index j would appropriately vary over categories like the car- porate, noncorporate, and housing sectors whee, for x if for no other reasons, different (gi0ss-of-tax) rates of return would presum ably prevaif, even in a full equilibriy erage quality can be Lal Baby, and the cou tribution of change in quliy to 2 between tand f+ 1 can be calculated as ZW)(AQ,..f Q)). Thus, the contribution of quality change THE AMERICAN ECONOMIC REVIEW manent 1938 is already built into the first summation in (2), but can he separately identified if we so choose. ‘A focus on lnuman capital could fead us t0 a slightly different way of breaking down Bw, AL. Here we could choose some “basic wage” w*, ideally the wage of some well- defined category of relatively unskilled labor, ‘Then we could divide the remuneration x; oF any given category into a part w* which was 2 reward for “raw labor" and another part Gs; — w*) which we would identify as the rewatd to the human capital of a typical worker of type i Using framework like (2) has tong been the standard for careful professionals. Pio~ neered by Zvi Griliches (1960, 1963). it was utilized by Edward F. Denison (1967) and John W. Kendrick (1973, 1976, 1977), among ‘others, This approach has cen further devel- ‘oped and carfied (0 a high art by Dale W. Jorgenson and Griliches (1967 ), Jorgeason et al (1987), and Jorgenson (1995) ‘The main point co be made here is thst once By direct conteibution moan what people ate paid for. Doctors earn more than arses, and engineers more than drafismen, These and similar differences are captured in Ei.AL,, which can be positive even if EAL, is 2ero, just from an upward re- shuling of the same labor force. A truly ac- curate measurement of type (2) would capture all the subtle differences of quality that exist in a modern Ishor force and would give each 4 weight corresponding 10 the (gross-of-tax) . This is a quite important point for it tus to zero in on the residual as repre- senting “technical change,"" ““TEP improve permil ment,” and “real cost reduction.” ‘There is no analytical reason to prefer one of the above three terms over another, in re~ ferting to the residual 2’. But J ant going a bit ‘out on a limb to say that a term like “technical change" leads most economists fo think of iz- ventions, of the products of research and de-VOL. #8 80.1 velopment (R&D), and of what we mis echnical innovations. On the other band, improvement, once purged of the changes in the quality of fabor and/or the direet eontei- bo afin capal mes e o, extrait difoen Kindo TR and ly elensieaa ony mind, makes one chink like an entrepreneur oF 4 CEO, or a production manager. T think it would be perfectly fair to char- acterize my presentation today as a paean in praise of “teal cost reduction’”’ asa stmdard label for R’, Labels do not change the tn- derlying reality, tut they may change the ‘way we look at itand the way we think about it. They algo can lead us to understand it bet- ter. Thinking in terms of real cost reduction has certainly done all this for me, as I have ‘tied to sort out the many puzzles and com- plexities that surround the process of eco- nomic growth Let me try (0 take you down the path I trav led. In the first place, reat cost reduction {RCR) is probably an the mind of most busi- ness executives, production managers, tc. at some point or another in any given week, et alone in any given month or year. It is a major ppath (0 profit in good times, and 4 major de~ ‘ence against adversity in bad tines. Most U.S firms that have downsized in recent years did so with RCR in mind. Sa, too, did the firms that computerized their payrotis and other ac- counts. And sa also did thase. who shifted to What they considered more modem manage- ment techniques. I recall going through a clothing plant in Central America, where the owner informed fic of a 20-percent reduction in teal costs, Following upon his installation af background music that played as the scam- stresses worked. And then there is the stary of two Chilean refrigerator firms that ended up as parts of a single conglomerate at one point. The new management reduced the number of models from something like 24 to two, making ‘agreements to import other models while ex porting these two. The end result was that out- ppat more than doubled, while the lahor force was cut to less than half, and even the capital HARBERGER: & VISION OF THE GROWTH PROCESS 5 api wack weit dow And ‘we all have seen ‘cases where, say, an office's real costs were reduced when a mavinet oF & manager was Te placed by someone more reasonable) But we" hhave also seen cases where real costs were re- duced when a very lax manager was replaced id (say, a decade). ‘Once one accepts this propasition as true, the ‘question then arises: Why would anybody try Co settie on just one underlying cause of real cost reduction? The answer, F think, is mind: the framework ia which one is thinking ‘atthe rnoment, The pioneer writings ofthe re sent endogenous growth literature can, (chink, be said to reflect a kind of annoyance at some- thing like Ror R being considered exogenous. ‘There was an urge to surmount tat inelegance by somehow making the residual endogenous. ‘And in a siniple growth model that meant gen ‘erating a feedback from the rest of the model to the residual. A 1001 feedbacks would be aut of the question, but one feedback would work just fine. Thus Paul Romer (1986) focused on 4 feedback through “knowledge,” with the stock of knowhedge shifting production func tions all over the economy; Robert B. Lucas, Jc, (1988) focused on “human capita,” not ‘ons ditect and remunerated productivity, but ‘on the externalities that each inerease in the stock of human capital were presumed to gen- crate. These single feedbacks achicved the Timited purpose of endogenizing R or R” within a specified model, but they did not rep- resent very well the multifaceted nature of real cost reduction as we observe it in actuality. And, in point of fact, both the cited authors in their more recent writings display a deep rec- ognition of the subtlety and complexity of the growth process, not really capable of heing ‘captured through a simple feedback mecha- nism. (See Romer, 1990, 1994a, b; Lucas, 1993.) So, real cost reduction is multifaceted and everywhere around us. Where does that get us? Or how can we get anywhere in the face of such complexity’?4 THE AMERICAN ECONOMIC REVIEW MARCH 1996 ‘Taats 1Growre Bazaxooin Trearite Rest Cast Renucrion As AbOeTV ‘edition Indusey ax) @ L ‘a00 ‘0. 2 00 20h, a 500 Sisob, All hens oor sab, “he mae ade oe a quick appreciation of Js, assume tat lta factor productivity ew ty 80 percent in ane industry over a decade, ‘by 6 percent in another industry, and by 50 percent in a third, If thir initial value adcted amounted to $100 billion, $200 billion, and $300 billion, respectively, then the real cast reduction of the frst was $80 billion, that of the second was $120 billion, and that of the thied $150 billion. So we can say that, mea sured at initial prices, he real cost eduction of the three together was $350 billion over the decade in question. I truly think that the notion of real cost reduction being additive in this way came (o my mind, and is easily seen by ‘others, just as a consequence of the label. The idea of additivity does not follow nearly so easily from the labels “technical advance” fd “total factor productivity.” ‘Anyway, this Vision of the growth process ‘opens up many new vistas and gives us many ‘new challenges. To me it gives life to the re~ sidual, viewed as rel cost reduction, in a way that remote macroeconomic externalities never did, It gives the residual Dody, in the sense thatthe nuraber of dollas saved by real cost reduction is a tangible and measurable ‘Quantity. I gives the residual a name (cel cost reduction), an address (the firm), and a face (the face of the entrepreneur, the CEO, the production manager, etc) And, finally, we shall sce that tuere can be vasly diffevent ex- pressions on that face, even as we move from firm (a frm in a given industry, as the TEP expetience of a period moves from sharply positive to devastatingly negative. Cum sural Ce sta ay tule ate o @ o sea, sion, tone 00h, 5200 $3006, ‘80h, 20th 600, 50h, 514006, 0006. 1. Yeast versus Mushrooms: Part Table 1 is based on the numerical example just given, plus the inforntation that the. re maining industries (say, in the economy) to- gether had an initial value added of $1,400 billion and expericnced real cost reduction of '$150 billion over the period. Setting out data in the format of Table 1 allows us to. make statements like “15 percent ($300 b/$2,000 bof the industries (measured by initial value sdded) accounted for 40 percent ($200 b/ $500 b,) of the real cost reduction (RCR) of the period” and “30 percent ($600 6 /$2,000 'h} of the industries accounted for 70 percent £350 b./$500 b.) of the period's RCR.”” stumbled on this way of presenting data on real cost reduction in the course of writing. a background paper (Harberger, 1990) for the World Bank's Warld Development Report of 1991, Once I saw it, T immediately embraced it because it helped me communicate co others what I call the "yeast versus mushrooms" believe that a “yeast” process fits best with very broad and general externalities, like externalities linked to the ‘growth of the total stock of knowledge or of ‘human capital, or brought about by economies of scale tied to the scale of the economy as aVOL. 88 NO. UARBERGER: A VISION OF THE GROWTH PROCESS 5 ‘Taare 2—Concentnacian ob TEP Qxawnn AMONG US, InouscHtes 1958-1967 1G sts (2) £6 (5) BILO8 OF 1958 Dower) ‘TPP growth over pertod ¢ Lumber & Weed Prost on ailtoad Teanspor. 063 “Teale Mill Prods 061 lecrical Mache 958 “Tuanspor Equipment as Chemicals ous Public Utes on, Peasleam and Coal oat uber and Products oat Mining oat Comnmnication ao ‘Tate am “There follow 18 more ids the combined reals of which are aa ‘Absolute arco ‘teal cum apeey) Cun. reduc! sum of indus | sum of ox 3 a 42 2 © 21 261 ast 480 532 ah am nan 249 loss 4101630 5.19 1366 9.30860 70s gsm _ santo 39 asst 904990, 465 sist tan 90 17 me 0 0 La ase 0 67.0 520 ne 7860 36 so #860 93 764016500 1 710 23980 40440 ‘Nov: Tp 10 percent these percentges ate contbutons to GDP af industries ranked acon vo thee presente of "TEP growth over period) of isren account for 30 percent of ttal TEP contribution, “Top 22 percent these perceniage te Coinhulionst0 GDP of indusnes caked contin to thet percent xe of TEP greet Over period) of indus acount for 52 percent of toa TEP contribution, ‘Top 40 percent (hese percentages ate contabutions to GDP of idstiescznked acotding Co thet percent ae of THP such Over period) ofindusines account for 70 peteeat of tou TFP contnbution Sources: Reedtek ané Grossman (1980), GDP data fora U.S, rations! aezourt, reductions stemming from 1001 different causes, though I recognize that one can build scenarios in which even 1001 causes could whole econoray. remember being impressed, when I first saw some early industry estimates of TRP improve- ment, by their tendency to industry concentra- tion, ‘For years 1 told my students that the 1920's were the decade of cars and rubber tires, the 1930's the decade of refrigerators, the 1940°s that of pharmaceuticals {especially ities), and the 1950°s that of television, (note that the with telecommunications and computers tak- ing over in recent decades. But these were just impressions, not based on any systematic ap- proach. My real tenaround came in the course, of writing my 1990 paper, where I presented 4 series of tables based on Kendrick and Elliot, 8. Grossman's (1980) work. Table 2 is an example, ‘Table 2 has the same format as Table |. Col- umn (1) presents the familiar measure of the percentage by which TEP gress, ot rcal costs ‘were reduced, duting the period in question percentages apply to the period 1958-1967 as a whole; they are not annual6 ‘THE AMERICAN ECONOMIC REVIEW rates). To turn these percentages into dollar amounts of real cost saving aver the period, ‘one multiplies them by base-period real GDP [col. (4)]. The results are shown in column (2). Columns (3) and (5) are the cumulative sums of columns (2) and (4), respectively Working with these figures one can make statements like those at the bottom of the table—ie, the top 10 percent of industries accounted for 30 percent of total real east re- duction; the top 22 percent of industries (mea~ sured by initial value. added) accounted for ‘more than balf of total real cost reduction, Readers will notice that at the foot of each column in the table is an entry referring to 18 auditional induscis, which together accounted foronly 10 percent ofthe total TEP contribution, while their combined share of initial output was atmast 60 percent of the total, Using the analogy wits yeast and mnsbrooms, the results of my calculations using the Kendriek-Grossman data pointed very clearly to a “‘mushrooms’” interpretation. Not only were the contributions to RR highly concentrated in a relatively few industig3, these industries also were very different as one shifted from decade span to decade span. The top four branches in percentage of real cost reduction during, 1948— 1958 were Communications, Public Utilities, Farming, and Miscellaneous Manufacturing. in 1958-1967 trey were Lumber, Railroad Trans- Port, Textile Mill, and Electrical Machinery. in 1957-1976 they were Finance, insurance. & Real Estate, Apparel, Communications, and Chemicals. Only Communications appears twice Timon won pocaral preting alized externality due t0 i Broved edition won beard jsily tsi it So also ¢proba- tly) would economies of seale associated with the scale ether of the frm or of the industry "oc aseview of he coven ss of analysis of RAD expends ard their amgnct em eonamuc growth See Gelehes (1994) MARCH 1988 Such economies are nat likely to jump wildly around from one industry to the next, from pe~ rod to period. One would expect them: to em body characteristies of the productive process that would be relatively stable over time; hhence they should show a reasonably high de- see of persistence, over time, in terms of the ‘TFP experience of particular industries, No economist can look at Table 2 without trom Edgar Robles, 1997), shows the quasi- Lorene curves for a 20.indusiry breakdown of the U.S. manuucturing sector over four suc cessive five-year periods ‘Wha strikes ane iosmediately about Figure Lis the charactedstc “overshooting.” Thave marked with the first vertical lime the poist ‘where te rising curve crosses 100 percent on the vertial axis. The interpretation is that in 1970-1975 the cumulative real cost reduction of just 25 percent of manufacturing industies (oieasured by inital value added) was equal Co the otal RCR for manufacturing as a whole. ‘Alter that chere are other industries producing ‘another 40 percent ofthe total, but their cane {cibution is offset by stil her industries with negative RCR cing the period—————7 ‘Corresponciing to the 25-percent figure for 1975, we lave around 12 percent for 1975— 1980, 48 percent for 1980-1985, and 40 per- cent for 1985~ 1991. ‘These are the fractions of manufacturing industry which by themselves, were able to account for te full amount real, cost reduction during the respective period, in manufacturing industry as a whole, “The second vertical line in each panel of Figure, 1 marks the maximum point of the curve( The interpretation is that about 64 per- ent of industries enjoyed real cost reduction during 1970-1973) with the remaining 36 per- cent suffering real cost inereases (dec TEP), For the subsequent peciods, the sponding figures are 65(35) percent, 78(22) percent, and #2(18) pereent. Here the fis fi lire is the percent of industries enjoying reat 0st reduetions the figures in parentheses rep- resent those experiencing declining TEP. Some interest attaches to the ordinate ofthe maximum point on each curve. In the fest pe- riod, TFP growth ended up accounting for closeYor. 88 NO. 1 HARBERGER: A VISION OF THE GROWTH PROCESS 7 A. 1970-1995 . 1980-1985 1975-1980 D. 1988-1990 rouke 1, Paoetsss o° TP Grewrit AMONG US, MANUFACTURING BRANES (0 170 percent of the RCR for total manufactur- ing, In 19751980 this figure was about 240 percent, in (980~1985 only about half tha, and in 1985-1991 a little more than 125 peccent. ‘The trouble i that when the aggregate TEP con- tribution is relatively small, the cumulative total of the positive contsibutions is a large multiple of thar aggrezote, while when the aggregate is large, this multiple tends to be smaller, Thus, for 1970-1975 and for 1975~1980, the total RCR, in manufacturing as a whole was only about 23 percent of iatial manufacturing Value added. In Contrast the total RCR forall manufacturing was ‘almost [0 percent of initial manufacturing value ‘added in 1980-1985, and about 7.5 percent in 1985-1991. ‘The problem obviously becomes greatly com- pounded if the real cost reduction for the agare~ ‘ate (in this ease total manufacturing) tums out to be negative. Special conventions would have to be established to make clear the interpretation ‘of Lorenz-like diagrams jn such cases, T believe T have hit on a felicitous way of solving all these problems, and at the same (ime creating an even beter, clearer visual rep- resentation of the degree of concentration or dispersion of real cost ceduction among the components of an aggregate. The ides is simm- ply ( relabel the vertical axis of the axis would he calibrated accordingly. Thus, by looking at the slope of a simple chord, we could visually assess how rapid was the TP growth of the aggregate in ques Figure 2 is presented simply for didactic purposes. Here we have a hypothetical indus- (cial branch made up of four industries, A, B,6 THE AMERICAN ECONOMIC REVIEW cum. Sum Fissl Cost Raduetion nts of ita value ad MARCH 1988 ot [Cum Rate ot TEE Grown 008 lou 2, k.ustaative TEP Growra Proms (Sesvise Sunser Dracnans) mulative real cost reduction (a real dollar amount) and plot it against cumulative initial real value added. Then we scale the vertical axis so as to comply with whatever metric we have decided upon for the TFP growth rate (in the example, a 30-deptee line representing a Leperceat anual TRP growth rate), and the bborizoutal axis so as to add up to 100 percent. In the lower panel of Figure 2 T give ex- amples to show how these diagrams cope with the problems of a low TEP. growth rate (the overshoot for the ease of 0.25-percent growth ‘would show up peaking at over 400 percent inVOL 58ND. 1 HARBERGBR: A VISION OF THE GROWTH PROCESS ’ 6 ee 6 | + = ora a “ Flours 2 TRP Gaqwrs Peomies rom US. Manuencruaine0 THE AMERICA ECONOMIC REVIEW 4 Lorenaetype diagram) and of negative TEP srowth (where itis hard to even conceptualize a Lorere-type picture). T fist presented these diagrams before a large audience at the Westera. Economic As sociation meetings in Seatle July 1997) and for thar presentation coined the label of "“sun- rise diagrams” on their analogy with the sun rising over a hill. That same evening, Yoram Barzel suggested that where the aggregate slope is negative, we aj term "sunset ,"" which Timmediarely accepk Figure 3 presents a set of suntise-sunset diagrams based on Jorgenson et al. (1987 pp. 188-90). These cover 32 industrial sectors (heir 35 minus Agriculture, Trade, and Government Enterprises). I think the utility of sunrise-sunset diagrams needs no further championing once these pictures are exam- ined and digested, Practically all variants ae id i (1953— 1957 and 1969-1973); negative growth with large and moderate overshoots (1966— 1969 and 19731979}; moderate growth with, small (1979-1985), medium (1960— 1966), and large (1948-1953) overshoots. ‘One striking fact that emerges from this set of pictures is how variable across periods is the negative contribution of the losers. If the Josers had only contributed zero change in ‘TEP, we would nave had cumulative TFP con tributions of about 0.8 percent per annum in 1948-1953, in 1957—1960, and in 1960— 1966, And the ather periods would not have been much different: about 0,7 percent in 1953-1957 and in 1969-1973, 0.6 percent in 1966-1969, and 0.5 percent in 1973-1979 and 1979-1985, Instead of this narrow range of cumulative contributions, we have an actual distribution that goes from ~0.9 percent in 1973~1979 through around 0.1 percent in 1953-1957 and 1969-1973 to over 0.5 percent in 1960-1966 and 1979-1985. ‘Does this not suggest that we make & major research push trying to improve our under- standing of the phenomenon of negative. TPP ‘growth? What syndromes characterize the fms and industries experiencing. it? How much of it stems from external shocks like international prices? How much of it from competition ‘within the industry? How much of it represents firms struggting 10 survive, yet experiencing, ARCH 1998 output levels well below their previous peaks {and presumably below installed capacity)? How much of i€ represents things like “Isbor hoarding’ as firms go theaugh periods of adversity? TH Yeast versus Mushrooms: Part IT hope that in the previous section I have made a convincing case concerning: (a) the usefulness of sunrise-sunset diagrams, (b) the aptness of the ‘yeast versus mushrooms”” di chotomy, and (c} the pervasiveness with which the mushroom side of that dichotomy seems (0 come out ahead when the GDP is broken, down into industries or industrial branches for TRP analysis. The grand design that emerges from the studies reported here, and from just about all the other industry & a that I recall a cr is that +; (iti) the losers are ‘a very important part of the picture most of ihe Gla an conus ically toe vara tons ae asarve in agaregae TEP perfor mance; and (iv) there is little evidence of Fesbtees om ped w ped of tee: ets in TEP performance. “he above resis, Chink, very i teresting Cine sense of pguing our ear iy) iy strong in rtf th inp Satlone about the nature of the growth process) an very cobut (nthe seas na thoy have wide epolieaby ovr sitet dd cs calyeed y siferc authors wang feast sompeat diferent methods) Bt these rela, no tar ate ule comple with what I might call an “industry view"* othe TH say this the ay Lemgocl, Iboked atthe growth process wll Suite recession tba was eed ny Moe sbout rer eh and autos i Be 1520's, ehigertos and other household tpplancs in ine 1990's, phasnaceuteals ia Te BSS se The fnege th 1 had fe Bind was one of yout within eich indy thd ushvooms between indusiles a Sonny of TP capt by fs iin indus depending on that Induvou. a8 No. 1 HARBERGER: A VSION OF THE GROWTH PROCESS u os Cum Sum Cum, Real Rate of Cost TEP Reductign | Growh f.0s772 ‘0086 é 3 4 Porcentito of initial Value Added Fount 4, TEP Growra Prone wy Mexicas MANUFACTURING Sect. (1892 ESTABLIKMENTS, 1984-1994) fuck in the technological draw, side by side With highly diverse experience between indus- tries because the distribution of technical ad- vances had wide dispeesion, even for periods as long as a decade. Getting access to data atthe firm level per~ mits one to explore whether this view is eom- patible with the actual experiences of firms and industries, We are just in the early stages of this exploration, but I think the result is quite clear already; namely, the “mushrooms” sary pre- vails just as much among firms within an in- dustry as it does among industries within a sector or broader aggregate. I will present here ‘only a taste af the evidence from the United ‘States (on which our systematic work just re- cently got started). Our massive evidence ‘comes from the Mexican manufacturing sector, for which Leonardo Torre (1997) has analyzed data from a sample of over 2,000 firms. A small fraction of these fiems were lost owing (0 miss- ing data, but same 1,900 firms remained inthe sample that Torre finally worked with, These Firms were divided into 4 branches of industry, 0 that on average we have about 43 firms pec branch. ‘There are really too many ways (© present such a mass of information as is contained in ‘Torre's study. What I will do here is give the ‘aggregate picture in Figure 4, and then show in Figures 5A~C three fast-growing branches, three of around median grow, and three from. among the slowest-growing branches ‘To complement these figures. I finaly pres- ent, in Figures 6-D, certain summary statis- tics from the suntise-sunset diagrams ofthe 44 bbranches that Tome studied. Here Figure 6A fives the distribution of average rates of TEP growth among the 44 industries. Figure 63 shows the distnbution of peak cumulative con- tributions, i.e., what the TEP contribution ‘would have been had all tue negatives been zet0s, Figure 6C displays the percentile of firms (by intial value added) marking the bor- derline between positive and negative TEP growth. And finally, Figure 6D shows, for branches with positive TFP growth, the per ‘centile of firms which, by themselves, account for 100 percent of the industry's TFP growth. ‘This evidence almost seems to replicate, for firms within an indusiry, what was found in the previous section for industries within theRn {HE AMBRICAN BCONOIKIC REVIEW ARCH 1998 Mator Manufoouing: St Orcs {66 Etabiahments] 26 cum, Bes 09 | ae ore 7 ‘aoe Growth 4 + ° | I | I I i i | | | I 3 Rok 2 2 y Pence iti Value Add Mevcen Mantachsing. Cement Establahments} oo cum Fate ares i es? Growth < oso ° ° | T ~ i i 8 | | : | 06 I | e ¥ & ¥ T Percent of iia Value Aled Mascon Manufacturing Othe Wood Prd. [5 Estaishments] ce Manetaing Ober Wore Prods. 18 Eveaienments) cum, Rate ot TF. 3 one oat ° + | | | | | oe t 6 Perce ital aun Adal 7 Fours 5A. TEP Grown: Proms ror Fast-Grownes Branches (Mexican Mavcunacumi, 1986-1984) Sunset diagrams, great influence of firms with negative TEP growth in determining the TEP ‘outcome for an industry, and a small or mod- crate fraction of firms accounting for 100 per- cent of the TEP growth of an industry (when that growth is positive}, with the comptemen- tary fraction being winners and losers whose efforts end up just offsetting each other. Itre- mains (0 try to give some interpretation 10 those resuls. JHE, “Just Rrrors” or t's 3 Jumgle Gut There?” ‘The first question that will enter the mind of many economists on looking at the evi dence presented so far is: how much of whatvor a 0 1 HARAEEGER A VSION OF Tut GROWTH PROCESS 2 Nevcon Manchin ats [6 Estates] 6 com F005 on : of TF 0253 oun ° . ° i i os i | <8 e Perea ot ital vaua Accel i Modean Manus: Paper [90 Estbtsrens ce cm reo atee os Sonn T ° a 0078 | a I i a bo e Frc ial Vt Ad ‘ aoicn Maul Tle Eabngs) ce ura Fa 03 atee ° 182 Grom 9 Seo0 8 8 TE pucdlvaat witavane nsssl! lovee 38, TAP Grown Prosttes oa Mani Gaga BeANcHts OMEMCAN MANUFACTURING, 1984-1994) Wwe have seen and emphasized migh« simply be the result of ercors of observations? This js by no means a frivolous question. For one can actually create frequency distributions ‘of rates of TEP increase which contain ex- actly the same information as the sunrise sunset diagrams previously presented. The “only trick isto count as the unit of frequency ‘not one fem (out of an industry aggregate) ‘or one industry (out of some larger agate ate) but, instead, say, 1 percent ofthe total value added of the aggregate, Thus a firm with 20 percent of the value added of an in- dustry would appear with 10 times the ‘weight of a firm accounting for 2 percent of the value added of that industry. In such = chart, the cumulative frequency (say, 68 per- cent) above ATP = 0 would represent the projection on the horizontal axis of the ‘maximum point on a sunrise diagram. Its4 TH AMERICAN BCONOMIC REVIEW ance 1599 Macon Manutacing Spwing [27 Esta shen & cum Fao ae Siow —t g08 ° é | pass et Menican gufacing Woaing Apatal (20 Fstabsheors} os nee ° 008 o 3 Perce ot ital Vatu Adit Motcan Margtactring Sprteic Rosin (95 Estabahents qT 5 porate ot isan aaa Fuse SC. TRP Gowri Paoris Fon Stow -Growine BRantciss (MEXICAN MANUFACTURING, 1984-1994) complement (32 percent) would represent (he initial value added associated with neg- ative TFP performance dutiag the period, 1f, then, all ¢he information could be gen- ‘erated by 2 properly designed frequency dis tribution of rates of TEP growth, coukd it not all be the result of chance alone-—more spe- cifically, of errors of measurements? [ really think not—my favorite quip on this is that ‘white noise does not sing a tune.”” That is, if ‘we can rationalize what we see in terms of an analytical framework which embodies well- established economic principles and sensible presumptions about underlying relationships and facts, this is itself strong evidence against the white noise hypothesis. Nonetheless, we have to face the fact that errors of observation of some magnitude cer- ainly do exist, and we must recognize that they’can cloud our perceptions and bias our results, What Lam going to do here is consider frequency distributions of firms. TEP is mea-VoL. 88 NO. LHARBERGER: A VISION OP THE GROWTH PROCESS is Frequtrey * 4 rr i a TeP Growth Rate FRouns 6A. Averace ANMUAL TEP Ghar Reve: MExicaN MaNOACTURE 1984-1994 eequence DasrGunion, 44 moustHiaL BRANCHES sured in two ways—ane using value added by set of firms on the one hand, and che other using “output” by those same firms, mea sured through dividing value added by sep2- rate estimated firm-by-firm price (of value added) indexes p). For these purposes we can conveniently think in terms of logarithms, so. let y= observed value added of firms p, = estimated fimlevel price index; estimated output, true value added]; true price index]: and @=T, ~ x, [tcue output of firm) We would like to have data on g; and its variance 05 = 07 = lage + oF I we simply work with observed value added ‘sour quantity variable, we get 3 =o} + 03 (assuming ax. = 0). ‘F we worked with the measured y, we get s}soitoitoito}—tom (assuming i and é tobe stricly random). ‘My presumptions are as follows: (i) We can estimate value added quite ac- ccurately atthe firm level. Hence the pre- sumption that ¢2 is smal (ii) In most industries, there is considerable vatiety among the firms and their prod- ucts, Hence, except in cases of industries with very homogeneous products, we should not expect 03 to be small, Hence, Fexpect o2 > o3 (iii) Finally, we Rave the presumption that, a Jeast at the level of firms within an in- dustry, aur < 0. We know that firms choose to operate in regions of the de- mand carve where they consider theelas- ticity facing them to be greater than one. But also, in an analysis of the growth process, one would expect the big gains in value added to accrue to those firms ‘in an industry which were passing along to consumers some of the fruits of eur- rent or past real cost reductions. ‘These three presumptions lead me to the con- clusion that 07 is likely to understate the true variance of output 3 (because —2a;¢ > 0 and 3 > a2), and that o? is likely to overstate oF (only the covariance terms with é and 1, which were assumed to be zer0, could make it otherwise). And since Torre worked with real.6 [THE AMERICAN ECONOMIC REVIEW L Frequency MARCH 1998 i. 64s “ee ‘TER Growth Rate OVE OE, MAXIMUM A¥ERAcE AnNuAL TEP Gowri RARE (Max. Oaowate oP TRP Pape) Mexican MaNPACTIRING, 1984~ 1994 Frequency DIsTHBUNON 44 INDUSTAIAL WANCHES value added as his quantity variable, this sug- gests that, if anything, the substitution of the “teue quantity variable™ 4 for observed vale added v would have given cesults with greater dispersion of TEP, and consequently Ereater overshooting in his sunrise-sunset diagrams ‘The above demonstration should be taken as merely suggestive. I is not importagt to me that Torre's results underestimate the variabi ity ofthe different firms’ TEP experience. Itis only important that measurement error should not be the principle determinant of those re sults. On this feel very confident In my view, icreally is “a jungle out there,” with winners and losers in every period—good as Well as bad. ‘As Uhave noted earlier, we are only just e= sinning a systematic study of TFP among U.S. firms, so Tcan offer no disalay comparable to Torre's However, Rables (1997) did examine the experience of 12 fiems in the U.S. oil industry. His results are sunmacized in Figuce 7. But Robles (ells basically the same story as Torre. ‘Three firms out of the 12 were more than suf- ficient to generate the real cost reduetion ex- perience by the total stoup. Halt (or almost half) of the firms had negative TFP growth in ‘each period, And the cumulated amount of this negative TEP growth was sizeable when mea- sured against the (otal TFP performance ofthe industry ‘What [see in TFP performance is quite anal- ‘ogous fo what {sec in the stock market pages of te newspaper. There are winners and losers ‘every day, every month, and every year. The pains and losses come from all sors of causes. ‘World price shocks can drive firms into negative ‘TEP performance if the consequent output re- duetions are greater than the reductions of in- pats. So, too, can cyelica ar secular deebines in demand, including those caused by the suecess- Tul ations of competitors ‘When firms are under stress, they typically fight to stay alive. Maybe they fight for too Tong in some eases, in the sense that less of society's resources would be wasted if theyVoL. &8.NO. 42 10 Frequency 1 ats HARBERGER: 4 VISION OF THE GROWTH PROCESS 0 50. Ss Percentiles Flour 6. Penoaxraas work Pesrive TEP Gaon: Mexican MaNURRcrumino 1984— 1994 Frequency Distuaurton o° Peecaatues, 44 INpusTAaL Deans ‘were to quit earlier in response (o a challenge that turns out to be deadly. But they do. not recognize the challenge as deadly, so they keep strugeling to survive. [helieve this is part ‘of the nature of entrepreneurs, CEOs, and business leaders in general. They would not he where they ate, doing what they are doing, of they were ready to quit at the first sign of a challenge, They are fighters hy nature, and they probably wauld not have achieved sue- ‘cess if they were nat. wee ‘New challenges come Sean ferent firms think of diferent ways to respond to them. Some (like Intel and Mierosoft) end up winners; others (Montgomery Ward and ‘Apple?) end up losing. Butt may aot be that they just waited passively and tried ta fight 10 sunvve in the face of negative. Shocks. They may have had quite innovative ideas, with de- cent prioe probabilities of success, but in the fend success did not come. ‘Ths, negative TEP Petcare a et Se come simply from reed he is saying 1s, yes, it's a jungle out there, but the processes ofthat Jungle are at the core of the dynamics of market-oriented economy. They are what got Us la where we are, and they hold the best promise for further progress in the future.” In my opinion, Schumpeter sav through 10 the essence of the problem, but itis not wise for us ta be fatalistic in accepting his vision We cannot lose by making a major effort 10 Understand the process of TEP improvement where. it happens—at the level of the firm, This is all the more true because of the 2 The idea of creative desrucin ha cme up ecene lexan, a antes of formal deli dite for ‘his eapr's fous on gowth scecomcne ane the tise economic Smerpretation of st really (see Cane M. Growsmae and Elkanaa Helpran, 1991, 1994, Pailiope Aghion ans Pecer Hove, 1992) Feran economeanc tbl) ceuptaicing te vatibily of pecfomanee arang Cems see Tneques Maaesse aad Gliees (195 pp. 200-048 ‘THE AMERICAN ECONOMIC REVIEW 12 a § 4 3 2 t a mance 1998 50. 7 100 Percentiles Ficune 6D, Percents Wiexe Cum, Sist TAP Conressuricn = 100 Percen: Mexican Maneacruning 1994-199 pervasiveness of negative as well as postive TEP petformance among the components of almost any aggregate. By learning more about this aspect of the aggregate picture, we may stumble upon ways 0 “accentuate the postive and eliminate the negative”’ parts of the TFP story. But that is too quixotic a gos) to take a3 the point of focus right aow. To me, the pres eat task is simply to get hold ofthe huge mass of information that is available at the firm level and squeeze i¢ hard enough to wring out as much understanding and as muuch insight as IV. Some Observations an Methods ‘and Research ‘What | am about to say in this section is not ‘meant (0 consist of direct implications of what Jhas gone before. Instead, I think of the earlier parts of this paper as building a case for a cer- tain vision of the economy, and of how the: forces of growth work within it This vision in ‘aura Teads one to think in different ways not ‘nly about he grawth process itself but about how we, a5 economists, might hest advance ‘our study and understanding of it, and how policies might be molded so as better to pro- mote it. @ tis abways wise t0 study the compo- nents of growth separately. The cate of in- vestment, the rate of return on capital, the rate ‘of growth of the labor force in numbers or im hoirs worked, the contribution of hursan eap- ital or of the increment in average quality of Johor, and the residual representing real cost reduction—al these are sufficiently different, and potentially sufficiently disjoint from each other, to merit their being treated separately. T would give special emphasis to the following tere points Gi) The worthwhileness of measuring the tate of ectum and emphasizing its role in the growth process, Gi} The importance of focusing on invest. ment rather than saving in studying theVoL #8.N0.1 HARBEROER: A VISION OF THE GROWPH PROCESS Ww ‘Cum sum Foal Cost eduction cum, Rate of Growth one62 ‘cur sum Real Cost Reduction Percentile of initial Value Added A. 1970-1981 ozataa 5 Percentile el rita! Value Added 8 1982-194 Rouen. TEP Qeow Proms Fon US. Om Bins2 THE AMERICAN ECONOMIC REVIEW process of growth, Saving is an interest- ing (topic in its own right, but the more “open economy” is the situation being studied, the less saving has to do with investment. Saving takes on great im- portance in clased-economy models fo- cused on aggregate growth, in which case it is equal (0 investment. It gets to bbe almast meaningless as one focuses an the growth of cities and regions, or an firms and industries. (Gii) The importance of viewing the residual 5 an umbrella covering reat cost redue- tioas of all kinds, and of recognizing that ‘we ate closer to home thinking that RCR takes 1001 forms than that ican be well represented by one or two or three aggregate style variables, (b) In principle, the accumulation of ha ‘man capital by the tabor force shoutd he rep~ resented in the labor contribution of the growth equation, or in a bifurcation of this contribution ints one due to raw labor, the fothes to human capital. fis in a term like E,w.AL, that one captures the shifting skill ‘composition of the labor force. Jn particular, we capture here the higher wages that are the fruits of investment in education and training, which are the benefits that the workers them- selves perceive. These should be kept sepa rate from any externalities education might have It is important to try to keep this intemalized part ofthe story out ofthe residual so that we ‘ean straightforwardly interpcet the residual as real cost reduction, (©) To study externalities due to education, training, or human capital, we should not be ‘content with broad generalizations such a5 “TEP growth is higher in entities with lots of ‘human capital per worker."* We should try to figure out how this extemality works. Is it higher for firms with high incidence of human capital? I it higher for industries or sectors? ‘Or are human capital externalities more spatial in nature, making more efficient the economic hte of the cities, provinces, states, or nations which have high concenteations of human cap- ital? And if this is a fitful trail to pursue, at what type and size of geographical units do these extemalities typically work? MARCH 1998 (A) The same, goes for economies of scale We should not be satisfied with vague attri« Inutions of economies of scale, say, at the levet of the national economy Instead, we should pursue the matter. If the economies of scale are national, through what channels do they ‘work, and what evidence do we have (0 look atto see them in operation? In particular, what is their connection to real cost reductions ‘where they really happen—ie, at the level of the firm? Economies of scale at the levels of the firm and the industry are easier to visual- ize, Here, to, however, the tsk is to check them out—to See if the feal cost reductions of firms are linked to the initial sizes of chose firms themselves, of of the industries in which they operate, and of the direction (up or down} in which output is moving. Ce} Perhaps most important of all, we should ceally try to rake full advantage of ev- idence at the fom level. I think particularly of identifying considerable numbers of outstand- ing cases of TEP improvements and TEP de- cline, and studying them one by one (0 tty 19 ferret out the sources of their big real cast ree ductions and real cost increases, You can be pretty sure, if there have beea big seal cost reductions in a firm, some people in that firm have a pretty good idea of where those cedu tions came fram and how they were accom~ plished, By capturing this grassroots evidence, ‘we can put some added discipline into our ru- minations about the nature of TFP at the ag- aregate level. We must follow up on the sort ‘Of work pioneered by Jacob Schmookler (1966) and Edwin Mansfield (1995). In gen- eral, our aggregate story should be compatible with, and comfortably contain, what we see at the grassroots level. In particular our overall picture of TEP improvement should comfort- ably accept the overwhelming evidence of the “ mushrooms" rather than “yeast” nature of the peacess, (E) Special urgency applies to the study of declining (oral factor productivity at botk the fumand the industry levels. The pervasiveness ‘of declining TFP is perhaps the most profound, ‘conclusion fo emerge from the empirical links ‘hat [ have reported here. As a profession, we obviously have been aware of its existence at the industry level for viewally all studies thatVOL 88 NO.1 HARBERGER: A VISION OF THE GROWTH PROCESS ar sive a breakdown by industry reveal it, Yet fo and, (o a degree even on economic growth in my knowledge, we have barely scratched the general, I believe that the key words are ‘“oh- surface in studying it [find ithard to think of structing” and ‘‘enabling."” We know from ‘more fertile soil for future research on the pro- sad experience how easy itis for governments, cess of economic growth, (o adopt policies that get im the way af eco- omic growth and even turn it negative. We (g) 1 do not think that we gain much by know, 100, that there is no ‘silver bullet,"” no trying to express the relation between policies single magic key that by itself opens the door and economic growth by a series of regres- to a paradise of prosperity and growth. sions. Cross-country growth regressions seem Broadly speaking, the easiest stating point for hopelessly naive to Vongtime observers of the a successful growth experience is a once- growth process like myself To us, there isto prosperaus economy that has suffered fram ‘much to question in regression lines that draw bad policies. Releasing that economy from its, ‘uct of their slope from the differences be- tramnmels, correcting an accumulation of past tween Sudan and Switzerland, between Bang- mistakes, can sometimes set in mation a pro- Iadesh and Brazil, or between Ceylon and longed episode of astounding growth. A shift Canada. In contrast i¢ seems much more sen- from poticies that obstruct to policies that en- sible to took at episodes within individual able growth seems to lic atthe heart of growth. counties and to search for common elements “miracles like those of Taiwan, Spain, Ko- that characterize the passage from bad to good rea, Brazil, Indonesia, Malaysia, and China growth experiences within each of the number among others) Of countries, and for those clements that seem ‘The springboard for the following listing of to deseribe the good growth experiences on the policy implications is the interpretation of the fone hand and the bad ones on the other. [think growth residual as representing real cost re ‘we cam reach in this way a good appreciation duction and the ready acceptance that in the of the nature of the growth process, withont — eal world RCR comes in 1001 different resorting to the straitjacket of regression lines forms. that seem to draw from comparisons among, very disparate countries, lessons that are sup- posed to be meaningful for countries like Bangladesh, Ceylon, and the Sudan—as well Hence, policies that i accu as others at different levels—as each strives perception of real costs are ipso facto inimical to take the next upward step in the climb (o- to growth Infatin isthe mast cbvious, prob- ‘ward modernization, ably the most pervasive, and almost cercainly “My view of eross-country growth regres- the most noxious of suck policies. Hf Thave any sions is somewhat less negative ¢o the extent expertise based on experience in economics, it that they focus on the components of growth fas to be in the first-hand observation of pto- (rate of investment, rate of return, and real cesses of serious inf jou to. ftatrolucion ipercula) rather than on the take ny woe for the man eos Saat ‘Overall growth rate. There is also subtle dis- eerie near Uietian to be drawn between two ways of presenting cross-countty regressions~—(i) "The most serious cost Mplanng” shy and’ how some coonias (ot infatton le the blaring of econome sro fas than crs (rot eaomonenety, \egen peraptons of tlt aries Th Sto") as spl) simmarzang w sres of apron ces Indidual pice jut a lea ct acing he owes we cient ayn and st voy eifeent ste. cbedrve (ar pretcbieandmsusfar iste: igh prodot pie snd’ rnp cow or inert leat as ame) mulls an idaion clamoring fr ao i Tehtdena wb nade Tassos dees (7 Smee eptie ‘Show iat, whea sucha sgl canes Sy cut eho oda cone no Inappecachig the questo ofthe ntuence ha arth poeta apa ees eon ot pay oneal contrucion npartculae Grucan thet pate pa of sdf o2 ‘THE AMBRICAN ECONOMIC REVIEW the ongoing inflation. Without exception, in my own observations, the higher the rate of inflation, the worse is its effect in blurcing agents" perceptions of relative prices, In an inflation at, say. 20 percent to 50 percent per ‘year, people see prices as in a moming haze; in one of 20 percent to 50 percent per month, they see them ay in # London fog. Many pirical studies exist showing that serious in- flations are seriously inimical to growth. (See. William Easterly, 1996.) The clouding of perceptions of relative prices is an important reason way—for it gets in the way of sue- cessful real cost reductions at the evel of the individual firm, Infation also inhibits growth in other, per~ CB? raps more obvious ways @py aivering energies from more produc- tive activities to the search for mecha- nisms of inflation protection; bby reducing (often very drastically) peo- ple’s real monetary balances, thus im- pacting negatively on the real amount of ‘credit the banking system provides to the productive sector: and CGB somewhat related to bosh (a) and (b)- by causing people (both “here” and abroad) to invest abroad some of the funds they would otherwise have in- vested ‘‘here,”” or (what is very close to the same thing) by accumulating hhoards of hard currencies as an inflation hedge, (b) A second policy implication is, inthe words of my friend and longtime collabora- tor Emesto Fontaine, avoid ‘prices that lie” GEER RRTEEEEMAking shout safe. tion, we focus on the blurring of the signals that the price system gives; here we focus on its giving wrong signals due to distortions that have been introduced, usually asa direct consequence of government policies. No good can or did come, in terms of economic efficiency, from tariffs of $0 percent and 100 percent and more, giving effective protec tion eften of 200'perceat and 300 percent and more, Nor can growth be fastered by heavy-handed price controls and interven tions in cedit markets. Tam not being a cebigiows purist here—just as big distortions have big costs, small distor- MARCH 1998 tions typically have small costs, and all econ- ‘omies are distorted to some degree. The message here is thar economies have 19 pay the price for the level of distortions they ‘choose to have, and that one of te important componcats of that price is that distortions create situations where what is tray a saving ofpprivate casts isnota genuine saving of costs JSrom the point of view of the economy ax a whole (e) Just a8 bad, and often even worse than, ditect distortions, are the excess costs imposed ‘on an economy by ill-conceived regulations ‘and bureaucratic hurtles, Hernando DeSoto (1989) has made the exposure of these rram- mels in Peru into what has become virtually his life's worle Clear rules of toe game are fan essential and integcal part of a well~ functioaing market economy, but al too easily nese get supplemented hy others that make investment, production, marketing, sales, new product development, et., more cosy. Labor laws have been particulary troublesome, often adding artificially to the cost of labor and giv- ing firms 2 strong incentive to avoid hiring new workers, simply because of the high costs associated with any later dismissal of them ‘But there aré loeds of other items —the need {for approvals, sometimes a dozen of more, before undertaking some investment or some new venture; regulations that oné Way or an- other impede new entry, so as to protect stroug vested interests (small retailers being pro- (ected against supermarkets in many coun- tries); and the complexity of tax codes and their enforcement, which imposes large com Pliance costs on business firms and indi- viduals. Somehow, countcies interested in promoting growth should find ways of paring their regulatory frameworks dowm to. those rules and requirements that are really justif- able in texms of their costs and benefits 10 the economy and society at large. (2) Although international trade distor- tions (tantts, quotas, licenses, prohibitions, etc.) might be subsumed under points (b} and (c). their impostance merits a separate head- ing. The move to openness (from a protec tionism that sometimes bordered on auarchy ) hhas been one of the main frallmarks of the growth miracles of the past half-century [seevot. 88 NO. 1 Schastian Edwards (1993) and Anne ©. Krueger (1985, 1997)]. Just as inflation has costs beyond the area under the demand curve for real cash balances, so protectionism seems to have burdens that 30 beyond the standard triangle-trapezoid-rectangle measures of the costs of trade distortions. There are at least two quite natural explanations: frst, that openness helps grease the wheels of the intemational tcansfer of more modcra technologies, and second, that firms that may once have relaxed in ease and comfort behind high protective barriers end up having to sink or swim once forced (0 compete in a much more open- economy setting. Under either explanation, trade liberalization opens up new paths of real cast reduction, thus providing additional im- pts to economic growth, Gd re recent wave of privatizations among both developed and developing econ- ‘mies may have important effects in enabling real cost reductions that otherwise might have been delayed, or not have happcned at all. It is, I believe, fair (0 say that in most countries state-owned enterprises operate under a series, of constraints that seriously get in the way of teal cost minimization in a comparative-static sense and of real cost reduction in a dynamic sense, These constraints sometimes limit the salaties of executives, sometimes impose ‘onerous conditions an the firm. as it employs ower-skilled workers, often limit the capacity of the firm to shut down inefficient lines of production, and almost always make it diffi- Culeto fire workers, ete, To my mind, however, Dethaps the worst altribute of state-owned en- terprises is the cthos that often evolves inside of them—an ethos whete middle managers are well advised to “leave well enough alone,” “‘not rock the boat,” and “not invite trouble,” This ethos fies inthe face ofa vision of the growth process that gives a huge role to the search for teal cost reductions atthe grass- coots level, and that recognizes the tumult that accompanies “creative destruction” in all its forms. I thus must applaud the contemporary trend toward privatization. If [harbor any qualms in this conneetion, they concern the ddogree to which many privatizationshave been carried outin too much haste and with too litle care, often motivated by purely fiscal consid- rations rather than by @ general search, for HARDERGER: A VISION OF THE GROWTH PROCESS 28 economic efficiency. This may have led to gra tuitous transfers of wealth in some instances and to the planting of newly private enterprises in soil that was not properly prepared (e-g., sill lacking a sound cegulatory framework for electricity rates, or intelligent rules promoting competitiveness in at least some aspects of telecomrnunications, ex.) (£) One cannot complete a list like this without mentioning something that most of us simply take for granted —a sound legal and institutional framework in which indeviduals are protected against arbitrary incursions on their property and other economic rights. This very basic poiat—recently much emphasized by Douglass C. North (1996), Robert J Barrow and Xavier Sala-i Martin (1994), “Maancur Olson, Je. (1996) and Barro( 1997) — is atleast potentially a vital element fora sus- tained process of successful economic growth If itis tue that spurts of growth have some times occurred in the absence of such a frame work, it is also true that most cases of sustained growth over long periods of time hhave benefited from a sound institutional and legal environment (g) Somewhat related to the above is the element of political consensus concerning the broad outlines of economic policy. We have leamed from experience that very admirable policy reforms can take place, yet end up hav. ing little effect. This can happen because 4 ‘new government comes in and reverses the re form. But it can also happen hecause people fear that a new government will come in and reverse the reforms later on. At the moment, ‘the Chilean economy is one of the jewels of economic growth (and general economic stc- cess) in Latin America. Many people point to the thoroughness and pervasiveness of Chile's economic reforms over the lst two decades ot 0, But not so many point to the fact that the teform package has temiained essentially intact ‘through several changes of ministers, and even ‘more important, through two presidential elee- tions in which the winners came from the op- posite side of the political feace from the government that initiated the reforms. The ‘confidence in the economic order of things in- stilled by this sequential endorsement of the basic framework of economic policy has to be2 ‘THE AMERICAN ECONOMIC REVIEW ‘one of the important reasons for Chile's con- tinued, very impressive economic perfor- mance. And itis important, also, in the context of this paper. Living in a world in which real cost reductions are a key dynamic force pro- ducing economic growth, we must look to the motivations and preacenpations of those who take the critical decisions at the level of the firin, For these decisions, itis not only impor- tant thatthe policy flamework be good 20%; the expectation that it will stay good i the future is also important. Otherwise, invest- ments will tend to be limited 10 those with short horizons and payment periods, and much soil, fertile wi tonger erm economic oppor- tunities, will go unplowed. VL. A Vision of the Growth Process Let me now try fo summarize my own vi- sion of the growth process-—the majar ele- ‘ments of which have been presented here. In the first place we have the five standard pillars ‘Commenting om these in tur, I would note that increases in the labor force kave taken on new meaning i many countries as labor force participation tates (particularly those of women) have increased. Whereas with a con- stant participation rate, the growth rate of the labor force is just a proxy for the growth cate ‘of popvlation, important inereases in. labor Force participation can lead, just by them: selves, to significant increases in measured real income per head. ‘Concerning increases in the stock of human capital, my conviction is that most of their ccontsibution to growth is, on the whole, well measured by market wages, as in the expres- sion E,w,AF,. ‘This does not deny the exis- tence of exicenalities due to au increased ‘human capital stock; i staply judges their in- fluence on the growth rate to be nuodest in comparison with the effects of education, training, learaing-by-doing, ete. that can be (and typically are) internalized by those who MARCH 1998 receive them. We therefore look forthe effects ‘of buman capital accumulation maialy in the term E,4,AZ,, and only (via externalities } as ‘one of many'lements underlying the growth residual RY. ‘The rate of investment is a veteran on the stage of growth analysis. What } would em- phasize here isthe importance of maintaining a clear separation between the rate of invest” ment and the rate of saving Modety-ctike “Grove of Te representative consumer) in which saving and iavestment are always equal are nat much use even for analysis at the national level in our modern, interde- pendent world. They are even less useful as fone goes down to smaller geographical regions, and simply cease to make sense as ‘one studies the growth process at the levels of the industry and the frm ‘The rate of return ¢o investment has in many ways beea the orphan of our gcowth analysis, having been masked from view by ‘out typical representation of capital's couti- bution to the growth rate as 5(AK/K). Here te rate of retumn (p + 6) is totally hidden from view. I deeply urge that more of us get into the habit of representing this same term as (p + 5)(AK/y). 1 want to see more alten- tion paid to the rate of return because it plays such a central role in the motivation of eco- nomic agents, and also because changes in it are such an important element in understand: ing and explaining the growth cesidual, R’ ‘Fable 3 helps explain why I feel this way. This table is adapted from Harald Beyer's (1996) work. He carried out an anslysis of the growth experiences of 32 couatries rang ing from Sri Lanka to the United States om ‘he ineome scale, and front Ieeland fo Austea- lia on the geographic seale. In Table 3 we ptescat results for his ten counteies with the highest and for his ten couréries withthe low- est GDP growth rates from 1971-1994. Inthe second ealumia the calculated avecage anaval rate of retuen is shown. In the third coluran We bave capital's conteibution ta the gcowth (p+ 8)(K!9)], and inthe nal cal fuman the estimated average annual rate of TFP ‘growth, all over the same time per ‘Table 3 shows an unequivoca) teadency for fast-prowing countries to be experiencing high rates of relum as welt as high capital contn- buvons and high rates of TEP improvement.VoL #8 NO. 1 LHARBERGER: A VISION OF THE GROWEH PROCESS 2s ‘Trove 3—Growris Raves, Res OF RETURN, Nb RATES O° TP INFROVEMENT {SHEACTED WOM A SAMOS 22-CCAERES, 1972 ( Ten eet ‘cop owing counries growth Taiwan am Korea a Toss 148 ong Keng 791 ceador ss opis sn ibabme 40 CCalama an lesan 498 tele 40 Mein 3s 0 oor gout 2 am 2 256 29 a2 South Aiea 26 Dena bis Used Kaptan aa Sweden 1.8 Metin as Mean Dai 291) Rate of) Capieal cont. TEP \ rium) warosthete (gosh ate 150 381 365 a2 430 238 as 348 296 0 336 228 140 270 036. a6 299 in 16 2a on us 199 on 9a 195 un 67 170 036 na8s 284 ne 29 Rate of Capital cont 9 growth ete st 1D 6s Le 63 07 6a 1.06 10 La ass on Lao 23 15 Le -a9 1 Lo as 96 995 an 43 as 2a 600 Loe ons as 08 oe Source: Beyer (1990), Tables TL. theough 11.3% also Appendtx for cates of reture is all the more interesting because in the calculation of TFP a higher level of the rate of return operates to reduce the calculated TRP Gc, Ap is a positive component of R" and should presuriably be positively correlated with it but R” is found by subtracting e AK from Ay; hence, in a sense, the level of @ The eandard expression fc the residual, R > fy ~ wae (po B)dK sas "dual eeplesenation whichis: should presumably be negatively correlated with R"}, What we are seeing here, in my opinion, is a genuine syndrome. in which all sorts of good things go together. Strong real R= Da + Haig + 8) — Bap. Ths form srply says thatthe fees of eal cost ceducton have to go some heteelter to werkews (Late) aft eomers of espa [Hao +8) er, i the form of lower pics tothe ay Ups cuwrmess7 ‘THE AMERICAN ECONOMIC REVIEW cost reductions and high rates of return create attractive investment opportunities which, sehen acted upon, bring about a high capital contribution 0 growth It sould be no surprise that under such circumstances the GDP growth, tate itself tends to be high. Tt should Tikewise bbe no surprise that the opposite syndrome — with weak real cost reductions and low rates of return producing fewer interesting invest ‘ment opportunities shoul’ end up being as- sociated with a low capital contribution and a low GDP growth rate. Finally. we come to the residual ! itself To me, the biggest message here is to recog- ize the muliplicity of sources from which it can (and [ believe does) come, and the addi tivity that nevertheless is its attribute, I think that the term real cost reduction very neatly captures both these aspects in a way that ren- ders it preferable to terms like TEE improve ment and technical advance— preferable. not in the sense of a mechanical definition (for which all three are. equally good), but in the sense of better conve¥ing the undedying na ture of the process to one’ listeners or readers. ‘The next step isto recog ve iain pillas, at least three ( ey eg eet ‘cape the conclusion that the great bulk of the actio ‘with the growth process takes place at the Tevel fence, I “feet we Should focus wich mor focus much more. study than wwe have in the past on what happens at this level. And when we are not working atthe fm, level, we should pay a lot of attention to what hhappens af lesser levels of disaggregation like industries ard industrial branches. Key insights flow from taking this kind of, focus. Few economists are aware of the per- vasiveness with which suprise-sunset dia- sarams are characterized by overshooting, or of the importance that firms or industries with real cost increases (ie,, reductions in TEP) play in determining the aggregate rates of real cost reduction that we see in such diagrams. Here we have only seratched the surface in digesting the evidence. But | find impressive the degree co which the data of Table 3 seem 1 point t a growth syndrome in which high rates of return, high rates of investment, high rates of real cost reduction, and high rates of MARCH 1998 output growth al go together. ¥ see in this re- sult the likelibood that real cost reductions are the big driving force, generating bigh rates of retura and calling forth high rates of invest reat and ftigh output growth. This intecpee- tation is compatible with many exercises that Thave performed over the years in which [ have tried to contrast high-growth with low- syowth experiences. In such exercises, as in Table 3, the difference in rates of real cast re- duction has (ypically been a major “Source™ of the difference in growth rates. Also impressive in the agalysis of the Jorgenson data isthe degree to which the vary- ing experiences of U.S. manufacturing in dif- ferent decades desives from different degrees of bad experience (real cast increases) rather than different degrees of good experience (real ‘cost reduction). Itis as if the creative part of Schumpeter's “ereative destruction" was mote steady (for these decades in U.S. man- facturing) than the destructive pact, whetting cur (or at least my) appetite to look deeper, inguicing into way this was 50, ‘The Mexican data at the firm level were somewhat more recalcitrant than the Jorgenson data by industrial branch, but they nonetheless sive us a clear picture of lots of winness and Tots of losers, with the losers being strongly characterized by falling real value added and/ or by falling real rates of reuum.* +0 Tome's obsevations with eegetive TEP yeowth, ‘over theee fourths nad negative growth in real value Seed, an over one-half ac falling eal rates of zien {ss than perce showed inreases both in el vale sed and ig tr real cae of rou. Many of Tome’s ‘anomalies of negative TEP improvement ogee wi postive teal utpt grou ser fore very hgh tates of {eure (2 being ented to the observed increases CAR) mike capt tok, "This puns toa puobler that extends to al (or nearly 1) srowiaccotabng Eamewcrks. Implicily, hey a fign to rew icsestment 2 warginal product bated oh the average oberved gross rte of rouen, (@ + 4) 0 average Soserved capital share 4. This makes Ile sense in eases ‘where the observed pis fr hove op far belo the going marke rates, Frm eang 50 percent real resent thei? Histories Investrens are vey Unliely to be "seculang” sich a igh expected eld on new investerts, Say, Tims gome through periods of teal accouminy loses tay often al be vet, but te er ik they ie expecting (or typically getting) negative reuns on nei new investments (AR), The are Think, gd re one for uo experiment st erative ways a elVOL 58NO. 1 rationalizing and/or eliminating barriers and controls, they may also lead to an increased pace of investment and increased rates of re tum. In this view, the connection between 00d policy and growth is not mectnical, and thus cannot easily be captured in regeession- type analysis, but that does mot stop it from. being of vital importance. want to give special weight to another cole of “growth-enabling”” policy actions, which hhas (© do with how policies, the effects of Which might typically be considered to be comparative static, can nonetheless turn out 0 influence economic growth rates over ex- tended periods of time. ‘This story begins with a cecognition that ‘vost developing counties have typically used production techniques that were. “backward” in elation to those used by the advanced econ- comies. One way to verity this is to imagine integrally replicating, say.a U.S, factory in In- dia, and manning it there with Indian workers equivalent in skill to their U.S. counterparts. The combination of tower construction costs, and much lower operating costs (mainly ‘wages) would permit this hypothetical aew In- dian firm ( undercut the prices of both the U.S. firm of which i¢ was a eopy and the typ- ical Indian firm curently active in the same industry. This says that the typical Indian firm. is operating on an "inferior" production func- tion, If, as [ believe, the difference in effi ciency between U.S. and developing-country firms is typically large, there is much room for quite rapid improvements in the developing countries as they leam how to “adopt and leg the ae of eum (2) imputed wo new investment ‘Atgucients can be develope for using abitary Bit "ser ‘ble rates (like 10, 15. <1 20 percent), a fr applying te iavesumects hy individual ms ardor indus tse ofetum 2 that ae obbiced nol fe Use specie as fr indusires bu eam brad scar eg sauactrig) {rezonomy wide averages. Tam paring ese avenues in ‘ork that will be reported alter tine HARBERGER: A VISION OF THE GROWTH PROCESS 7 adapt" already-known techoiques from the advanced countries. J would assume thatthe incentives for ‘con vergence” are always present, but that they have typically run into barricrs and trammels of many kinds in the poor countries of the world, Reducing the barriers and loosening the trammels permits the more rapid convergence ta techniques that are closer to the frontier of knowledge, The way I sce the influence of policy in ‘growth, itis simply not true tha implementing enabling policies typically permits a quantam jump from the old to the truly modem. It is imate accurate to describe it as speeding up ‘what will in any case be a very lengthy pro- cess. Personally, I like the analogy to 2 hy- draulic system in which a large vessel with a high water level and lots of water is connected toa much smaller and narrower vessel with a tmuch lower level of water, Physical laws dic- tate tendency for the water levels o equalize im the end. Bat this can take a very long time if the tube connecting the two vessels is tiny, ‘or is clogged by extraneous matter. The policies that we consider good for growth have the attribute, in this analogy, of removing the extraneous matter and /orenlarg- ing the connecting tube. Buteven with the best modernizing policies, the tube remains small enotigh so that it takes many decades for a Country to pass from poor to middle income oF from middle income to rich. If somehow the hydraulic connection could be made so large as to bring about an almost instantaneous full adjustment of the water level, then we would say that good policies mainly represent level adjustments. But observing even the best of real-world growth experiences, J think we hhave to conclude that the adjustment is going (o be extended over a lengthy period in any event, thus causing the big observable result of beiter policies to be a higher growth rate ‘over an extended petiod rather than. a discrete jump to a totally different level. This way of looking at the world also leads to some observations on the current literature dealing with convergence. I have long been mystified by allusions to an ultimate conver- ‘gence of growth rates among countries, oran ultimate convergence of levels of output pet head, To me, the natural convergence is prod- et by product, not country by counzy. And8 ‘THE AMERICAN ECONOMIC REVIEW among products there may be some where current techniques will never be further improved. Those products will have no real cost reductions or TEP improvesments in the futlre, while others will enjoy huge advances of productivity, My guess is that unlucky countries (Bbutan, Nepal, Mongolia”) may always lag way behind the pack, while luckier ‘ones (Taiwan, Argentina, Brazil?) may one day hope to be among the world’s leaders. So convergence comes through as a general ten- dency, and quite sucely a general possibility, for the production techniques used in making any given product to improve as enterprises using “backward” techniques leam of better ‘ones, and even more important, learn about hhow to put them into practice. Ido not believe that much more than this can be said of con- vergence as areal-world force, Wage rates for given types of labor tend toward a cough ‘equality across regions in a country because ‘of the ease of migration in response to per~ ceived wage differences. The forces at work internationally are both weaker and more ‘complex, but the big message here is that he improvement of technique in any one in- dustry does nothing to improve the wages of Tabor in just that industry. After an improve ment, the industry may end up hiring more or Jess labor, but will presumably choose the amount of labor 50 as to bring marginal pro- ductivity into correspondence with market ‘wages. So technological improverment has an effect on wages via supply and demand in the rational labor market, not heaugh any direct link from technical improvement to wages ‘for given skills, ete.) at the level of the in- dustey or the firm. As my final point, let me return to the ‘thought thatthe justification for perfecting the functioning of the market system does not lie only in reducing the efficiency costs associated with each period’s operation of the economy. Perfecting equntry’s economic policy docs rot only cause it to move from 4 path at around, say, 90 percent of its potential output to another equal to 95 percent of potential, with the time path of potential output being somehow given in advance. That gain would certainly he a worthwhile gain, and it would amply justify a lot of hard work involved in achieving it, But that gain is still fundamen- tally comparative static in nature. MARCH 1998 ‘What I hope to have evoked ia this paper is a sense that the perfecting of economic pro- ‘cesses can also in nearly all cases be justified 15 greasing, the wheels of the. constant search for new avenues of real cost reduction, To the ‘extent that economic reforms do 50, they be- ‘come vehicles for bringing an economy to a point where, year after year, new, cheaper, and better ways are found of doing things, not just in so-called “production” but also’ in such mundane areas as merchandising, sales, - nance, insurance, and many more ‘Some years ago, in a book that {edited called World Economic Grovth (1984), I wrote an essay called “Economic Policy and Economic. Growth,” ia which T listed 13 lessons" that thought followed from the papers presented in the volume, recounting the growth experi- ences of countries as disparate a5 Ghana and ‘Taiwan, or Japan and Sweden. These lessons— basically focused on thinking about policies in terms of their economic costs and benefits— could easily be read as 2 reprise of the old comparative-static story. But they were not meant as such—it is quite relevant that they appeared in an essay concerned with World eco ‘nomic growth. The point was that these sensible policies emerged as part of a consensus of se~ rows economists, each an expert in his partic~ lar country’s history, focusing attention on the process of economic growth. ‘A few years later, and as the theme of a different concept, John. Williamson (1990) coined the term, the “Washington consen- sus.” Williamson listed ten poins, covering territory very similar to my 13 lessons. He also produced a pithy summary that captures the essential thrust of both his and my listings “Macroeconomic prudence, outward orienta tion, and domestic liheralization.”” He, too, and the members of the Washington profes- sional establishment whose. apparent consen- sus led to Williamson's list, were not just thinking of comparstive-static gains as they reached theie conclusions about policy. Teey ‘were thinking about ways to move economies from slow growth, stagnation, and even in some cases negative growth, to a healthy, prosperous flowering of economic progress. Similar views were more recently affirmed by Stanley Fischer (1993), To me, the dynamics of real cast reduction ase atthe very least an important piece of whatvot. #8 40.1 people have in mind witen they is efficiency oriented policies as essential ingredients of a program promoting economic growth. It is policies of this type that give the right sigaals {o the CEOs and the managers down the fine, that take away trammels that impede their quest for ceal cost reductions, and that create ait environment in which Schumpeter’s pto- cess of “creative destruction’’ can work its wonders APPENDIX ON METHODOLOGY The vision of the growth process presented in this paper leads one almost inevitably co some methodological twists—-twists which, if they are not new, at Teast differ in significant ways from what I take to be the most common practice in breaking economic growth down info components. (2) To measure the real rate of return to capital, one must express the numerator (real dollars’ of retum) and the denominator (the

You might also like

• The Subtle Art of Not Giving a F*ck: A Counterintuitive Approach to Living a Good Life

  

  From Everand

  The Subtle Art of Not Giving a F*ck: A Counterintuitive Approach to Living a Good Life

  Mark Manson

  Rating: 4 out of 5 stars

  4/5 (5819)
• The Gifts of Imperfection: Let Go of Who You Think You're Supposed to Be and Embrace Who You Are

  

  From Everand

  The Gifts of Imperfection: Let Go of Who You Think You're Supposed to Be and Embrace Who You Are

  Brene Brown

  Rating: 4 out of 5 stars

  4/5 (1093)
• Never Split the Difference: Negotiating As If Your Life Depended On It

  

  From Everand

  Never Split the Difference: Negotiating As If Your Life Depended On It

  Chris Voss

  Rating: 4.5 out of 5 stars

  4.5/5 (845)
• • Magazines
  • Podcasts
  • Sheet music
• Principles: Life and Work

  

  From Everand

  Principles: Life and Work

  Ray Dalio

  Rating: 4 out of 5 stars

  4/5 (609)
• The Glass Castle: A Memoir

  

  From Everand

  The Glass Castle: A Memoir

  Jeannette Walls

  Rating: 4.5 out of 5 stars

  4.5/5 (1717)
• Grit: The Power of Passion and Perseverance

  

  From Everand

  Grit: The Power of Passion and Perseverance

  Angela Duckworth

  Rating: 4 out of 5 stars

  4/5 (590)
• Sing, Unburied, Sing: A Novel

  

  From Everand

  Sing, Unburied, Sing: A Novel

  Jesmyn Ward

  Rating: 4 out of 5 stars

  4/5 (1104)
• Hidden Figures: The American Dream and the Untold Story of the Black Women Mathematicians Who Helped Win the Space Race

  

  From Everand

  Hidden Figures: The American Dream and the Untold Story of the Black Women Mathematicians Who Helped Win the Space Race

  Margot Lee Shetterly

  Rating: 4 out of 5 stars

  4/5 (897)
• Shoe Dog: A Memoir by the Creator of Nike

  

  From Everand

  Shoe Dog: A Memoir by the Creator of Nike

  Phil Knight

  Rating: 4.5 out of 5 stars

  4.5/5 (540)
• The Perks of Being a Wallflower

  

  From Everand

  The Perks of Being a Wallflower

  Stephen Chbosky

  Rating: 4.5 out of 5 stars

  4.5/5 (2104)
• The Hard Thing About Hard Things: Building a Business When There Are No Easy Answers

  

  From Everand

  The Hard Thing About Hard Things: Building a Business When There Are No Easy Answers

  Ben Horowitz

  Rating: 4.5 out of 5 stars

  4.5/5 (348)
• Bad Feminist: Essays

  

  From Everand

  Bad Feminist: Essays

  Roxane Gay

  Rating: 4 out of 5 stars

  4/5 (1018)
• Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future

  

  From Everand

  Elon Musk: Tesla, SpaceX, and the Quest for a Fantastic Future

  Ashlee Vance

  Rating: 4.5 out of 5 stars

  4.5/5 (474)
• Her Body and Other Parties: Stories

  

  From Everand

  Her Body and Other Parties: Stories

  Carmen Maria Machado

  Rating: 4 out of 5 stars

  4/5 (822)
• The Outsider: A Novel

  

  From Everand

  The Outsider: A Novel

  Stephen King

  Rating: 4 out of 5 stars

  4/5 (1866)
• The Emperor of All Maladies: A Biography of Cancer

  

  From Everand

  The Emperor of All Maladies: A Biography of Cancer

  Siddhartha Mukherjee

  Rating: 4.5 out of 5 stars

  4.5/5 (271)
• The Sympathizer: A Novel (Pulitzer Prize for Fiction)

  

  From Everand

  The Sympathizer: A Novel (Pulitzer Prize for Fiction)

  Viet Thanh Nguyen

  Rating: 4.5 out of 5 stars

  4.5/5 (122)
• Angela's Ashes: A Memoir

  

  From Everand

  Angela's Ashes: A Memoir

  Frank McCourt

  Rating: 4.5 out of 5 stars

  4.5/5 (441)
• Brooklyn: A Novel

  

  From Everand

  Brooklyn: A Novel

  Colm Tóibín

  Rating: 3.5 out of 5 stars

  3.5/5 (1947)
• The Little Book of Hygge: Danish Secrets to Happy Living

  

  From Everand

  The Little Book of Hygge: Danish Secrets to Happy Living

  Meik Wiking

  Rating: 3.5 out of 5 stars

  3.5/5 (401)
• A Man Called Ove: A Novel

  

  From Everand

  A Man Called Ove: A Novel

  Fredrik Backman

  Rating: 4.5 out of 5 stars

  4.5/5 (4771)
• Steve Jobs

  

  From Everand

  Steve Jobs

  Walter Isaacson

  Rating: 4.5 out of 5 stars

  4.5/5 (808)
• The World Is Flat 3.0: A Brief History of the Twenty-first Century

  

  From Everand

  The World Is Flat 3.0: A Brief History of the Twenty-first Century

  Thomas L. Friedman

  Rating: 3.5 out of 5 stars

  3.5/5 (2259)
• The Yellow House: A Memoir (2019 National Book Award Winner)

  

  From Everand

  The Yellow House: A Memoir (2019 National Book Award Winner)

  Sarah M. Broom

  Rating: 4 out of 5 stars

  4/5 (98)
• Devil in the Grove: Thurgood Marshall, the Groveland Boys, and the Dawn of a New America

  

  From Everand

  Devil in the Grove: Thurgood Marshall, the Groveland Boys, and the Dawn of a New America

  Gilbert King

  Rating: 4.5 out of 5 stars

  4.5/5 (266)
• The Art of Racing in the Rain: A Novel

  

  From Everand

  The Art of Racing in the Rain: A Novel

  Garth Stein

  Rating: 4 out of 5 stars

  4/5 (4208)
• A Tree Grows in Brooklyn

  

  From Everand

  A Tree Grows in Brooklyn

  Betty Smith

  Rating: 4.5 out of 5 stars

  4.5/5 (1929)
• Team of Rivals: The Political Genius of Abraham Lincoln

  

  From Everand

  Team of Rivals: The Political Genius of Abraham Lincoln

  Doris Kearns Goodwin

  Rating: 4.5 out of 5 stars

  4.5/5 (234)
• Yes Please

  

  From Everand

  Yes Please

  Amy Poehler

  Rating: 4 out of 5 stars

  4/5 (1902)
• A Heartbreaking Work Of Staggering Genius: A Memoir Based on a True Story

  

  From Everand

  A Heartbreaking Work Of Staggering Genius: A Memoir Based on a True Story

  Dave Eggers

  Rating: 3.5 out of 5 stars

  3.5/5 (231)
• The Woman in Cabin 10

  

  From Everand

  The Woman in Cabin 10

  Ruth Ware

  Rating: 3.5 out of 5 stars

  3.5/5 (2522)
• Fear: Trump in the White House

  

  From Everand

  Fear: Trump in the White House

  Bob Woodward

  Rating: 3.5 out of 5 stars

  3.5/5 (738)
• Wolf Hall: A Novel

  

  From Everand

  Wolf Hall: A Novel

  Hilary Mantel

  Rating: 4 out of 5 stars

  4/5 (3811)
• John Adams

  

  From Everand

  John Adams

  David McCullough

  Rating: 4.5 out of 5 stars

  4.5/5 (2409)
• On Fire: The (Burning) Case for a Green New Deal

  

  From Everand

  On Fire: The (Burning) Case for a Green New Deal

  Naomi Klein

  Rating: 4 out of 5 stars

  4/5 (74)
• The Light Between Oceans: A Novel

  

  From Everand

  The Light Between Oceans: A Novel

  M.L. Stedman

  Rating: 4.5 out of 5 stars

  4.5/5 (789)
• Manhattan Beach: A Novel

  

  From Everand

  Manhattan Beach: A Novel

  Jennifer Egan

  Rating: 3.5 out of 5 stars

  3.5/5 (792)
• The Constant Gardener: A Novel

  

  From Everand

  The Constant Gardener: A Novel

  John le Carré

  Rating: 3.5 out of 5 stars

  3.5/5 (104)
• The Unwinding: An Inner History of the New America

  

  From Everand

  The Unwinding: An Inner History of the New America

  George Packer

  Rating: 4 out of 5 stars

  4/5 (45)
• Rise of ISIS: A Threat We Can't Ignore

  

  From Everand

  Rise of ISIS: A Threat We Can't Ignore

  Jay Sekulow

  Rating: 3.5 out of 5 stars

  3.5/5 (137)
• Little Women

  

  From Everand

  Little Women

  Louisa May Alcott

  Rating: 4 out of 5 stars

  4/5 (104)
• OvercapacityReport2016_R3

  

  Document62 pages

  OvercapacityReport2016_R3

  Frann Ara

  No ratings yet
• Frederick GVC Mapping

  

  Document25 pages

  Frederick GVC Mapping

  Frann Ara

  No ratings yet
• Azmeh - Nadvi - Asian Firms Restructuring Global Value Chains - 2014

  

  Document11 pages

  Azmeh - Nadvi - Asian Firms Restructuring Global Value Chains - 2014

  Frann Ara

  No ratings yet
• Key Value Chain Steel Euromines Final 0

  

  Document23 pages

  Key Value Chain Steel Euromines Final 0

  Frann Ara

  No ratings yet
• 1 s2.0 S109095162100033X Main

  

  Document10 pages

  1 s2.0 S109095162100033X Main

  Frann Ara

  No ratings yet
• When Do Value Chains Go Global

  

  Document21 pages

  When Do Value Chains Go Global

  Frann Ara

  No ratings yet
• Jep 23 1 221

  

  Document19 pages

  Jep 23 1 221

  Frann Ara

  No ratings yet
• Johnson 2018 Measuring Global Value Chains

  

  Document32 pages

  Johnson 2018 Measuring Global Value Chains

  Frann Ara

  No ratings yet
• Dijk GlobalValueChains 2012

  

  Document25 pages

  Dijk GlobalValueChains 2012

  Frann Ara

  No ratings yet
• Values As Timeless Foundations of Todays Economy

  

  Document19 pages

  Values As Timeless Foundations of Todays Economy

  Frann Ara

  No ratings yet
• #1162 - 292 308 - Schneider

  

  Document17 pages

  #1162 - 292 308 - Schneider

  Frann Ara

  No ratings yet
• Evolutionary Theorizing in Economics: Richard R. Nelson and Sidney G. Winter

  

  Document50 pages

  Evolutionary Theorizing in Economics: Richard R. Nelson and Sidney G. Winter

  Frann Ara

  No ratings yet
• Evolution of Social Behavior: Individual and Group Selection

  

  Document30 pages

  Evolution of Social Behavior: Individual and Group Selection

  Frann Ara

  No ratings yet
• Jep 15 4 69

  

  Document81 pages

  Jep 15 4 69

  Frann Ara

  No ratings yet
• Evolution and Human Nature: Arthur J. Robson

  

  Document21 pages

  Evolution and Human Nature: Arthur J. Robson

  Frann Ara

  No ratings yet
• Jep 15 4 3

  

  Document8 pages

  Jep 15 4 3

  Frann Ara

  No ratings yet
• Jep 15 4 87

  

  Document26 pages

  Jep 15 4 87

  Frann Ara

  No ratings yet