Professor Emeritus David Lim, Griffith University, Brisbane, Australia

One of the most important concepts in economics is comparative advantage—the ability to produce goods and services at a lower opportunity cost than others, which forms the basis for exchange. The Heckscher-Ohlin theory suggests that this ability comes from differences in factor endowments, where a country will do well producing goods and services that use relatively more of its abundant resource. Thus, a developing economy with abundant labour and little capital and technology should produce labour-intensive and low-technology goods. By implication, it should not offer investment incentives that subsidize capital. And whatever capital it has should be used to the full. This article looks back at the work I did on these issues from the mid-1970s for Malaysia.

Capital subsidies and capital intensity

Malaysia offered many incentives to attract investment, especially foreign direct investment (FDI), in the period under review. The operative legislation was the Investment Incentives Act of 1968. Its most important provision for firms that were granted ‘pioneer status’ was the tax holiday, whose duration depended on the level of investment, so the larger this was, the longer the firm paid no company tax. Added to this were accelerated capital depreciation, investment allowances, and subsidized bank loans, which directly subsidized capital expenditure. A priori, by reducing the cost of investment, these incentives would, other things being equal, encourage the establishment of capital-intensive activities. This is something that labour-abundant countries such as Malaysia should not do.

The data from Malaysia's 1972 Survey of Manufacturing conducted by the Department of Statistics–Malaysia, showed that pioneer firms were much more capital intensive than the rest at the industry-group level for 12 such groups. In general, empirical evidence shows that this greater capital intensity was caused by the provision of capital subsidies. 1A caveat of this simple two-variable regression is that large firms, especially foreign ones, might have been given capital subsidies even if they did not seek them because of their size and their degree of foreign ownership. If so, then just using capital subsidy would have introduced biases into the relationship, as it incorporated the effects of scale and reputation via their influence on the cost of capital.

It is obvious that as the size of firms increases, they are likely to be more capital intensive as it is easier to achieve quality control for mass production using ‘machine-paced’ rather than ‘operator-paced’ techniques of production when there is a shortage of skilled workers. The higher the degree of foreign ownership of the firms, the more likely they are more capital intensive because foreign firms prefer using the technology they are familiar with, regardless of factor cost and scale considerations. They could also be reluctant to get too involved in the social and political problems associated with employing local labour. I then re-estimated the simple two-variable regression by introducing the two additional variables mentioned above.

Industrialization in Malaysia
EHM video ‘How Malaysia Has Progressed Over the Past 100 Years’

The results were mixed, with half showing the granting of pioneer status had, via their subsidy to capital, increased the capital intensity, and half not (Lim, 1980, pp. 22–30). However, there was no ambiguity on the positive impact of the scale of operation and, to a smaller extent, the degree of foreign ownership. This was an unsatisfactory finding on the impact of capital subsidies on capital intensity, and the smallness of the sample of 12 observations added to the tentativeness of the results.

I returned to the topic a year later when data from the 1973 Census of Manufacturing by the Department of Statistics–Malaysia became available. This has a bigger sample of 26 industry groups and a 100 per cent, not the 75 per cent, coverage of the 1972 survey. I was also able to use the data I collected for an earlier World Bank study on capital utilization (touched on later) for 350, or 10 per cent of, manufacturing establishments for 1972 from 28 industry groups (Lim, 1976b, pp. 209–217). I used the same estimating equation. Using this dataset, the results on the influence of pioneer incentives in determining greater capital intensity were clearer (Lim, 1981, pp. 376–382). However, as with the earlier study, by far the most important determinant was the scale of operation. There was also some evidence that foreign firms preferred using technology that was familiar to them.

Many years later in 1989, I was given the opportunity to revisit the issue with far superior data from a project, conducted by the International Labour Organization, United Nations Development Programme, and the Economic Planning Unit–Malaysia, on the impact of economic policies in Malaysia on employment creation in different economic sectors. I was given manufacturing and had data for nearly 5,000 pioneer status and non-pioneer status establishments for 1979, the last year the Department of Statistics–Malaysia collected information on pioneer status establishments. As the establishment was the unit of observation, there was no need to artificially combine establishments from technically different production lines into one industry group, or to use an inappropriate weight to derive variables at the industry-group level.

This new set of data allowed me to refine the estimation further by redefining the variables measuring capital intensity and size, dummy variables for capital subsidy and foreign ownership, and introducing a dummy variable measuring the effect of corporation. Incorporated establishments, unlike sole proprietorships and partnerships, were more likely to be granted capital subsidies.2
The study showed that the capital-based incentives produced greater capital intensity (Lim, 1992, pp. 705–716). In addition, the higher the pioneer incentive in the form of capital subsidy, and the larger the size of the firms, the greater the capital intensity of the firms. Also, incorporated establishments had higher capital intensity than sole proprietorships and partnerships. The degree of foreign ownership, however, had no effect on capital intensity of firms. The methodology was repeated for the 1976 data for nearly 4,000 establishments. This time, all the above variables above were important determinants of capital intensity. In particular, granting pioneer status produced greater capital intensity, a conclusion supported by Fong and Lim (1984, pp. 396–418).

Capital subsidies and capital utilization

The dominant thinking in the 1960s was that countries were, and remained, poor because they lacked capital, and to make them grow faster all one had to do was to increase their investment ratio. The exact amount was estimated using the Harrod-Domar model connecting the growth rate of output to the savings ratio and the capital-output ratio, a measure of capital productivity.3  This relationship was introduced in the 1930s to analyse the conditions needed for steady economic growth in a Europe beset by unemployment and instability after the Great Depression, but also used widely in poor economies to estimate the level of capital needed to achieve a target growth rate of output. This capital-centred approach simply says that growth depends on the quantity of capital available and the productivity of that capital.

During this period, Malaysia’s five-year development plans, and those of many developing economies, were formulated on this basis. Once the needed investment level has been estimated, policies are introduced to increase the level of capital from domestic savings, FDI, and foreign aid, and to raise efficiency in its use. The corollary of a capital-centred approach to growth is that the capital available is utilized heavily. I was able to find if this was the case in 1973 in a World Bank project on capital utilization in manufacturing in Colombia, Israel, Malaysia, and the Philippines. The first thing we did was to seek an objective measure of capital utilization.

We rejected the widely used McGraw-Hill measure because production managers are not given a definition of full capacity when asked the percentage of full capacity at which they are operating their plants. For example, a firm that believes full capacity is 50 per cent of the total number of hours available in a day, after time is taken out for maintenance, will report a higher level of utilization than one that sees it as 67 per cent. We also rejected the electricity measure, where utilization is the percentage of the rated capacity of the electric motor plants used to drive the capital equipment.  This 

Labour-abundant economy in Malaysia
Photo source:
Laura Low

was because, although nearly all modern factories operate on electricity, some factories use solar energy because it is cheaper than grid supply, which is not always reliable anyway. We decided instead on a time measure, where capital utilization is the percentage of 24 hours the factory is operated. This excludes the time used for maintenance, an engineering requirement that is more or less the same for equipment in the same industry. These conclusions were presented in a research seminar at Oxford and published (Lim, 1976a, pp. 149–159).

The data on capital utilization using the time measure were obtained by surveys at the establishment level. For Malaysia, we collected valid data for 350 establishments, around 10 per cent of the total number at the time, with these spread across the industrial sub-sectors according to their shares in total manufacturing net output. The results showed that the utilization level was lower than expected, which confirmed the casual observations. This measure, with others, was used to produce the actual, desired, and full levels of capital utilization in Malaysian manufacturing (Lim, 1977a, pp. 53–62).

We also spent much time identifying the reasons for the low level of capital utilization. The results came out in the bigger cross-country study published some years later (Bautista et al., 1983). As the survey had collected so much data, I was able to use it to carry out research on other related aspects of the Malaysian economy. One was to compare the economic performances of foreign and local companies. I found that foreign firms used their capital plants more than their local counterparts, not because of their greater X-efficiency, as popularly believed, but their greater scale of operation, capital intensity, and benefits derived from input of professional managers (Lim, 1976b, pp. 209–217).4  Foreign firms also paid higher wages again because of their greater size, and their tendency to pay wages more commensurate with those at home (Lim, 1977b, pp. 55–66). Another area was to compare the wages paid by incorporated private and public limited companies and those paid by unincorporated sole proprietorships and partnerships. The former paid more because of their greater scale of operation and efficiency, and because their greater public exposure ensured that they could not be seen to pay ‘sweat labour’ wages (Lim, 1978, pp. 75–81).

Concluding remarks

My analysis of national survey data suggested two main conclusions. First, Malaysia offered investment incentives that favoured capital, which resulted in more capital-intensive manufacturing activities than should be the case for what was at the time a labour-abundant and capital-scarce economy. Second, Malaysia had a low level of capital utilization for a country lacking capital. This seems ironic but it could help to explain why the investment-dependent duration of the tax holiday was dropped when the Investment Incentives Act of 1968 was replaced by the 1986 Promotion of Investments Act.

The provision of investment incentives also ironically did little to encourage FDI (Lim, 1983, pp. 207–212), a view shared by a Bank Negara Malaysia working paper (Mohd Shazwan Shuhaimen et al., 2017). FDI is decided fundamentally by proven economic performance and, where relevant, the presence of mineral deposits (Lim, 1983, pp. 207–212). Countries lacking these compete among themselves in giving more incentives, while all the time foreign firms decide where and how much to invest according to non-tax criteria, knowing full well that they would pay very little tax wherever they went (Shah and Toye, 1978).

Further reading
Bautista, R., H. Hughes, D. Lim, D. Moratwez, and F. Thoumi. 1983. Capital Utilization in Manufacturing in Developing Countries. New York: Oxford University Press.
Fong, C. O. and K. C. Lim. 1984. ‘Investment Incentives and Trends in Manufacturing Investments in Malaysia’. The Developing Economies, v. 22, No. 4.
Lim, D. 1976a. ‘On the Measurement of Capital Utilisation in Less Developed Countries’. Oxford Economic Papers, v. 28, No. 1.
______ 1976b. ‘Capital Utilization of Local and Foreign Establishments in Malaysian Manufacturing’. Review of Economics and Statistics, v. 58, No. 2.
______ 1977a. ‘Actual, Desired and Full Levels of Capital Utilisation in Malaysian Manufacturing’. Journal of Development Studies, v. 14, No. 1.
______ 1977b. ‘Do Foreign Companies Pay Higher Wages Than Their Local Counterparts in Malaysian Manufacturing?’ Journal of Development Economics, v. 4, No. 1.
______ 1978. ‘Sweat Labour and Wages in Malaysian Manufacturing’. Economic Development and Cultural Change, v. 27, No. 1.
______ 1980. ‘Tax Incentives and Resource Utilisation in a Small LDC’. Australian Economic Papers, v. 19, No. 34.
______ 1981. ‘Another Look at the Effect of Capital Subsidies on Capital-Intensity’. Australian Economic Papers, v. 20, No. 37.
______ 1983. ‘Fiscal Incentives and Direct Foreign Investment in Less Developed Countries’. Journal of Development Studies, v. 19, No. 2.
______ 1992. ‘Capturing the Effects of Capital Subsidies’. Journal of Development Studies, v. 28, No. 4.
Mohd Shazwan Shuhaimen, Nurul Hana Ahmad Ghazali, K. T. F. Wong, P. L. Loke, and Sarah Syamimi Mohd Suhaimi. 2017. ‘Rethinking Investment Incentives’. Kuala Lumpur: Bank Negara Malaysia, Staff Insights 2017/12.
Shah. S. M. S. and J. Toye. 1978. ‘Fiscal Incentives for Firms in Some Developing Countries: Survey and Critique’ in J. Toye (ed.) Taxation and Economic Development. London: Frank Cass.

1 The definitions are capital/labour ratio, measuring capital intensity, as the dependent variable and capital subsidy, measured by the share of the pioneer firms’ labour force/value added in the industry’s labour force/value added, as the independent variable. 

2 The independent variable capital intensity is now the replacement value of the establishment’s fixed assets per full-time employee. The pioneer incentive variable is now a dummy variable, with 1 for each establishment granted pioneer status and 0 for each not granted or did not seek it. The size of the firm is now measured by the establishment’s sales. Foreign ownership now is another dummy variable, with 1 for each foreign-owned or controlled establishment and 0 for each of the others. For the effect of incorporation, 1 is given to each incorporated establishment and 0 to each unincorporated one.

3 The equation used is g = s/k, where g is the growth rate of output, s the savings ratio, and k the capital-output ratio, a measure of capital productivity.

4 X-efficiency occurs when firms produce the maximum output with given inputs. The situation is highly likely in competitive markets.

c/o Asia-Europe Institute
University of Malaya,
50603 Kuala Lumpur

Generic Popup