Chapter 8. Diversification and the Economy: The Role of Government in Enhancing the Industrial Base

Reda Cherif, Fuad Hasanov, and Min Zhu
Published Date:
April 2016
  • ShareShare
Show Summary Details
Bill Francis, Iftekhar Hasan and Yun Zhu 

An individual economy develops and evolves along various paths. How different paths lead to a variation in economic prosperity and stability has always been a matter of curiosity to economists and policymakers. In this chapter, we look at how industry-level sectoral diversification affects economic growth and other measures of economic output, and how policymakers could help deepen industrial diversification.

We investigate a matrix of the economic outcomes that are normally overlooked or overshadowed by the pure size or the growth of an economy. Specifically, we look at poverty alleviation, technological innovation, flow of foreign capital, empowerment of women, and entrepreneurship. In general, we find that industrial diversification greatly improves economic output for all of these nonconventional measures.

Given the empirical evidence that industrial diversification leads to stable and healthy growth, we discuss a number of cases in which governments have initiated and supported small business development through direct and indirect capital supply, tax incentives, equity guarantees, and so on. We emphasize the strategic importance of government initiatives in the enhancement of sectoral diversification, without which the economy may take much longer and more social resources to deliver a similar outcome.

For policymakers, this chapter sends a strong message that the effort spent in supporting and enhancing industry sectors, which countries fall short on, will result in more stable and balanced long-term economic growth.

Diversification, Risk, and Growth

Any problem or debate on economic development starts with the measurement of GDP, which tracks annual growth, and by which the size of an economy can be compared across countries. However, policymakers and economists care more about the risk to growth and path of future development. As Lucas (1998) states, for advanced countries “growth rates tend to be very stable over long periods of time,” while for poorer countries “there are many examples of sudden, large changes in growth rates, both up and down.” Empirically, Ramey and Ramey (1995) find a negative correlation between variability of growth rates and average growth for Organisation for Economic Co-operation and Development (OECD) countries. Quah (1993) finds that for higher-income countries, the probability of falling to low-income status is small, consistent with the general observation that highly developed economies experience lower volatility, and less-developed economies the opposite.

Industrial diversification has been shown to be a necessary path for development. It reduces idiosyncratic shocks from individual sectors, and enhances capital accumulation, capital allocation, and technology improvement. In the beginning of the development process, industrial concentration may support GDP growth. But it does not guarantee a stable growth pattern, and is vulnerable to long-term macroeconomic shocks (Koren and Tenreyro 2013). Thus, higher variance of GDP resulting from sectoral concentration may entail a welfare loss that outweighs the benefits (Kalemli-Ozcan, Sørensen, and Yosha 2003). Macroeconomic fluctuations also result from shocks at the micro level. The growth of the financial industry, for example, contributed to the surge of macroeconomic volatility in the past decades (Carvalho and Gabaix 2013).

Reducing variability through diversification is achieved by technology improvement and capital supply. At the early stage of development, technology advancement can be considerably limited by education level and the scarcity of necessary resources. It is only at a certain stage of economic development that capital can be reallocated into education, research and development, and government-sponsored technology-driven projects that would accelerate growth and expand traditional industrial sectors into new areas. It is crucial that this step be carried out. If left to the private sector, an economy has a higher chance of falling into the trap of idiosyncratic risk shocks.

Capital supply is the second driving force. At early stages, lack of capital makes countries exploit their natural resource endowments and invest in safe but less productive projects. The limited opportunity for diversification makes the earlier stages of development highly random and subject to abrupt shocks in growth rates. Therefore, a typical development pattern will consist of a lengthy period of “primitive accumulation” with highly variable output, followed by takeoff and financial deepening, and, finally, steady growth. ‘‘Lucky’’ countries will spend less time in the primitive accumulation stage and develop faster (Acemoglu and Zilibotti 1997).

Diversification is a key step in development. It reduces volatility and transforms a simple economy, focused on a few industries, into a diversified and comprehensive economy with stable future growth. Technology and capital supply (including the development of the finance sector) are the two major factors that determine whether and how quickly a simple economy can evolve into developed status.

Industrial Development and the Economy

To provide an intuitive image of industrial development, we begin with a simple observation of the diversification patterns, and discuss a few important measures that capture various aspects of the development of an economy.

Diversification Patterns around the World

Figure 8.1 shows a plot of the time series diversification pattern of the United States, OECD countries, Gulf Cooperation Council (GCC) countries, and Middle East and North Africa countries.1 We use the Herfindahl-Hirschman index (HHI) of sectoral output as a common measure of diversification.2 We use industry and sectoral output data from the United Nations Industrial Development Organization, which has the most comprehensive coverage of the industrial output of 23 two-digit industrial sectors (127 in four-digit sectors) for 166 economies from 1963 onward.

Figure 8.1Global Diversification Model

Source: United Nations Industrial Development Organization.

Note: The Herfindahl-Hirschman index ranges between zero and one, with a larger number indicating higher concentration or less diversification. GCC = Gulf Cooperation Council; MENA = Middle East and North Africa; OECD = Organisation for Economic Co-operation and Development; US = United States.

Figure 8.1 shows that the United States and OECD countries had the highest and most stable diversification patterns in the past 50 years. The United States has the lowest concentration in our sample, with HHI of industrial concentration below 0.1. The HHI for OECD countries was just above 0.1 in 1963 and declined until the global financial crisis.

Middle East and North Africa countries experienced some fluctuations in their industrial concentration, partly coming from the GCC countries. The Middle East and North Africa countries together had an HHI of about 0.2, indicating reasonable diversification for developing or natural-resource-dependent entities. We also observed a slight increase in HHI since the early 2000s, partly due to the politics-induced change in economic environment post-2001 and partly due to the global financial crisis.

GCC countries have very low diversification and higher volatility. Their average HHI ranged from 0.4 to 0.5 over the past 50 years. Because of the limited number of industrial sectors and oil price volatility, the GCC experienced greater output volatility. Among all the countries in our sample, these have the lowest diversification.

For further insight into the industrial sectors in GCC countries, we plot the industrial output share of the top five industries of GCC countries in 2010. The results are shown in Figure 8.2. The sector coke, refined petroleum products, and nuclear industry dominates, with 73 percent of GCC industrial output. As shown in Figure 8.3, no single industry has a dominant share of worldwide industrial output, with the coke, refined petroleum products, and nuclear industry sector having an 8 percent share. This shows that a single-industry-dominated economy inevitably faces greater risk and external shocks.

Figure 8.2Shares of Top Industries in Total Output in GCC, 2010

Source: United Nations Industrial Development Organization.

Note: Percentages of industry output are in relation to total economic output. GCC = Gulf Cooperation Council.

Figure 8.3Shares of Top Industries in Total Output Worldwide, 2010

Source: United Nations Industrial Development Organization.

Note: Percentages of industry output are in relation to total economic output. This figure also includes a segment for coke, refined petroleum products, and the nuclear industry.

The Effect of Diversification

Here, we study the impact of diversification on economic outcomes beyond GDP. We specifically explore income dispersion, technological innovation, the net flow of foreign capital, the empowerment of women, and entrepreneurship. We consider all of these as the key outcomes of a healthy economy.


We use the industry and sectoral output data from the United Nations Industrial Development Organization to calculate diversification.3 We collect country-level characteristics from the World Bank’s World Development Indicators, and innovation information (patents) from EPO Worldwide Patent Statistical Database. The main variables of interests corresponding to the above economic outcomes are:

  • The Gini index and income share held by the lowest 10 percent of earners, to capture income dispersion.
  • Patents (log), representing the country-level innovation output.
  • Foreign direct investment (net, percent of total output) that proxies for the capital flow between a country and the rest of the world.
  • The female-to-male employment ratio, which is used as an economic proxy for women’s empowerment.
  • New business registrations, a proxy for the entrepreneurial environment.

Figure 8.4 shows the time-series distribution of these variables. OECD countries have a lower level of income dispersion, more innovation output, a better environment for international capital flows, lower gender inequality, and a better environment for entrepreneurs. We also observe the improvement in almost all of the variables in the non-OECD countries.

Figure 8.4Time-Series Distribution of Key Variables of Interest

Sources: EPO Worldwide Patent Statistical Database; and World Bank, World Development Indicators.

Note: OECD = Organisation for Economic Co-operation and Development.

Measurement of Industrial Diversification

To fully gauge a country’s industrial diversification, we employ a number of measures for the concentration of output across industrial sectors: (1) the HHI for sectoral output over a year, calculated as the sum of the squares of the shares of each industrial sector; (2) the maximum-minimum spread, calculated as the spread between the largest and smallest industrial sector shares; (3) the mean-median spread, calculated as the spread between the mean and median value of the industrial sector shares; and (4) the log-variance of sector shares, calculated as the log of variance of all the industrial sector shares.

Empirical Findings

The main results of the effects of diversification on various economic outcomes are presented in Table 8.1. We use ordinary least squares regressions, controlling for country characteristics such as GDP per capita, inflation rate, interest rate, and population (log), as well as year and country fixed effects. The main independent variables are four measures of industrial diversification; namely, the HHI for the sectoral output, the maximum-minimum spread, the mean-median spread, and the log-variance of sector shares. For simplicity, we suppress the control variables in the table.

Table 8.1Empirical Results of a Study of the Effects of Diversification on Select Economic Outcomes
VariablesGini IndexIncome Share Held by Lowest 10 PercentPatent Number (Log)
Maximum-minimum spread45.113***−5.239***−3.488**
Mean-median spread301.383***−21.048***−42.400***
Log variance234.323***−19.964***−19.182***
Country fixed effectYesYesYesYesYesYesYesYesYesYesYesYes
Number of observations4964964964965255255255252,2542,2492,2492,249
Adjusted R20.1200.1730.0820.2000.1350.2100.0600.1690.7030.7070.7170.714
VariablesFDI (Net, % of Total Output)Female-to-Male Employment RatioNew Business Registration
Maximum-minimum spread2.369−0.103−1.073**
Mean-median spread19.407**−0.664*1.515
Log variance14.030*−0.486*−2.522
Country fixed effectYesYesYesYesYesYesYesYesYesYesYesYes
Number of observations4434434434431,1451,1451,1451,145425425425425
Adjusted R20.3710.3680.3950.3840.5460.5450.5450.5460.9820.9820.9820.982
Note: FDI = foreign direct investment; HHI = Herfindahl-Hirschman index. Standard deviations are in parentheses.* p<.05, ** p<.01, *** p<.001.
Note: FDI = foreign direct investment; HHI = Herfindahl-Hirschman index. Standard deviations are in parentheses.* p<.05, ** p<.01, *** p<.001.

The results are, in general, consistent with our main argument that diversification advances social welfare. The statistical significance varies among different measurements of diversification. Specifically, industrial diversification alleviates poverty by reducing income dispersion and increasing the income of the bottom 10 percent of a population.4

We also find that diversification improves country-level innovation (annual filings of patents). It has some positive but a generally weak effect on cross-border capital flow. Foreign direct investment can be affected by factors other than industrial diversification, but our results provide some evidence that with more industries—and more evenly distributed industrial pattern—an economy is able to attract more capital inflows.

The World Bank’s World Development Indicators Database does not provide comprehensive coverage on the employment ratio of women in many countries. However, even within well-developed economies where the social status of women is relatively high, we still observe the positive influence of industrial diversification on women’s empowerment. The new business registration improves with industrial diversification. As diversification promotes services sectors, more opportunities for small business are created.

Industrial diversification enhances economic and social welfare in multiple dimensions, and, in the next section, we discuss how government should facilitate diversification.

Diversification and the Role of the Government

Advantages of Government Support

Without going into the debate over whether a small or large government is better for economic development, government can take roles to subsidize education and research and development, and to provide venture capital with fiscal and regulatory frameworks. Governments in OECD countries have decades of experience in implementing direct programs to mobilize venture capital in support of small and technology-based firms and to produce public benefits through innovation and job creation. Government involvement aims to remedy deficiencies in private capital markets, to leverage private sector financing, and to reduce aggregate risk by diversifying the economy.

Evidence is ample that government involvement can provide greater social rates of return than the private sector. Various schemes can bring public benefits by targeting small firms with good job creation potential, or firms that can develop technologies important to long-term growth.

The government can fill funding gaps that prevent viable small businesses from obtaining funding on reasonable terms. Access to traditional venture capitalists is not available for many small firms, even in the United States and other OECD countries, where financial systems are highly developed (let alone in countries with lower levels of investor protection and capital liquidity).

By providing a small amount of financial support, governments can certify firms and reduce information asymmetry—and they do not have to completely close the funding gap. The certification effect involves bringing in other financial institutions, such as venture capitalists, private equity firms, banks, and so on. In many OECD countries, governments usually provide a small portion of seed funding, and, in doing so, greatly enhance the chance that funded firms will enjoy long-term growth.

Sponsoring nascent industries and technology helps diversify industrial sectors and further reduce aggregate shocks. And this is consistent with the chapter’s argument that promoting new industries and supporting new technologies can be more effective with the help of government.

Governments therefore can and should play a pivotal role in industrial diversification. The next section looks at how this is done in OECD countries.

Financial Support for Industry Diversification in OECD Countries

There are two major types of support (shown in Table 8.2): to directly supply funding and to provide financial incentives, such as tax benefits, loans, and equity guarantees.

Table 8.2Types of Financial Support in OECD Countries
Direct Supply of Capital
Research fundingProvide research and development funding for small-sized firmsUnited States: Small Business Innovation Research
Government equity investmentMake direct investments in venture capital firms or small firmsBelgium: Investment Company for Flanders
Government loansMake low-interest, long-term and/or nonrefundable loans to venture capital firms or small firmsDenmark: VækstFonden (Business Development Finance) loan program
Financial Incentives
Tax incentivesProvide tax incentives, particularly tax credits, for investing in small firms or venture capital fundsUnited Kingdom: Enterprise Investment Scheme and Venture Capital Trusts
Loan guaranteesGuarantee a proportion of bank loans to qualified small businessesFrance: Société Française de Garantie des Financements des Petites et Moyennes Entreprises
Equity guaranteesGuarantee a proportion of the losses of high-risk venture capital investmentsFinland: Finnish Guarantee Board
Source: Organisation for Economic Co-operation and Development (OECD).
Source: Organisation for Economic Co-operation and Development (OECD).

A number of governments choose tax incentives, particularly investor tax credits, to stimulate particular types of investment. Incentives may be available for investments made directly in qualifying small companies or for investments made in qualified pooled vehicles. An important decision in program design is whether the tax incentive should be given at the front or back end, which is tied to any capital gains realized at exit. The first approach rewards all investors, whereas the second rewards only winners.

Another difference between the two approaches is that front-end incentives may cause behavior motivated primarily by tax-shelter considerations. For example, Canada gives tax incentives to hybrid public and private funds as so-called Labor-Sponsored Venture Capital Corporations (LSVCCs), whose asset growth was particularly high in the 1990s. At the end of 1995, LSVCCs represented 49 percent of Canada’s $6 billion of venture capital assets. In that year, hybrid funds raised $1.2 billion, whereas private funds raised only $0.3 billion. LSVCCs invest in Canadian small and medium-sized enterprises, and investments dedicated to early-stage deals, which represented 34 percent of their 1995 investments. The attraction of LSVCCs is that an investor receives a federal tax credit of 15 percent on up to $3,500 of an investment held for five years. In addition to the federal credit, investors in Ontario and Quebec, which account for the bulk of LSVCC funds, receive a 15 percent tax credit on these investments.

Most OECD countries offer some form of government-backed guarantee covering loans to small firms. Typically under these programs, the government guarantees a percentage of a qualified loan made by a financial institution. In the event of default, the loss incurred by the lender is only for the amount of the loan not covered by the guarantee. The aim of these programs is to encourage financial institutions, particularly commercial banks, to fund small firms that have viable projects but cannot meet collateral requirements.

The United States has strong support for research and development (R&D), especially for basic R&D (Figure 8.5). A major outlet of R&D funding is through financial support for small business.

Figure 8.5Financing Sources of Research and Development (R&D) in the United States, 2008

Source: National Science Foundation.

Since 1953, the U.S. Small Business Administration (SBA) has run the 7(a) Guaranteed Business Loan Program that guarantees long-term loans to start-up and high-potential companies. These loans are then guaranteed by the SBA for up to 75 percent of the amount provided by the commercial lender. Interest rates are negotiated between the borrower and the lending bank. The maximum amount currently guaranteed by the SBA is $500,000. Between 1980 and 1991, the SBA guaranteed $31 billion in loans through the 7(a) program. Its default rate was 17 percent on loans guaranteed in 1995/96, and this varied over the life of the program; in 1983/84 it was 25 percent, but declined to 9.5 percent in 1992/93.

The Small Business Innovation Research (SBIR) Program was set up to promote the technological innovation in small-sized firms in the United States. The program has provided 20–25 percent of total funding for early-stage technology firms. In 2010 more than $1 billion in research funds were granted; over half the awards were given to firms with fewer than 25 people and a third to firms with fewer than 10. A fifth of these firms were minority or women-owned businesses. A quarter of these firms in 2009/10 were first-time recipients. Some of America’s most dynamic companies have received support through federal programs, including Apple Computer, Compaq, FedEx, and Intel. In addition to funding firms, publicly sponsored funds provided early experience for many individuals who went on to lead independent venture organizations.

Empirically, Lerner (1999) shows that SBIR recipients enjoy substantially greater employment and sales growth, and were more likely to go on to receive private venture capital financing. The superior performance was confined to firms in regions with substantial venture capital activity, and was pronounced in high-technology industries.

Similar programs have been initiated in other OECD countries. For example, Germany’s government equity investment program “Beteiligungskapital für junge Technologieunternehmen” (BJTU) reduced the failure rate of companies that it financed to 17 percent. The Netherlands’ Technical Development Credits Scheme provides subordinated 10-year loans to firms for the development of new products, services, or processes. Here, repayment is based on firms’ revenues and, in the event of technical or commercial failure, the loan is forgiven. The United Kingdom’s Loan Guarantee Scheme has a repayment schedule of between 2 and 10 years. Companies less than two years old are eligible for a 70 percent guarantee on loans up to £100,000; older companies are eligible for an 85 percent guarantee on loans up to £250,000.


Economic development is not merely a matter of real GDP growth. It trickles down to other important aspects of society, such as inequality, innovation, and the empowerment of women. In this chapter, we demonstrate that industrial diversification is one of the most important economic variables.

By bringing supporting empirical evidence, we show that industrial diversification is correlated with poverty alleviation, technological innovation, the empowerment of women, capital supply, and entrepreneurship. All these are the common factors that we nowadays use to evaluate the prosperity of an economy and the health of a society.

Together with the empirical evidence, we review a number of practices in OECD countries in which government is involved in small business development through direct and indirect capital supply, tax incentives, equity guarantees, and so on. We argue that the strategic financing support of government is crucial for promoting small business and diversifying industrial sectors. Without this support, it may take much longer and more social resources to deliver a similar outcome, with a high possibility of falling into a slow and volatile growth pattern.


    AcemogluD. and F.Zilibotti.1997. “Was Prometheus Unbound by Chance? Risk, Diversification, and Growth.Journal of Political Economy105 (4): 70951.

    • Search Google Scholar
    • Export Citation

    CarvalhoV. and X.Gabaix.2013. “The Great Diversification and Its Undoing.American Economic Review103 (5): 16971727.

    Kalemli-OzcanS.B. E.Sørensen and O.Yosha.2003. “Risk Sharing and Industrial Specialization: Regional and International Evidence.American Economic Review93 (3): 90318.

    • Search Google Scholar
    • Export Citation

    KorenM. and S.Tenreyro.2013. “Technological Diversification.American Economic Review103 (1): 378414.

    LernerJ.1999. “The Government as Venture Capitalist: The Long-Run Impact of the SBIR Program.Journal of Business72 (3): 285318.

    • Search Google Scholar
    • Export Citation

    LucasR. E.1998. “On the Mechanics of Economic Development.Journal of Monetary Economics29: 6170.

    QuahD.1993. “Empirical Cross-Section Dynamics in Economic Growth.European Economic Review37 (2): 42634.

    RameyG. and V. A.Ramey.1995. “Cross-Country Evidence on the Link between Volatility and Growth.American Economic Review85 (5): 113851.

    • Search Google Scholar
    • Export Citation

OECD includes Australia, Austria, Belgium, Canada, Chile, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, Korea, Luxembourg, Mexico, Netherlands, New Zealand, Norway, Poland, Portugal, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Turkey, United Kingdom, and United States; GCC includes Bahrain, Kuwait, Oman, Qatar, Saudi Arabia, and United Arab Emirates; Middle East and North Africa includes Algeria, Bahrain, Egypt, Iran, Iraq, Israel, Jordan, Kuwait, Lebanon, Yemen, United Arab Emirates, Libya, Morocco, Oman, Palestine, Qatar, Saudi Arabia, Syria, and Tunisia. The list varies over time as countries joined at various times.


Specifically, in each country (or region) and each year, we first calculate the share of each industrial sector by dividing the sectoral output with the overall output of the economy. The HHI is then calculated by the sum of the squares of the shares of each industrial sector. The HHI ranges between zero and one, with a larger number indicating higher concentration or less diversification.


We mainly use the output from the two-digit industry sectors, since a large number of countries do not have comprehensive information on four-digit industrial sectors.


The results may be driven by the natural gap between developed and developing countries, because more developed economies, mostly under democratic systems, had lower income dispersion throughout the sample period. However, the inclusion of fixed effects should mitigate this problem.

    Other Resources Citing This Publication