Neoclassical Model Four Uncaged Tigers:

Neoclassical Model

Four Uncaged Tigers: The Eastern Asian Growth Phenomenon Revisited: A Neoclassical Interpretation

Beginning in the 1960s, the four nations of Hong Kong, Singapore, South Korea, and Taiwan became the first newly industrialized countries. High growth rates and a rapid rate of industrialization that extended through the 1990s helped pushed these nations from their position in the developing world into a position of being high-income economies. As we enter the second decade of the twenty-first century, these nations are still among the world's fastest growing industrialized economies globally. This paper examines the economic growth in these countries by considering different models for economic growth in nations that are making the transition from the Third to the First (or nearly First) World, focusing on the relative importance of increased capitalization and increased productivity along with other related factors such as the cost of labor and the level of education in the labor force.

Before proceeding, it is worth noting here that the term "newly industrialized country" is a term that has a specific economic definition. NICs are countries that in economic terms can not yet be counted among the First World nations but that are demonstrate along a number of economic indices that they are ahead of where they might be expected to be in comparison to other developing nations. As is generally true of NICs, these "Four Tigers" underwent rapid and significant economic growth that was in a significant way export oriented that paralleled (in time) rapid and substantial industrialization and urbanization. These four countries (as again is typical of NICs) also opened their markets. There are other aspects that generally obtain in NICs that I will not be analyzing directly here but that should be acknowledged as being in the background, such as strong political leadership.

"An Equally Remarkable Record of Factor Accumulation"

It should not be surprising that there are a number of different models and explanations for the economic growth in the four countries that are the focus of this paper. One reason for this is that economic growth and development are complex, which tends to encourage a multiplicity of models. The second reason why there are a number of different models has nothing to do with the facts on the ground but rather the philosophical assumptions and preferences of the economist. While economics is, of course, a social science that is concerned with describing the realities of the world, it is also a discipline that has embedded in its models certain prescriptive assumptions about how the world should be run. This fact -- that economics is always prescriptive as well as descriptive (in additional to being analytical) -- is something to bear in mind as we examine economic models of growth in these four nations.

The first model of growth in this area that I will examine is the one put forth by Young (1995), a model summarized in what he refers to as an intentionally dull paper. (This is, in fact, not true.) His overall assessment of the dynamics of growth in these nations follows:

It is this record of growth, along with its apparent association with the rapid growth of manufactured exports, that has led most economists to believe that productivity growth in these economies must be extraordinarily high, particularly in their manufacturing sectors. This view, however, ignores an equally remarkable record of factor accumulation (Young, 1995, p. 242).

Factor accumulation includes all those elements that go into production, including human capital, physical capital, knowledge. Young details the types of factor accumulation that have occurred in these four countries as the basis for his argument that the economic growth and development in these nations has been predicated on factor accumulation more than any other reason.

Young lists the important elements of factor accumulation in these nations as follows:

Labor input, primarily as a result of the increasing level of female participation in the workforce (which is in turn the result of a lowered birthrate)

Transfer of labor from one economic sector to another, that is from agricultural to industrial economic activity

Capital input has grown rapidly in the for countries: "although the investment to GDP ratio has remained roughly constant in Hong Kong, in the other NICs it has risen substantially over time."

Rapid human capital accumulation, which the percentage of workers with a secondary education or more in each country almost tripling. Young notes -- and this is an essential transition to my discussion of his overall assessment of growth in these nations -- that this increase in human capital in terms of increased education can be translated into a contribution to about 1% per annum additional growth in labor input in each of these economies" (Young, 1995, pp. 644-45).

So what does this mean? Are there clear conclusions to be drawn from this? Young argues that there is in fact only one rational answers, which is that the most accurate and parsimonious explanation of the rate and direction of growth in these nations is one based on factor accumulation. He is arguing, in effect, that economists have in some measure let themselves become bedazzled by the economic growth in these four nations and, as a result, have failed to sufficiently rigorous in their assessment of the underlying dynamics. In reading Young's research (in the paper cited here as well in his other work in this area), I am almost tempted to believe that his critique of other models of explanation for the economic growth in this region borders on thinking that other models are informed more by hope than facts, or perhaps even by something less rational than hope.

Demystifying Growth Rates

A purely rational approach to understanding the growth dynamics in these nations, Young, argues, is tantamount to demystifying the rate of growth as emanating from increased productivity.

All of the influences noted above-rising participation rates, intersectoral transfers of labor, improving levels of education, and expanding investment rates-serve to chip away at the productivity performance of the East Asian NICs, drawing them from the top of Mount Olympus down to the plains of Thessaly. In a companion paper [Young 1994], I use simple back-of-the-envelope calculations and large international data sets to show that, with regard to productivity growth in the aggregate economy and in manufacturing in particular, the NICs cannot be considered to be strong outliers in the postwar world economy (Young, 1995, p. 645).

Young's overall finding is that there is simply no good reason for assuming that there is any significant increased level of productivity: He notes that with the exception of Singapore, these NICs's productivity growth is "not particularly low" but also "not extraordinarily high" (Young, 1995, p. 671).

In summarizing his conclusion, Young forcefully makes a point that I alluded to earlier: Much of what goes into economic assessments is ideologically driven rather than factually based:

Underlying the pervasive influence of the East Asian NICs on both theoretical and policy-oriented research in the economics profession lies a common premise: that productivity growth in these economies, particularly in their manufacturing sectors, has been extraordinarily high (Young, 1995, p. 671).

Young's argument (and the data behind it) is an extremely -- with one important caveat (to which I will return). His work, too, has its own ideological core.

Collins and Bosworth (1996) in general support the findings of Young (1995), for they too attribute economic growth in this region of the world to factor accumulation far more than to increased productivity. In addition to supporting (and providing additional clarification) of Young's findings and suppositions. In addition, they provide an excellent assessment of the usefulness of growth accounting for this type of analysis. They write:

Growth accounting has been subject to recent criticism, because it can not identify the underlying fundamental causes of growth. However, this is not its objective. It provides a consistent decomposition of growth among its proximate sources which we believe is very informative.... Note also that growth accounting does not require us to take a stand on the appropriate underlying model of growth. We do not need to choose among a neoclassical framework in which technology is identical across countries and technical progress is exogenously determined and the many alternative frameworks in which technology may differ across countries and the accumulation of knowledge is an endogenous process (Collins & Bosworth, 1996, p. 5).

I appreciate the clarity of this analysis, although I have to call into account their argument that growth accounting removes the necessity of choosing between a neoclassical framework or an endogenous one. It seems to me that they are arguing (as is Young, although he does so less explicitly) that their work, and in general growth accounting is in fact ideologically pure. (Or perhaps they would be more likely to think of it as free of ideology.) I question the truth of this claim on the ground that no economic model is free of what are essentially ideological arguments. (And it is perhaps most important to look for ideology when an author claims to be perfectly neutral.) I will return to the strengths and limitations of growth accounting as a tool to use to assess the economic development of these nations below.

Growth Accounting

Growth accounting is an economic method designed to measure the relative and absolute contributions of different factors to economic growth and development. Developed by Robert Solow in 1957, this methodological approach disaggregates or decomposes the different elements of economic growth. The most important assumption of this method is that the gross output of an economy can be analyzed into increases in the range of factors (primarily increases in labor and in capital) and which cannot be accounted for by discernible changes in the utilization of these factors.

Another way of explaining Solow's model is this: The unexplained part of growth in an economy's GDP is best understood as a simple increase in productivity, with productivity being defined in common-sense terms as achieving a larger output without an increase in the input levels of any factor. Solow's model also suggests that this increase in GDP is the result of technological progress. This model has been used to assess a wide range of economies around the world and has tended to produce similar results, including the fact that in a range of situations the actual levels of economic growth cannot be accounted for by increases (or, conversely, decreases) in capitalization or labor force growth (or loss) rates.

Growth accounting allows for the extrapolation of different factors from a total economic profile to determine if factor accumulation is sufficient to explain economic growth. This form of methodological approach allows economists to isolate what we might call "a certain something extra" (because there is no reason that economics cannot support a certain amount of whimsy). This additional factor is called the Total Factor Productivity (TFP) (or the Solow residual) and serves as a measure of technological progress.

Krugman (1994) provides an elegantly commonsensiical explanation of growth economics:

We all do a primitive form of growth accounting every time we talk about labor productivity; in so doing we are implicitly distinguishing between the part of overall national growth due to the growth in the supply of labor and the part due to an increase in the value of goods produced by the average worker. Increases in labor productivity, however, are not always caused by the increased efficiency of workers. Labor is only one of a number of inputs; workers may produce more, not because they are better managed or have more technological knowledge, but simply because they have better machinery. A man with a bulldozer can dig a ditch faster than one with only a shovel, but he is not more efficient; he just has more capital to work with. The aim of growth accounting is to produce an index that combines all measurable inputs and to measure the rate of growth of national income relative to that index to estimate what is known as "total factor productivity."

Solow's initial use of this method was an attempt to determine the effect of technological growth on total economic growth. To do this (and this seems very simple in retrospect) he subtracted the growth rates of both labor and capital (both were weighted) from the growth weight of the total output. The residual (Solow posited) was the result of the growth of technology.

This initial use of growth accounting is now referred to as the primal approach and some economists have a significant doubt about its efficacy, which is that it is based on measurements of key economic inputs such as labor and capital. Measuring these input values can be very complicated -- in fact, are very complicated. This is true in the First World in which data are relatively clear-cut and relatively likely to be free of substantial manipulation and fraud for political reasons. Analogous data from developing nations are likely to be even more problematic since these figures are produced by government agencies that may lack expertise in calculating the needed figures. Or -- and this is far more likely than any lack of expertise -- is the fact that government agencies (and this is of course true in both the developing world and the developed world) may manipulate figures for a variety of political reasons. It is, of course, possible that governments may move beyond simple manipulation to outright mendacity in terms of their published statistics -- again, for a variety of either internal or external political reasons.

As a result of these facts, a dual approach of growth accounting was developed that is based not on quantities but on factor prices -- a shift based on the fact that factor prices are usually easier to measure in accurate ways. Prices tend to be much more accurate because they are determined at the market, where there are a range of incentives to get the prices right. (Of course, marketplaces are not immune from the possibility of manipulation, but for the moment we will set that aside.)

Hsieh (2002) summarizes this:

The advantage of using the national income identity rather than the cost function approach is that the national income identity derivation makes it explicitly clear that the equivalence of the dual and primal procedures do not depend on any assumptions about the underlying technology or market structure (p. 504).

Barro (1998) provides an excellent overview of the usefulness of growth accounting in the assessment of economies undergoing rapid change. Growth accounting, he writes "is generally viewed as a preliminary step for the analysis of fundamental determinants of growth and is especially useful if the determinants of factor growth rates are substantially independent from those that matter for technological change" (p. 1). He also underscores one of the most significant limitations of growth accounting, which is that factor prices coincide with social marginal products. (Although, as he notes, there are methods for calculating what occurs when this is not the case.)

However, while there are differences between primal and dual approaches in growth accounting that lead to different results, this difference can sometimes be more theoretically important than factually so, as Hlousek (n.d.) determines through a study of Czech economics:

The primal and dual measures of TFP growth rate should be the same with only the condition that output equals factor incomes. No other assumptions about the form of the production function, bias of technological change or relationship between factor prices and their social marginal products need to be made.The two measures will differ only if the national accounts are inconsistent with the data on factor prices.

One thing that is not clear from Hlousek's study is to what extent his analysis holds steady for developing nations. While, theoretically, economic models should in fact be applicable in different circumstances, the highly variable accuracy of economic data (as noted above) ensure that such transferability may not be possible. (Hlousek is fully cognizant of this, writing that "Further research will be also focused on other countries and cross-country comparisons of TFP growth rates.") However, setting this aside, it is important to acknowledge the similarity in accuracy between the two methods of growth accounting -- both of which methods determined that in the case of the Czech republic that TFP (Total Factor Productivity) is more important that factor accumulation in understanding Czech economic dynamics.

This paper examined two approaches to growth accounting: primal that is based on quantities of factor inputs and dual that is based on factor prices. The analysis used Czech time series of aggregate variables. The results of the exercise are quite satisfactory. Both estimates of TFP growth are very similar and dual approach is useful alternative to measuring TFP (Hlousek, n.d.)

Having providing an overview (albeit a brief one) of growth accounting, I now turn to a discussion of the ways this methodology interacts with neoclassical economic models.

Neoclassical Economic Models

Neoclassical economics is not a perfectly homogeneous entity, with different practitioners in different eras (and with different motivations) defining the subfield somewhat differently. However, there is certainly a core of beliefs and intellectual assumptions in the field. These assumptions include the following:

People are rational beings who have the ability to assess the value of different options and choose the most valuable among them

Individuals (being both rational and perceptive) work to maximize utility (or worth)

Businesses (being analogous to people and so capable of reason and intelligent choices) work to maximize their profits

Individuals interacting with the marketplace act independently of each other and in possession of complete and accurate information.

In other words, the marketplace is filled with individuals and groups who purse their own (and only their) interests in rational and accurate ways to increase the amount of goods or other forms of wealth that they have.

One of the essential methodological aspects of neoclassical economic models and analyses is that in such models market supply and demand are aggregated across firms and individuals, thus allowing for patterns to emerge through a reduction in natural variation (as well as what might be seen as less-than-fully rational behavior on the part of individuals or firms. Neoclassical models emphasize the importance (and, some would argue), the inevitability of equilibria.

Hsieh argues that growth accounting techniques strongly support a neoclassical explanation for growth in these four nations:

These studies suggest that factor accumulation has been the lead actor in East Asia's growth, many economists have reached the conclusion that the industrial revolution in East Asia can largely be explained as transition dynamics in a neoclassical growth framework. More broadly, these studies reinforce the message that a minimalist neoclassical growth model, perhaps augmented with human capital, is sufficient to explain why some countries are rich and others are poor.

Kim (2001) makes the link between Hsieh's argument above and implications for future policy decisions. Although I will discuss this more below, I would like to comment on it briefly at this point. What I am examining here is different possible explanations for the extremely rapid growth of the Four Tigers from the 1960s to the 1990s and the reduced growth in these nations since that time. While I would never argue that knowledge for its own sake is not by definition valuable, I would also argue that economic research is made more valuable if it can provide models for future economic policy.

At the core of this paper -- and, of course, of economic theory in general -- is the question of whether governmental guidance of (or interference with, depending on one's point-of-view) can reliably, predictably, and beneficently guide economic growth or should the markets be allowed to be untrammeled. Those who advocate for governmental involvement in economic policy favor endogenous models while those who view the marketplace as sufficient (or as nearly sufficient) refer to neoclassical models to explain the workings of the economic world.

Kim (2001) argues that neoclassical models are the most accurate ones for explaining the development of the economies of these four nations. Neoclassical economics models explain not only the feverish pace of growth over three decades but also the more recent slowdown (this slowdown being relative, of course). Moreover, Kim writes, that purely neoclassical modeling provides the appropriate lens to view future developments in this region.

Krugman's critical view on East Asian growth is based on a series of empirical studies that reported meager total factor productivity (TFP) growth in East Asia... These studies presented empirical findings that the growth rates of TFP in East Asian economies are not as spectacular as their output growth, which is surprisingly against our expectation. According to the neoclassical economic growth model, the growth based on only physical capital accumulation is destined to a halt as it reaches the steady state, since the growth rate is determined by the rate of productivity growth in the long run. Thus, the growth based on accumulation without efficiency improvement raises a serious concern about future growth (Kim, 2001, p. 1).

To understand the point that Kim is making -- that the best way to understand the pattern of growth in the region is to use a neoclassical economic model based on data supplied through growth accounting -- it is important to detail the implied contrast he is making. He is, in effect, arguing that neoclassical economic models and policies are sufficient to understand what is occuring in the four nations and that there is no need to call on endogenous growth theory.

Endogenous growth theory (which is sometimes also called new growth theory, although the model is now a quarter of a century old) diverges from neoclassical models in that endogenous models argue that government policies can have significant (and predictable) effects on the economy. While a neoclassical economist, for example, would argue that individuals and firms will innovate without any (added) incentive simply because the market rewards innovation, followers of endogenous models argue that innovation are more likely to occur if governments offer inducements for innovation -- inducements that will mesh with the market's incentives in a form of synergism.

I am arguing here that there is no need to rely on the arguments and suppositions of endogenous models to explain the rate of growth since the 1960s of Hong Kong, Singapore, South Korea, and Taiwan but rather that neoclassical models are sufficient (when joined with growth accounting methods) to describe as well as to explain the varying (but predictable) growth rates in the region.

One of the key findings made by the scholars whose work I have been citing (and backed up by a range of reports on these countries' economic activity is new technological options have been relatively unimportant in the four countries. The importance of new technologies in explaining growth is central to most endogenous models, and the lack of a need to include significant technological improvements in models to explain growth in this region is a key reason to prefer neoclassical models to endogenous ones.

Except -- to be fair in my analysis -- there is some doubt as to whether neoclassical models are in fact the most accurate models for the growth in these countries. While not conclusive, a study from the IMF suggests that increases in new technology may account for more of the growth rate in these four countries than I have been positing up to this point. Although I do believe that a neoclassical model is better (and even the author of the study that I am citing here -- Sarel [1996] notes that the data are inconclusive) I think it is important to acknowledge that technological innovation may account for a significant amount of the growth in the region. The relevant data are summarized in this table:

"http://www.imf.org/external/pubs/ft/issues1/fig3_4.gif" d?

(http://www.imf.org/external/pubs/ft/issues1/index.htm)

Sarel summarizes the importance of these data, and especially of Table Four. These data demonstrate that while each of the Four Tigers demonstrated an accumulation of both capital and labor at a rate that was considerably father than other economies both in the region and globally, this increase is insufficient to explain the total rate of growth. This leads Sarel to conclude that a significant amount of the growth in these nations must be attributable to growth and innovation in technology -- taking us off of the path of neoclassical economic modeling.

Except, perhaps not, as Sarel notes. When he subjects his data to a sensitivity analyses demonstrates that when even small but simultaneous changes occur in his parameters "together yield an estimate of productivity growth significantly lower than the baseline result." This leaves him unable to commit fully to either a neoclassical or an endogenous position, or as he writes: "the debate over the relative contribution to economic growth of factor accumulation vs. more efficient technology is still very much alive."

Seeing What We Expect to See

I have been making two related arguments here. The primary one of these is that a neoclassical model of economic growth is sufficient to explain the past half century of growth in the four nations under discussion. A related, or, rather, embedded argument, is that growth accounting procedures provide us with the kind of reliable, accurate, and comparable data that are needed to run neoclassical models. I believe that I have supported this argument, although I acknowledge that there is always more than one way to parse an economic data set. This is, in large measure and as noted above, because economic models are never pure math but are in fact the result of ideological preferences.