Economic freedom and unemployment around the world.

Over the past years, numerous empirical studies have analyzed how economic freedom affects levels and growth rates of GDP per capita (for a survey, see De Haan, Lundstrom, and Sturm 2006). In most cases, they found that economic freedom, in particular an increase in economic freedom, has a significant positive effect on both. However, so far it has not been analyzed whether and to what extent economic freedom affects unemployment. This paper is the first to tackle this issue. Section 1 describes the data used in this paper. Sections 2 and 3 present and discuss the regression results. Section 4 concludes.

1. Data

This paper applies the Economic Freedom of the World (EFW) index that has been developed by a group of North American economists under the auspices of the Canadian Fraser Institute with the aid of a worldwide network of further economists and institutes (Gwartney and Lawson 2005). The index has often been assessed to be a good measure of economic freedom and thus has been widely used in the empirical literature. It subdivides economic freedom into five areas: (i) size of government, (ii) legal structure and security of property rights, (iii) access to sound money, (iv) freedom to trade internationally, and (v) regulation of credit, labor, and business (for variable definitions and sources, see Appendix A). The two main authors of the index have explained their concept and measurement of economic freedom in each report (e.g., Gwartney and Lawson 2005) and elsewhere (e.g., Gwartney and Lawson 2003).

Economic freedom is complex and multidimensional. A major advantage of the EFW index is that it reflects all aspects of economic freedom and adequately groups them into five different areas. To provide information on the overall linkage between economic freedom and unemployment, we use the EFW summary index. Additionally, we use the five area indices. This allows us to analyze which economic freedoms have the strongest effects. As the correlation matrix indicates, there is substantial correlation among several of the area variables (Appendix B). Therefore, we estimate specifications that include these measures one at a time. However, the correlation matrix also indicates that the area variables are not highly correlated with each other. Therefore, we also present specifications in which we include them jointly. Estimating the effects of the area variables both separately and jointly allows us to check the robustness of our results for these variables.

The EFW index not only has the advantage that it covers all aspects of economic freedom and adequately groups them into five different areas. It also has the advantage that it is based largely on objective data. These data and the ratings derived on the basis thereof may be verified by anybody. Aspects of economic freedom that cannot be measured objectively (mainly in the areas legal structure and property rights and regulations) are measured using subjective data from the Global Competitiveness Reports of the World Economic Forum and the International Country Risk Guides of the PRS Group. The data in the Global Competitiveness Reports are based on annual surveys among approximately 4000 to 5000 senior business executives. The data in the International Country Risk Guides are computed from an in-house panel of experts of the PRS Group. In both cases, the data are compiled professionally. For example, the selection of respondents for the World Economic Forum's surveys is largely representative of the structure of each economy, and the respondents have excellent knowledge of their country's economy. The International Country Risk Guides are based on the assessments of experienced experts. (1)

For each variable included in the EFW index (a total of 38), the relevant data are translated into ratings with the aid of a formula that is published in the reports of the Fraser Institute. The rating scale ranges from 0 to 10, with 0 representing the lowest and 10 the highest degree of economic freedom. The area ratings for each of the five areas are calculated as the arithmetic means of the ratings for their respective components. In turn, the summary rating is the average of the five area ratings.

Applying the most recent methodology, which was further developed over the years, the EFW index was calculated for every fifth year from 1970 on plus for the years 2001, 2002, and 2003. As no unemployment data compiled according to a uniform methodology are available for years prior to 1980 (see the following discussion), the following regressions are based on data for the years 1980, 1985, 1990, 1995, 2000, 2001, 2002, and 2003. This covers almost a quarter of a century up to the most recent past. However, a drawback is that only a maximum of eight observations is available for each country. Certain countries, mainly transition countries and some developing countries, were included only in the ratings from the late 1980s or early 1990s.

The country group consists of no fewer than 87 countries (Appendix C). Whereas almost all previous cross-country labor market studies only covered industrial countries, this paper also includes a large number of developing and transition countries. The country sample thus is unusually large for a labor market study. However, three qualifications should be noted in this respect. First, as mentioned previously, unemployment data are available only from 1980.

Second, the EFW index even covers 127 countries, 40 of which could not be taken into account, as no labor market data are available for these primarily African countries. Third, for many of the developing and transition countries included in the sample, unemployment data are even more scant than EFW data, particularly for the 1980s. This is the main reason why the number of observations is always markedly below 696 (87 countries, eight years) in our main regressions (section 2).

To measure the effects on unemployment, we use not only the overall unemployment rate but also unemployment rates relating to female and young workers (for definitions and sources, see Appendix A). This enables us to determine not only to what extent economic freedom affects unemployment among the total labor force but also to what extent it affects two important demographic groups that often have above-average unemployment rates.

Almost all the unemployment data used in this paper come from the latest edition of the International Labour Office's (ILO's) Key Indicators of the Labour Market (International Labour Office 2005). All variables are exclusively based on labor force survey data. Thus, the data do not refer to registered unemployment. Instead, they are based on an international standard that defines the unemployed as all persons above a specific age who, during the reference period, were without work, currently available for work, and seeking work. Although national coverage of unemployment can vary with regard to factors such as age limits and criteria for seeking work, in the latest edition of its Key Indicators of the Labour Market, the ILO has undertaken great efforts to produce series that are comparable across countries. With regard to age limits, for example, most national series presented in this publication refer to the age-group 15 years and older. Furthermore, for the latest edition of its Key Indicators of the Labour Market, the ILO has "cleaned" the national time series to eliminate breaks in series. Thus, the data are comparable over time. Although the ILO's unemployment data are not completely harmonized across countries, they are comparable to the greatest extent currently possible.

To control for business cycle conditions, we normalized each country's GDP growth rate for its average growth rate over the previous 10 years. (2) In addition, we use year dummies to control for year-specific effects. We also control for the percentage share of children in the population. This share varies widely across countries, especially between developing and industrial countries. Large variations in this share are likely to affect the labor market.

Furthermore, we use two variables to control for the impact of geographical conditions. In a series of papers, Sachs and coauthors have demonstrated that levels and growth rates of GDP per capita are strongly correlated with key geographical variables (e.g., Gallup, Sachs, and Mellinger 1999; Sachs 2001). Our first geographical control is the share of land area in geographical tropics. Sachs and coauthors argue that tropical climates hinder production and development. One may thus hypothesize that they may also increase unemployment. Our second geographical control is the mean distance to the nearest ice-free coastline. A long distance is likely to increase transport costs for international trade, thus possibly increasing unemployment.

We also control for ethnic fractionalization. In ethnically heterogeneous societies, the group that comes to power may implement policies that lower output and increase unemployment (e.g., policies that prohibit the growth of industries dominated by the losing groups). Indeed, previous studies have found ethnic fractionalization to be inversely related to GDP per capita growth (e.g., Easterly and Levine 1997; Alesina et al. 2003). Finally, we employ a control variable for interstate and internal wars because they may severely disrupt the labor markets of the countries in which they take place.

In our main regressions, the coefficients are estimated using the random effects, feasible generalized least squares (FGLS) procedure that incorporates time-invariant country effects (Swamy-Arora method). This enables us to exploit both the cross-country and the time-series variation included in the sample while simultaneously controlling for unobserved country effects. (3) Allowing for cross-country differences in unemployment that reflect the influence of omitted variables is highly desirable, but the random effects method for doing so produces biased estimates if variables included as controls are correlated with country-specific error terms. Therefore, a Hausman test for misspecification of the random effects model is shown for each regression. As the results from this test indicate, none of our estimates is biased (Tables 1-3). Thus, in our case the random effects FGLS method is the appropriate choice. Finally, to correct for heteroskedasticity, we estimate robust t-statistics using the technique developed by White.

In addition to the control variables used, it would have been desirable to explicitly control for the impact of important labor market institutions (e.g., wage-setting systems, dismissal protection laws, wage replacement benefits). However, such data are not available for a large number of countries and years. To date, the Organization for Economic Cooperation and Development (OECD) has undertaken the most extensive effort to quantify labor market institutions and has developed a set of indicators.

However, the OECD data cover only some of their member countries and only a relatively short period of time. Applying these data would have reduced the number of observations to approximately 70. There are also only few data for the EFW component labor market regulations and for the indicators on which this component is based. Up to now, the data were published only for comparatively few countries and years in the reports of the Fraser Institute. (4) Although the following regressions thus do not explicitly control for the effects of labor market institutions, both the regulation variable and the country-specific random effects implicitly control for them, at least to some degree.

2. Main Results

Tables 1 to 3 present our main results. (5) The estimates for the economic freedom variable (measuring the impact of economic freedom in general) indicate that more economic freedom is likely to reduce unemployment both among women and young people (Tables 2 and 3). By contrast, the economic freedom variable is not statistically significant in the regression to explain the overall unemployment rate (Table 1).

The effects of economic freedom on female and youth unemployment appear to be substantial. Take the United States and Italy as examples. The United States achieved one of the best results. On average over the eight years under review, its EFW summary rating was 8.1. Italy's rating, at 6.5, was noticeably lower. If Italy had enjoyed the same extent of economic freedom as the United States, its unemployment rate would have been 2.4 percentage points lower among women and 4.2 percentage points lower among youths, ceteris paribus. Of course, these figures (as the ones in the following paragraphs) should be interpreted with some caution. However, they illustrate that restrictions on economic freedom are likely to involve substantial costs in terms of higher unemployment.

The estimates for the area variables (measuring the impact of specific aspects of economic freedom) provide differentiated insights into the importance of individual economic freedoms to the labor market. According to our estimates for the size of government variable, a smaller government sector is likely to lower unemployment among the total labor force as well as among female and young workers (Tables 1-3). The coefficients are always statistically highly significant, both in the specifications that include this variable separately and in the specifications that include it jointly with the other four area variables.

Once again, the effects are likely to be substantial. For example, if Italy's government sector (rated 4.0) had been as small as in the United States (rated 6.9), its unemployment rate would have been 2.3 percentage points lower among the total labor force, 3.1 percentage points lower among women, and 4.4 percentage points lower among young people, ceteris paribus. (6) Our results for the size of government variable are in line with numerous theoretical studies that argue that a larger government sector increases unemployment via various channels. Most important, a larger government sector crowds out the private sector, reducing the international competitiveness of the relevant economy (Alesina and Perotti 1997). In addition, high taxes, which are an unavoidable implication of a large government sector, reduce the profitability of private investment (Alesina et al. 2002). Our results are also in line with previous empirical studies on the unemployment effects of government size (Abrams 1999; Christopoulos and Tsionas 2002; Feldmann 2006).

The results for the legal structure and property rights variable are statistically significant in each case too, indicating that a stronger rule of law (i.e., a more independent judiciary, a better protection of property rights, and so on) lowers unemployment (Tables 1-3). The structural features of the legal system appear to have a noticeable impact on the labor market as well. For example, if Italy (rated 6.8 in this area) had enjoyed the same strength in the rule of law as the United States (rated 8.5), its unemployment rate would have been 1.4 percentage points lower among the total labor force, 1.9 percentage points lower among women, and 1.7 percentage points lower among youths, ceteris paribus.

This insight is novel. The labor market effects of the structural features of the legal system have not been previously analyzed. However, there are strong theoretical arguments supporting our finding. In a legal system characterized by the rule of law, labor, like other resources, tends to be allocated efficiently so that low unemployment is achieved. Under the rule of law, people have a strong incentive to be gainfully employed, either self-employed or as employees, because the income they earn in their employment is legally secured and in case of a judicial dispute they can expect to push through their claims in court at relatively low cost. Correspondingly, under the rule of law, enterprises have a strong incentive to hire staff because the proceeds resulting from the employment of workers are legally secured and enterprises can expect to be able to push through their claims in court at relatively low cost in case of a judicial dispute. Furthermore, as positive externalities of innovations are being internalized when property rights are secured, economic progress under such a legal regime is constantly driven forward. Thus, relatively low unemployment may be anticipated in the long term as well. Conversely, a lacking rule of law undermines both people's incentive to take up gainful employment and enterprises' incentive to hire workers and innovate, resulting in relatively high unemployment.

We do not find any statistically significant effects of the sound money and freedom to trade internationally variables (Tables 1-3). In the light of the previous literature, this is not surprising. For example, inflation may exert both an adverse sand effect and a favorable grease effect on unemployment. Previous empirical studies have found evidence for both (e.g., Akerlof, Dickens, and Perry 1996, 2000; Groshen and Schweitzer 1999; Wyplosz 2001). As far as freedom to trade across national boundaries is concerned, although both a liberalization of foreign trade and a liberalization of international capital movements tend to improve allocative efficiency and thus reduce unemployment in the long term, they may increase unemployment in the short term. For example, if some industries have previously been strongly protected from foreign competition, a reduction of tariffs and regulatory trade barriers may lead to a substantial loss of jobs in these industries. A liberalization of international capital movements may increase unemployment in capital-exporting industrial countries in the short term if jobs are relocated to developing countries. In any case, most empirical studies find the unemployment effects of both foreign trade and international capital movements to be small. (7) This is in line with our statistically insignificant estimate for the freedom to trade internationally variable.

The regulation variable is statistically significant in a single regression only. According to this regression, less strict regulation is associated with lower unemployment among young people (Table 3). However, this result is not robust to the inclusion of the other four area variables. Similar to the liberalization of international economic relations, a comprehensive liberalization of national credit, labor, and product markets may have an adverse impact on unemployment in the short term and lower unemployment only in the long term (Blanchard and Giavazzi 2003). Initially, enterprises may reduce excess staff, causing unemployment to rise. Unemployment may fall only gradually thereafter as lower market entry barriers encourage the entry of new firms and as workers' reduced bargaining power leads to a reduction in real wages. Indeed, according to previous empirical studies, an extensive deregulation of individual industries may involve a substantial reduction in the number of jobs in the respective industries (e.g., Davis and Wilson 1999). However, many empirical studies also find less strict product and labor market regulation to be generally associated with lower unemployment rates (e.g., Feldmann 2003; International Monetary Fund 2003; Nickell, Nunziata, and Ochel 2005). This is in line with our statistically significant estimate for the regulation variable.

3. Supplementary Results

As argued in the previous section, the impact of an increase in some economic freedoms may be adverse in the short term. It may take several years for the beneficial long-term effects to prevail. Even if there are no adverse short-term effects at all, an increase in economic freedom may lower unemployment only with a time lag of several years. For example, employers may be reluctant to hire additional staff until they can be reasonably sure that a recent strengthening of the rule of law will not be reversed. In this section, we explore the long-term impact of economic freedom by analyzing its effects on long-term changes in unemployment rates. We define a change in the unemployment rate as the average unemployment rate over 2000-2003 minus the average unemployment rate over 1980 and 1985.

In our supplementary regressions, we use each of the six EFW variables (the summary variable and the five area variables) in two different forms: (i) change in the respective rating, also defined as average over 2000-2003 minus average over 1980 and 1985, and (ii) average rating over 1980 and 1985. The long-term effects of economic freedom should be reflected primarily in the coefficients on the change variables of economic freedom. If economic freedom makes a difference, countries with substantially higher EFW ratings in the latter part of the period than in the earlier years should have lower unemployment rates during the recent years than during the earlier ones. However, long-term effects may also be reflected in the average rating over 1980 and 1985 variables of economic freedom. A high (low) level of economic freedom during these years may have triggered a decrease (an increase) in unemployment in the ensuing years.

In our supplementary regressions, we use the same controls as in our main regressions. Except where controls are time invariant (e.g., distance to coastline), they are also defined as change variables (in the same manner as the dependent variables and the variables of interest). Thus, change in GDP growth gap, for example, reflects the difference between both periods in business cycle conditions. We use one additional control: the level of the respective unemployment rate during the earlier years. Countries with high initial unemployment rates are more likely to experience large drops in unemployment than countries starting at low levels of unemployment.

Four caveats should be noted. First, as data on economic freedom and particularly on unemployment are lacking for many developing and most transition countries for the 1980s, the number of countries is much smaller than in our main regressions. Second, because of the small number of observations in our supplementary regressions, we have to refrain from estimating the area variables jointly. We can estimate them separately only. Third, as there is just one observation per country in our supplementary regressions, we are unable to control for unobserved country effects. Fourth, because of a lack of adequate data (section 1), we are unable to control for the effects of changes in labor market institutions. However, some of these effects should be reflected in the coefficients on the change in regulation variable. These four drawbacks are the main reason why we regard the regression results presented in this section as supplementary.

According to our results, the coefficients on change in economic freedom are strongly negative and statistically significant in the regressions to explain both change in the female and change in the youth unemployment rate (Tables 5 and 6). Thus, an increase in economic freedom from 1980 and 1985 to 2000-2003 is associated with a substantial fall in unemployment for these demographic groups over the same period. Furthermore, a higher level of economic freedom in 1980 and 1985 is correlated with a decline in unemployment over the period to 2000-2003. The latter result is statistically significant not only for female and young workers but for the total labor force as well (Tables 4-6).

Most of the results for the two size of government variables, either in their change or in their initial level form, are statistically insignificant. According to the sole significant result, a smaller government sector in 1980 and 1985 is correlated with a fall in youth unemployment over the ensuing years (Table 6). By contrast, most of the results for the two legal structure and property rights variables are statistically significant. Specifically, a strengthening of the rule of law, as reflected in an increase in the respective EFW area rating, is associated with a fall in unemployment both among the total labor force and among women (Tables 4 and 5). Furthermore, a stronger rule of law in 1980 and 1985 is associated with a fall in unemployment over the period to 2000-2003. This result is statistically significant for the total labor force as well as for women and young people (Tables 4-6).

Whereas none of the results for the two sound money variables are statistically significant, most of the results for the freedom to trade internationally and the regulation variables are. According to these results, a more comprehensive liberalization of international trade and capital movements from the first half of the 1980s to the beginning of the current decade is associated with a fall in unemployment among female and young workers over the same period (Tables 5 and 6). Furthermore, a more liberal regime for international exchange in 1980 and 1985 is associated with a decline in unemployment over the period to 2000-2003 both among the total labor force and among women and youths (Tables 4-6). Similarly, a more comprehensive deregulation of credit, labor, and product markets is correlated with a fall in unemployment among the total labor force as well as among women (Tables 4 and 5). Finally, more flexible regulations in 1980 and 1985 are associated with a fall in unemployment among the total labor force as well as among women and youths (Tables 4-6).

Overall, the results from our supplementary regressions indicate that economic freedom is likely to have a favorable long-term impact on unemployment, particularly among women and young people. The most important economic freedoms likely to exert beneficial unemployment effects in the long term are a strong rule of law, freedom to trade internationally, and flexible regulations governing credit, labor, and product markets.

4. Conclusion

This paper is the first to provide insights into the effects of economic freedom on unemployment. According to our regression results, economic freedom is likely to substantially reduce unemployment, especially among women and young people. A small government sector and a legal system characterized by an independent judiciary, impartial courts, and an effective protection of property rights clearly seem to have beneficial effects. In addition, there are indications that freedom to trade across national boundaries and a light regulatory burden may also lower unemployment, though apparently in the long term only.

Appendix A: Definitions and Sources of Variables (8)

Distance to coastline: Mean distance to nearest ice-free coastline, measured in thousands of kilometers. Source: Center for International Development (2001); author's calculations.

Economic freedom: Summary index of the Economic Freedom of the World (EFW) index, consisting of the following five indicators: size of government, legal structure and security of property rights, access to sound money, freedom to trade internationally, and regulation of credit, labor, and business. See definitions of these indicators. All indicators carry equal weights. The scale of the EFW index ranges from 0 to 10, with 0 representing the lowest and 10 the highest degree of economic freedom. Source: Gwartney and Lawson (2005).

Ethnic fractionalization: One minus the Herfindahl index of ethnic group shares, reflecting the probability that two randomly selected individuals from a population belong to different groups. The definition of ethnicity involves a combination of racial and linguistic characteristics. The classifications reflect the judgments of ethnologists and anthropologists on the appropriate definition of ethnicity. Source: Alesina et al. (2003).

Female unemployment rate: Unemployed women as a percentage of the female labor force. Labor force survey data. Source: International Labour Office (2005).

Freedom to trade internationally: Subindex of the EFW index, consisting of the following five indicators: taxes on international trade, regulatory trade barriers, actual size of trade sector compared to expected size, difference between official and black market exchange rate, and international capital market controls. The ratings for the indicator taxes on international trade are based on the following three equally weighted subindicators: revenue from taxes on international trade as a percentage of exports plus imports, mean tariff rate, and standard deviation of tariff rates. The ratings for the indicator regulatory, trade barriers are based on two equally weighted survey questions from the World Economic Forum's annual Executive Opinion Surveys. The first statement says, "In your country, hidden import barriers (that is barriers other than published tariffs and quotas) are an important problem (lower end of the scale) or not an important problem (higher end of the scale)." The second statement says, "The combined effect of import tariffs, license fees. bank fees, and the time required for administrative red-tape raises costs of importing foreign equipment by 10% or less (higher end of the scale) or by more than 50% (lower end of the scale)." For the indicator actual size of trade sector compared to expected size, regression analysis is used to derive an expected size based on the population and geographic size of the country and its location relative to the concentration of world GDP. For the indicator difference between official and black market exchange rate, a formula is used that allocates a rating of 10 to countries with a domestic currency that is fully convertible without restrictions and that allocates ratings that decline toward zero when exchange rate controls are present and the black market premium increases toward 50%. The ratings for the indicator international capital market controls are based on two equally weighted sources of data. First, they are based on answers to the following question from the Executive Opinion Surveys: "Foreign ownership of companies in your country is rare, limited to few cases, and prohibited in key sectors (lower end of the scale) or prevalent and encouraged (higher end of the scale)." Second, they are based on an index of capital controls that is derived from periodical IMF reports. All five indicators carry equal weights. The rating scale ranges from 0 to 10, with higher values indicating more freedom to engage in international exchange. Source: Gwartney and Lawson (2005).

GDP growth gap: Annual percentage growth rate of real GDP during the current year minus average annual percentage growth rate of real GDP during the previous 10 years. Source: World Bank (2006); author's calculations.

Legal structure and property rights: Subindex of the EFW index, consisting of the following five indicators: judicial independence, impartial courts, protection of intellectual property, military interference in rule of law and the political process, and integrity of the legal system. The ratings for the first three indicators are based on results from the World Economic Forum's annual Executive Opinion Surveys. With regard to judicial independence, the survey statement says, "The judiciary is independent and not subject to interference by the government or parties in disputes." With regard to impartial courts, the statement says, "A trusted legal framework exists for private businesses to challenge the legality of government actions or regulation." With regard to protection of intellectual property, the statement says, "Intellectual property protection in your country is either weak or nonexistent (lower end of the scale) or equal to the world's most stringent (higher end of the scale)." The ratings for the indicators military interference in rule of law and the political process and integrity of the legal system are calculated using data from the PRS Group's International Country Risk Guides. These data are computed from an in-house panel of experts of the PRS Group. All five indicators carry equal weights. The rating scale ranges from 0 to 10, with higher values indicating a stronger rule of law (i.e., a more independent judiciary, a better protection of property rights, and so on). Source: Gwartney and Lawson (2005).

Population aged 0-14: The percentage of the total population that is in the age-group 0-14 years. Source: Directorate General of Budget, Accounting, and Statistics (2005); World Bank (2006).

Regulation: Subindex of the EFW index, consisting of the following three indicators: credit market regulations, labor market regulations, and business regulations. The ratings for the first indicator are based on the following five subindicators: percentage of deposits held in privately owned banks, competition from foreign banks (based on the denial rate of foreign bank license applications and the share of foreign bank assets in total banking sector assets), percentage of credit extended to the private sector, avoidance of interest rate controls and regulations that lead to negative real interest rates, and interest rate controls. While the first four subindicators are based on objective data, the fifth subindicator is based on replies to the following question from the World Economic Forum's annual Executive Opinion Survey: "Interest rates on bank deposits and/or loans are freely determined by the market." The ratings for the second indicator are based on five subindicators. The first four subindicators come from the Executive Opinion Surveys. The four survey statements are "The minimum wage, set by law, has little impact on wages because it is too low or not obeyed"; "Hiring and firing practices of companies are determined by private contract"; "The share of labor force whose wages are set by centralized collective bargaining is low"; and "The unemployment benefits system preserves the incentives to work." The fifth subindicator is based on objective data; it measures the use of conscripts to obtain military personnel, including the duration of military conscription. The ratings for the third indicator are also based on five subindicators. The first subindicator, price controls, measures the extent to which businesses are free to set their own prices. The other four subindicators come from the Executive Opinion Surveys. The four survey statements are "Administrative procedures are an important obstacle to starting a new business"; "Senior management spends a substantial amount of time dealing with government bureaucracy"; "Starting a new business is generally easy"; and "Irregular, additional payments connected with import and export permits, business licenses, exchange controls, tax assessments, police protection, or loan applications are very rare." All subindicators are averaged to calculate the ratings for the three indicators. In turn, the subindex regulation is the average of the ratings for the three indicators. The rating scale ranges from 0 to 10, with higher values representing less strict regulation. Source: Gwartney and Lawson (2005).

Size of government: Subindex of the EFW index, consisting of the following four indicators: general government consumption spending as a percentage of total consumption, transfers and subsidies as a percentage of GDP, role of state-owned enterprises in the economy and government investment as a percentage of total investment, and top marginal income and payroll tax rates (and income threshold at which the top marginal income tax rate applies). All indicators carry equal weights. The rating scale ranges from 0 to 10, with higher values indicating a smaller government sector. Source: Gwartney and Lawson (2005).

Sound money: Subindex of the EFW index, consisting of the following four indicators: average annual growth rate of the money supply (M1) in the last five years minus average annual growth of real GDP in the last 10 years, standard inflation variability in the last five years (based on the standard deviation of the annual GDP deflator), recent annual consumer price inflation rate, and freedom of citizens to own foreign currency bank accounts domestically and abroad. All indicators carry equal weights. The rating scale ranges from 0 to 10, with higher values indicating a higher quality of money (no excess money growth, low and relatively stable inflation rates, and currency convertibility). Source: Gwartney and Lawson (2005).

Tropical area: Share of land area in geographical tropics. Source: Center for International Development (1999, 2001).

Unemployment rate: Unemployed as a percentage of the labor force. Labor force survey data. Source: International Labour Office (2005).

War: Dummy variable for interstate and internal wars in the respective country. Source: Centre for the Study of Civil War (2005).

Youth unemployment rate: Unemployed aged 15-24 years as a percentage of the labor force in the same age bracket. Labor force survey data. Source: European Commission (2005); International Labour Office (2005); Organization for Economic Cooperation and Development (2005, 2006).

Appendix B: Correlation Matrix

                                                Legal
                                              System and
                      Economic    Size of      Property    Sound
                      Freedom    Government     Rights     Money

Economic freedom        1.00
Size of government      0.39        1.00
Legal system and
  property rights       0.71       -0.15         1.00
Sound money             0.84        0.21         0.48      1.00
Freedom to trade
  internationally       0.80        0.14         0.58      0.58
Regulation              0.78        0.33         0.53      0.53
GDP growth gap         -0.06       -0.06        -0.05      -0.06
Population aged
  0-14                 -0.59        0.09        -0.70      -0.43
Tropical area          -0.29        0.32        -0.55      -0.26
Distance to
  coastline            -0.28       -0.16        -0.22      -0.23
Ethnic
  fractionalization    -0.34        0.04        -0.42      -0.29
War                    -0.17        0.05        -0.19      -0.16

                        Freedom to                    GDP
                           Trade                     Growth
                      Internationally   Regulation    Gap

Economic freedom
Size of government
Legal system and
  property rights
Sound money
Freedom to trade
  internationally           1.00
Regulation                  0.55           1.00
GDP growth gap              0.03          -0.03       1.00
Population aged
  0-14                     -0.59          -0.33      -0.03
Tropical area              -0.32          -0.08       0.02
Distance to
  coastline                -0.15          -0.29       0.08
Ethnic
  fractionalization        -0.24          -0.19       0.07
War                        -0.15          -0.10       0.02

                      Population   Tropical   Distance to
                      Aged 0-14      Area      Coastline

Economic freedom
Size of government
Legal system and
  property rights
Sound money
Freedom to trade
  internationally
Regulation
GDP growth gap
Population aged
  0-14                   1.00
Tropical area            0.70        1.00
Distance to
  coastline              0.24        0.08        1.00
Ethnic
  fractionalization      0.60        0.55        0.32
War                      0.10        0.12        0.10

                           Ethnic
                      Fractionalization   War

Economic freedom
Size of government
Legal system and
  property rights
Sound money
Freedom to trade
  internationally
Regulation
GDP growth gap
Population aged
  0-14
Tropical area
Distance to
  coastline
Ethnic
  fractionalization         1.00
War                         0.08          1.00

Appendix C: List of Countries

Albania, Algeria, Argentina, Australia, Austria, Bahamas, Bangladesh, Barbados, Belgium, Bolivia, Botswana, Brazil, Bulgaria, Canada, Chile, China, Colombia, Costa Rica, Croatia, Cyprus, Czech Republic, Denmark, Dominican Republic, Ecuador, Egypt, El Salvador, Estonia, Finland, France, Germany, Greece, Guatemala, Honduras, Hong Kong, Hungary, Iceland, India, Indonesia, Ireland, Israel, Italy, Jamaica, Japan, Jordan, Latvia, Lithuania, Luxembourg, Malaysia, Malta, Mauritius, Mexico, Morocco, Namibia, Netherlands, New Zealand, Nicaragua, Norway, Pakistan, Panama, Paraguay, Peru, Philippines, Poland, Portugal, Romania, Russia, Singapore, Slovak Republic, Slovenia, South Africa, South Korea, Spain, Sri Lanka, Sweden, Switzerland, Syria, Taiwan, Thailand, Trinidad and Tobago, Turkey, Ukraine, United Arab Emirates, United Kingdom, United States, Uruguay, Venezuela, Zambia.

Received June 2005; accepted July 2006.

References

Abrams, Burton A. 1999. The effect of government size on the unemployment rate. Public Choice 99:395-401.

Akerlof, George A., William T. Dickens, and George L. Perry. 1996. The macroeconomics of low inflation. Brookings Papers on Economic Activity 1:1-59.

Akerlof, George A., William T. Dickens, and George L. Perry. 2000. Near-rational wage and price setting and the long-run Phillips curve. Brookings Papers on Economic Activity 1:1-44.

Alesina, Alberto, Silvia Ardagna, Roberto Perotti, and Fabio Schiantarelli. 2002. Fiscal policy, profits, and investment. American Economic Review 92:571-89.

Alesina, Alberto, Arnaud Devleeschauwer, William Easterly, Sergio Kurlat, and Romain Wacziarg. 2003. Fractionalization. Journal of Economic Growth 8:155-94.

Alesina, Alberto, and Roberto Perotti. 1997. The welfare state and competitiveness. American Economic Review 87:921-39.

Blanchard, Olivier, and Francesco Giavazzi. 2003. Macroeconomic effects of regulation and deregulation in goods and labor markets. Quarterly Journal of Economics 118:879-907.

Center for International Development. 1999. geodata.xls (Excel file). Cambridge, MA: Center for International Development at Harvard University.

Center for International Development. 2001. physfact_rev.xls (Excel file). Cambridge, MA: Center for International Development at Harvard University.

Centre for the Study of Civil War. 2005. Armed Conflict Dataset, Version 3-2005b. Accessed 5 March 2006. Available http://www.prio.no.

Christopoulos, Dimitris K., and Efthymios G. Tsionas. 2002. Unemployment and government size: Is there any credible causality? Applied Economics Letters 9:797-800.

Davis, David E., and Wesley W. Wilson. 1999. Deregulation, mergers, and employment in the railroad industry. Journal of Regulatory Economics 15:5-22.

De Haan, Jakob, Susanna Lundstrom, and Jan-Egbert Sturm. 2006. Market oriented institutions and policies and economic growth: A critical survey. Journal of Economic Surveys 20:157-91.

Directorate-General of Budget, Accounting, and Statistics. 2005. Statistical yearbook of the Republic of China 2004. Taipei: Directorate-General of Budget, Accounting, and Statistics.

Easterly, William, and Ross Levine. 1997. Africa's growth tragedy: Policies and ethnic divisions. Quarterly Journal of Economics 112:1203-50.

European Commission. 2005. Employment in Europe 2005: Recent trends and prospects. Luxembourg: Office for Official Publications of the European Communities.

Feenstra, Robert C., Gordon H. Hanson, and Deborah L. Swenson. 2000. Offshore assembly from the United States: Production characteristics of the 9802 program. In The impact of international trade on wages, edited by Robert C.

Feenstra. Chicago: University of Chicago Press, pp. 85-122.

Feldmann, Horst. 2003. Labor market regulation and labor market performance: Evidence based on surveys among senior business executives. Kyklos 56:509-39.

Feldmann, Horst. 2006. Government size and unemployment: Evidence from industrial countries. Public Choice 127:451-67.

Gallup, John Luke, Jeffrey D. Sachs, and Andrew Mellinger. 1999. Geography and economic development. CID Working Paper No. 1. Cambridge, MA: Center for International Development at Harvard University.

Groshen, Erica L., and Mark E. Schweitzer. 1999. Identifying inflation's grease and sand effects in the labor market. In The costs and benefits of price stability, edited by Martin Feldstein. Chicago: University of Chicago Press, pp. 273-308.

Gwartney, James D., and Robert A. Lawson. 2003. The concept and measurement of economic freedom. European Journal of Political Economy 19:405-30.

Gwartney, James D., and Robert A. Lawson. 2005. Economic freedom of the world: 2005 annual report. Vancouver: Fraser Institute.

International Labour Office. 2005. Key indicators of the labour market. 4th edition. Geneva: ILO.

International Monetary Fund. 2003. World economic outlook. April. Washington, DC: International Monetary Fund.

Kravis, Irving B., and Robert E. Lipsey. 1988. The effect of multinational firms' foreign operations on their domestic employment. NBER Working Paper No. 2760.

Lipsey, Robert E. 1999. Foreign production by US firms and parent firm employment. NBER Working Paper No. 7357.

Miles, Marc A., Kim R. Holmes, and Mary Anastasia O'Grady. 2006. 2006 index of economic freedom. Washington, DC: Heritage Foundation.

Neven, Damien, and Charles Wyplosz. 1999. Relative prices, trade and restructuring in European industry. In Trade and jobs in Europe: Much ado about nothing?, edited by Mathias Dewatripont, Andre Sapir, and Khalid Sekkat. Oxford: Oxford University Press, pp. 33-59.

Nickell, Stephen, Luca Nunziata, and Wolfgang Ochel. 2005. Unemployment in the OECD since the 1960s. What do we know? Economic Journal 115:1-27.

Organization for Economic Cooperation and Development. 2005. Employment outlook. Paris: Organization for Economic Cooperation and Development.

Organization for Economic Cooperation and Development. 2006. Labour Market Statistics--Indicators. Accessed 11 March 2006. Available http://www.oecd.org.

Pryor, Frederic L. 1999. The impact of foreign trade on the employment of unskilled U.S. workers: Some new evidence. Southern Economic Journal 65:472-92.

Sachs, Jeffrey D. 2001. Tropical underdevelopment. NBER Working Paper No. 8119.

Thorbecke, Willem, and Christian Eigen-Zucchi. 2002. Did NAFTA cause a "giant sucking sound"? Journal of Labor Research 23:647-58.

World Bank. 2006. WDI Online. Accessed 7 and 22 March 2006. Available http://www.worldbank.org.

Wyplosz. Charles. 2001. Do we know how low should inflation be? In Why price stability?, edited by Alicia Garcia Herrero, Vitor Gaspar, Lex Hoogduin, Julian Morgan, and Bernhard Winkler. Frankfurt am Main: European Central Bank, pp. 15-33.

(1) The Heritage Foundation has also published an index of economic freedom since 1995 (e.g., Miles, Holmes, and O'Grady 2006). This index is solely based on the assessments of in-house experts. Thus, it is far less objective and transparent than the EFW index. The latter is clearly superior from a methodological point of view.

(2) The output gap would have been the best indicator to control for the impact of business cycle fluctuations. However, data on this variable are available for industrial countries only.

(3) In our case, the fixed effects model is not a feasible alternative to control for unobserved country effects. As most of the variability in the EFW ratings comes from differences across countries rather than from changes in rating scores through time, fixed effects estimates would be very imprecise. They only use the time-series variation within countries. In any case, the fixed effects coefficient estimates (not reported here) are similar to the random effects coefficient estimates presented in this paper.

(4) There is another reason why the number of observations would have been very small when using either the OECD or the EFW data on labor market institutions. In order to completely cover the area regulation, the variables credit market regulations and business regulations would have been needed in addition to the data for labor market institutions. However, up to now, the data for these two components of the EFW index have also been published only for comparatively few countries and years in the reports of the Fraser Institute.

(5) To diminish the impact of outliers, we use the natural logarithms of the dependent variables.

(6) The numerical illustrations involving area variables are based on the regressions that estimate their coefficients jointly. These regressions are probably more reliable, as they explain a larger percentage of the variation of the dependent variables than the ones that estimate the area variables separately (Tables 1-3) and as these variables are not highly correlated with each other (Appendix B).

(7) Empirical studies on foreign trade include Neven and Wyplosz (1999), Pryor (1999), and Thorbecke and Eigen-Zucchi (2002). Empirical studies on international capital movements include Kravis and Lipsey (1988), Lipsey (1999), and Feenstra, Hanson, and Swenson (2000).

(8) The variables measuring changes through time, indicated by the prefix Change in, are defined as average over 2000 2003 minus average over 1980 and 1985. The variables measuring averages over 1980 and 1985, indicated by the postfix average over 1980 and 1985, are defined as indicated.

Horst Feldmann, Department of Economics and International Development, University of Bath, Bath BA2 7AY, United Kingdom; E-mail [e-mail].

Table 1. Regressions to Explain the Log Unemployment Rate *

                                      (1)          (2)          (3)

Economic freedom                    -0.06
                                   (-1.38)
Size of government                               -0.07 ***
                                                (-3.85)
Legal structure and property                                  -0.08 ***
  rights                                                     (-3.39)
Sound money

Freedom to trade internationally

Regulation

GDP growth gap                       0.00         0.00         0.00
                                    (0.19)       (0.37)       (0.03)
Population aged 0-14                 0.01 *       0.01 **      0.01
                                    (1.89)       (2.15)       (1.38)
Tropical area                       -0.24 **     -0.11        -0.36 ***
                                   (-2.23)      (-1.25)      (-4.05)
Distance to coastline                0.02         0.03        -0.01
                                    (0.15)       (0.20)      (-0.06)
Ethnic fractionalization             0.49 *       0.50 *       0.46 *
                                    (1.84)       (1.91)       (1.68)
War                                  0.04         0.06         0.02
                                    (0.34)       (0.53)       (0.17)
No. of observations                508          508          506
No. of countries                    87           87           87
[R.sup.2]                            0.17         0.18         0.19
Standard error of regression         0.33         0.33         0.32
F-statistic                          7.39 ***     7.87 ***     8.30 ***
Hausman test                        15.59        11.67        8.17

                                      (4)          (5)

Economic freedom

Size of government

Legal structure and property
  rights

Sound money                          0.01
                                    (1.03)
Freedom to trade internationally                 -0.01
                                                (-0.11)
Regulation

GDP growth gap                       0.01         0.00
                                    (0.91)       (0.22)
Population aged 0-14                 0.01 **      0.01 **
                                    (1.99)       (2.15)
Tropical area                       -0.24 **     -0.26 **
                                   (-2.20)      (-2.09)
Distance to coastline                0.04         0.04
                                    (0.29)       (0.24)
Ethnic fractionalization             0.46 *       0.53 ***
                                    (1.78)       (2.66)
War                                  0.10         0.07
                                    (1.10)       (0.68)
No. of observations                509          507
No. of countries                    87           87
[R.sup.2]                            0.17         0.17
Standard error of regression         0.35         0.33
F-statistic                          7.20 ***     7.15 ***
Hausman test                        16.79       15.73

                                      (6)          (7)

Economic freedom

Size of government                               -0.09 ***
                                                (-4.86)
Legal structure and property                     -0.09 ***
  rights                                        (-3.64)
Sound money                                       0.01
                                                 (1.02)
Freedom to trade internationally                  0.03
                                                 (0.59)
Regulation                           0.08         0.06
                                    (1.45)       (1.55)
GDP growth gap                       0.01         0.00
                                    (1.21)       (0.61)
Population aged 0-14                 0.01*        0.01 **
                                    (1.86)       (2.53)
Tropical area                       -0.24 *      -0.23 ***
                                   (-1.81)      (-2.69)
Distance to coastline                0.09         0.00
                                    (0.73)       (0.00)
Ethnic fractionalization             0.44 *       0.46 **
                                    (1.77)       (2.46)
War                                  0.12         0.07
                                    (1.18)       (0.74)
No. of observations                506          502
No. of countries                    87           87
[R.sup.2]                            0.18         0.24
Standard error of regression         0.35         0.32
F-statistic                          7.90 ***     8.25 ***
Hausman test                        13.33        21.62

(a) Feasible generalized least squares estimates with country-specific
random effects (Swamy-Arora method). All regressions
are based on data for the years 1980, 1985, 1990, 1995, and
2000-2003. Heteroskedasticity-consistent t-statistics in
parentheses (White method). All regressions also contain year
dummies and a constant term.

* Statistically significant at 10%.

** Statistically significant at 5%.

*** Statistically significant at 1%.

Table 2. Regressions to Explain the Log Female Unemployment Rate (a)

                                      (1)         (2)         (3)

Economic freedom                    -0.11 **
                                   (-2.25)
Size of government                              -0.08 ***
                                               (-3.08)
Legal structure and property                                -0.08 ***
  rights                                                   (-3.10)
Sound money

Freedom to trade internationally

Regulation

GDP growth gap                       0.01        0.01        0.01
                                    (1.24)      (1.44)      (1.14)
Population aged 0-14                 0.01        0.01 **     0.01 *
                                    (1.30)      (2.26)      (1.91)
Tropical area                       -0.24 *     -0.12       -0.40 ***
                                   (-1.71)     (-0.87)     (-3.17)
Distance to coastline               -0.06       -0.02       -0.07
                                   (-0.35)     (-0.12)     (-0.46)
Ethnic fractionalization             0.64 ***    0.65 ***    0.60 ***
                                    (2.84)      (2.74)      (2.66)
War                                  0.14        0.19 **     0.15 *
                                    (1.59)      (2.34)      (1.77)
No. of observations                483         483         482
No. of countries                    85          85          85
[R.sup.2]                            0.17        0.17        0.17
Standard error of regression         0.36        0.36        0.36
F-statistic                          6.80 ***    6.62 ***    7.01 ***
Hausman test                        11.58        9.69        5.87

                                      (4)         (5)

Economic freedom

Size of government

Legal structure and property
  rights
Sound money                          0.01
                                    (0.67)
Freedom to trade internationally               -0.05
                                               (-1.02)
Regulation

GDP growth gap                       0.02        0.01
                                    (1.45)      (1.32)
Population aged 0-14                 0.02 **     0.01
                                    (2.35)      (1.44)
Tropical area                       -0.26       -0.26 *
                                   (-1.50)     (-1.75)
Distance to coastline               -0.03       -0.02
                                   (-0.16)     (-0.12)
Ethnic fractionalization             0.63 **     0.70 ***
                                    (2.09)      (3.90)
War                                  0.23 ***    0.17 **
                                    (3.00)      (2.20)
No. of observations                484         482
No. of countries                    85          85
[R.sup.2]                            0.16        0.16
Standard error of regression         0.38        0.36
F-statistic                          6.23 ***    6.24 ***
Hausman test                        10.26       11.29

                                      (6)         (7)

Economic freedom

Size of government                              -0.08 ***
                                               (-3.15)
Legal structure and property                    -0.09 ***
  rights                                       (-3.07)
Sound money                                      0.02
                                                (1.22)
Freedom to trade internationally                -0.02
                                               (-0.54)
Regulation                           0.02       -0.00
                                    (0.39)     (-0.01)
GDP growth gap                       0.02        0.01
                                    (1.50)      (1.47)
Population aged 0-14                 0.01 **     0.01
                                    (2.41)      (1.60)
Tropical area                       -0.25       -0.22 *
                                   (-1.43)     (-1.72)
Distance to coastline               -0.00       -0.06
                                   (-0.03)     (-0.36)
Ethnic fractionalization             0.60 **     0.60 ***
                                    (1.98)      (3.20)
War                                  0.22 **     0.14 *
                                    (2.47)      (1.81)
No. of observations                482         479
No. of countries                    85          85
[R.sup.2]                            0.16        0.21
Standard error of regression         0.38        0.36
F-statistic                          6.33 ***    6.61 ***
Hausman test                        10.95       14.94

(a) Feasible generalized least squares estimates with country-specific
random effects (Swamy-Arora method). All regressions
are based on data for the years 1980, 1985, 1990, 1995, and
2000-2003. Heteroskedasticity-consistent t-statistics in
parentheses (White method). All regressions also contain year
dummies and a constant term.

* Statistically significant at 10%.

** Statistically significant at 5%.

*** Statistically significant at 1%.

Table 3. Regressions to Explain the Log Youth Unemployment Rate (a)

                                      (1)         (2)         (3)

Economic freedom                    -0.10 ***
                                   (-2.67)
Size of government                              -0.06 ***
                                               (-3.68)
Legal structure and property                                -0.03 *
  rights                                                   (-1.90)
Sound money

Freedom to trade internationally

Regulation

GDP growth gap                       0.01 ***    0.01 ***    0.01 ***
                                    (2.59)      (2.62)      (2.93)
Population aged 0-14                 0.00        0.01        0.01
                                    (0.22)      (0.65)      (0.54)
Tropical area                       -0.16       -0.09       -0.26 **
                                   (-1.26)     (-0.66)     (-2.35)
Distance to coastline               -0.17       -0.15       -0.15
                                   (-1.22)     (-1.09)     (-1.18)
Ethnic fractionalization             0.58        0.61 *      0.57
                                    (1.62)      (1.67)      (1.62)
War                                 -0.03        0.02       -0.01
                                   (-0.31)      (0.26)     (-0.10)
No. of observations                395         395         395
No. of countries                    81          81          81
[R.sup.2]                            0.35        0.34        0.34
Standard error of regression         0.28        0.28        0.28
F-statistic                         14.57 ***   14.06 ***   14.24 ***
Hausman test                        13.87       11.97       14.85

                                      (4)         (5)

Economic freedom

Size of government

Legal structure and property
  rights
Sound money                         (0.02)
                                   (-1.44)
Freedom to trade internationally                -0.04
                                               (-0.70)
Regulation

GDP growth gap                       0.01 **     0.01 ***
                                    (2.35)      (2.90)
Population aged 0-14                 0.01        0.00
                                    (0.56)      (0.35)
Tropical area                       -0.19       -0.18
                                   (-1.51)     (-1.12)
Distance to coastline               -0.14       -0.15
                                   (-0.99)     (-1.04)
Ethnic fractionalization             0.59 *      0.70 **
                                    (1.68)      (2.12)
War                                 -0.02       -0.02
                                   (-0.21)     (-0.18)
No. of observations                395         394
No. of countries                    81          80
[R.sup.2]                            0.33        0.34
Standard error of regression         0.28        0.28
F-statistic                         13.48 ***   14.12 ***
Hausman test                        13.77       13.45

                                      (6)         (7)

Economic freedom

Size of government                              -0.06 ***
                                               (-3.97)
Legal structure and property                    -0.04 *
  rights                                       (-1.79)
Sound money                                     (0.01)
                                               (-0.54)
Freedom to trade internationally                -0.02
                                               (-0.28)
Regulation                          -0.04 *     -0.00
                                   (-1.89)     (-0.02)
GDP growth gap                       0.01 ***    0.01 ***
                                    (2.66)      (2.59)
Population aged 0-14                 0.01        0.00
                                    (0.69)      (0.32)
Tropical area                       -0.20       -0.13
                                   (-1.50)     (-1.06)
Distance to coastline               -0.15       -0.20
                                   (-1.09)     (-1.41)
Ethnic fractionalization             0.60 *      0.67 **
                                    (1.69)      (2.00)
War                                 -0.00       -0.00
                                   (-0.03)      (0.05)
No. of observations                395         394
No. of countries                    81          80
[R.sup.2]                            0.33        0.37
Standard error of regression         0.28        0.28
F-statistic                         13.49 ***   12.37 ***
Hausman test                        11.16       20.36

Feasible generalized least squares estimates with country-specific
random effects (Swamy-Arora method). All regressions
are based on data for the years 1980, 1985, 1990, 1995, and
2000-2003. Heteroskedasticity-consistent t-statistics in
parentheses (White method). All regressions also contain year
dummies and a constant term.

* Statistically significant at 10%.

** Statistically significant at 5%.

*** Statistically significant at 1%.

Table 4. Regressions to Explain Change in the Unemployment Rate (a)

                                    (1)            (2)          (3)

Change in economic freedom        -1.23
                                 (-1.34)
Economic freedom, average         -2.25 ***
  over 1980 and 1985             (-3.34)
Change in size of government                     0.13
                                                (0.19)
Sire of government, average                     -0.05
  over 1980 and 1985                           (-0.10)
Change in legal structure and                                -1.05 *
  property rights                                           (-1.89)
Legal structure and property                                 -1.32 ***
  rights, average over 1980                                 (-3.45)
  and 1985
Change in sound money

Sound money, average over
  1980 and 1985
Change in freedom to trade
  internationally
Freedom to trade
  internationally, average
  over 1980 and 1985
Change in regulation

Regulation, average over
  1980 and 1985
Unemployment rate, average      -0.79 ***     -0.78 ***    -0.79 ***
  over 1980 and 1985            (-10.70)       (-8.59)     (-11.43)
Change in GDP growth gap           0.13         -0.05         0.04
                                  (0.89)       (-0.33)       (0.31)
Change in population aged          0.27          0.04         0.25
  0-14                            (1.50)        (0.19)       (1.46)
Tropical area                      1.48          1.85        -0.52
                                  (1.34)        (1.11)      (-0.38)
Distance to coastline              0.30          0.69        -0.64
                                  (0.21)        (0.44)      (-0.39)
Ethnic fractionalization           1.09          2.08         0.70
                                  (0.44)        (0.80)       (0.25)
Change in war                      2.86          1.99         3.26
                                  (1.07)        (0.55)       (1.23)
No. of observations               45            45           45
[R.sup.2]                          0.76          0.66         0.79
Standard error of regression       2.98          3.58         2.83
F-statistic                       12.55 ***      7.46 ***    14.27 ***

                                     (4)           (5)          (6)

Change in economic freedom

Economic freedom, average
  over 1980 and 1985
Change in size of government

Sire of government, average
  over 1980 and 1985
Change in legal structure and
  property rights
Legal structure and property
  rights, average over 1980
  and 1985
Change in sound money               0.31
                                   (0.53)
Sound money, average over          -0.50
  1980 and 1985                   (-1.01)
Change in freedom to trade                     -0.94
  internationally                              (-1.35)
Freedom to trade                                -1.57 ***
  internationally, average                     (-2.90)
  over 1980 and 1985
Change in regulation                                        - 1.65 *
                                                            (-1.82)
Regulation, average over                                     -2.28 ***
  1980 and 1985                                             (-3.05)
Unemployment rate, average         -0.81 ***    -0.79 ***    -0.73 ***
  over 1980 and 1985             (-10.58)     (-11.41)      (-8.86)
Change in GDP growth gap            0.08         0.05         0.06
                                   (0.51)       (0.33)       (0.41)
Change in population aged           0.20         0.20         0.32
  0-14                             (1.02)       (1.10)       (1.56)
Tropical area                       1.96 *       0.72         2.49 *
                                   (1.83)       (0.54)       (2.00)
Distance to coastline               1.20        -0.75         0.52
                                   (0.96)      (-0.62)       (0.25)
Ethnic fractionalization            1.94         2.23         1.63
                                   (0.87)       (0.89)       (0.53)
Change in war                       4.03         2.52         2.52
                                   (1.42)       (0.80)       (1.02)
No. of observations                45           45           44
[R.sup.2]                           0.73         0.75         0.74
Standard error of regression        3.18         3.08         3.17
F-statistic                        10.51 ***    11.41 ***    10.61 ***

(a) Ordinary least squares estimates. Change: is defined
as average over 2000-2003 minus average over 1980 and 1985.
Heteroskedasticity-consistent t-statistics in parentheses
(White method). All regressions also contain a constant term.

* Statistically significant at 10%.

** Statistically significant at 5%.

*** Statistically significant at 1%.

Table 5. Regressions to Explain Change
in the Female Unemployment Rate (a)

                                   (1)           (2)          (3)

Change in economic freedom        -2.97 ***
                                 (-4.02)
Economic freedom, average         -3.21 ***
  over 1980 and 1985             (-3.86)
Change in size of government                   -0.22
                                              (-0.28)
Size of government, average                    -0.62
  over 1980 and 1985                          (-1.23)
Change in legal structure and                                -1.28 **
  property rights                                           (-2.10)
Legal structure and property                                 -1.78 ***
  rights, average over 1980                                 (-4.16)
  and 1985
Change in sound money

Sound money, average over
  1980 and 1985
Change in freedom to trade
  internationally
Freedom to trade
  internationally, average
  over 1980 and 1985
Change in regulation

Regulation, average over 1980
  and 1985
Female unemployment rate,         -0.74 ***    -0.74 ***     -0.77 ***
  average over 1980 and 1985    (-10.71)      (-8.97)      (-13.47)
Change in GDP growth gap          -0.03        -0.19         -0.12
                                 (-0.12)      (-0.72)       (-0.53)
Change in population aged          0.20        -0.12          0.19
  0-14                            (0.82)      (-0.42)        (0.86)
Tropical area                      0.58         2.39         -1.86
                                  (0.33)       (1.12)       (-1.12)
Distance to coastline              0.49         1.80         -0.68
                                  (0.27)       (0.60)       (-0.32)
Ethnic fractionalization           3.35         4.74          3.25
                                  (0.87)       (1.01)        (0.85)
Change in war                      5.42 *       6.11          8.29 **
                                  (1.76)       (1.65)        (2.31)
No. of observations               42           42            42
[R.sup.2]                          0.83         0.75          0.84
Standard error of regression       3.94         4.70          3.82
F-statistic                       16.96 ***    10.83 ***     18.24 ***

                                   (4)         (5)            (6)

Change in economic freedom

Economic freedom, average
  over 1980 and 1985
Change in size of government

Size of government, average
  over 1980 and 1985
Change in legal structure and
  property rights
Legal structure and property
  rights, average over 1980
  and 1985
Change in sound money             -0.07
                                 (-0.10)
Sound money, average over         -0.43
  1980 and 1985                  (-0.70)
Change in freedom to trade                     -2.32 **
  internationally                             (-2.62)
Freedom to trade                               -2.52 ***
  internationally, average                    (-3.48)
  over 1980 and 1985
Change in regulation                                         -3.89 ***
                                                            (-2.76)
Regulation, average over 1980                                -3.46 ***
  and 1985                                                  (-3.23)
Female unemployment rate,         -0.73 ***    -0.77 ***     -0.75 ***
  average over 1980 and 1985     (-8.47)     (-12.57)      (-10.48)
Change in GDP growth gap          -0.18        -0.05         -0.15
                                 (-0.61)      (-0.25)       (-0.71)
Change in population aged          0.02         0.14          0.32
  0-14                            (0.06)       (0.55)        (1.09)
Tropical area                      1.22        -0.06          2.04
                                  (0.69)      (-0.03)        (0.99)
Distance to coastline              1.05        -1.29          1.43
                                  (0.42)      (-0.85)        (0.54)
Ethnic fractionalization           4.92         5.39          3.70
                                  (1.20)       (1.65)        (0.74)
Change in war                      6.92         5.77          5.51
                                  (1.67)       (1.62)        (1.68)
No. of observations               42           42            41
[R.sup.2]                          0.75         0.83          0.83
Standard error of regression       4.68         3.95          3.92
F-statistic                       10.94 ***    16.83 ***     16.97 ***

(a) Ordinary least squares estimates. Change is defined as
average over 2000-2003 minus average over 1980 and 1985.
Heteroskedasticity-consistent t-statistics in parentheses
(White method). All regressions also contain a constant term.

* Statistically significant at 10%.

** Statistically significant at 5%.

*** Statistically significant at 1%.

Table 6. Regressions to Explain Change
in the Youth Unemployment Rate (a)

                                    (1)          (2)          (3)

Change in economic freedom        -5.92 **
                                 (-2.33)
Economic freedom, average         -5.03 ***
  over 1980 and 1985             (-3.75)
Change in size of government                   -1.31
                                              (-1.53)
Size of government, average                    -1.34 *
  over 1980 and 1985                          (-1.91)
Change in legal structure and                               -2.23
  property rights                                          (-1.28)
Legal structure and property                                -2.51 ***
  rights, average over 1980                                (-2.86)
  and 1985
Change in sound money

Sound money, average over
  1980 and 1985
Change in freedom to trade
  internationally
Freedom to trade
  internationally, average
  over 1980 and 1985
Change in regulation

Regulation, average over 1980
  and 1985
Youth unemployment rate,          -0.35 ***    -0.43 ***    -0.43 ***
  average over 1980 and 1985     (-2.97)      (-3.38)      (-3.01)
Change in GDP growth gap           0.04         0.01        -0.10
                                  (0.13)       (0.02)      (-0.34)
Change in population aged 0-14     0.55 *      -0.17         0.70 *
                                  (1.90)      (-0.58)       (1.79)
Tropical area                      7.57 **      9.37 **      3.25
                                  (2.79)       (2.81)       (1.12)
Distance to coastline              5.62         5.41         4.70
                                  (1.60)       (1.42)       (1.34)
Ethnic fractionalization          -6.76        -5.15        -5.67
                                 (-1.01)      (-0.80)      (-0.86)
Change in war                    -11.90 **     -6.72        -8.44 **
                                 (-2.43)      (-1.09)      (-2.42)
No. of observations               32           32           32
[R.sup.2]                          0.71         0.59         0.70
Standard error of regression       4.41         5.26         4.50
F-statistic                        5.93 ***     3.45 ***     5.60 ***

                                    (4)          (5)          (6)

Change in economic freedom

Economic freedom, average
  over 1980 and 1985
Change in size of government

Size of government, average
  over 1980 and 1985
Change in legal structure and
  property rights
Legal structure and property
  rights, average over 1980
  and 1985
Change in sound money             -0.63
                                 (-0.40)
Sound money, average over         -1.46
  1980 and 1985                  (-0.97)
Change in freedom to trade                     -2.89 *
  internationally                             (-1.83)
Freedom to trade                               -2.72 **
  internationally, average                    (-2.21)
  over 1980 and 1985
Change in regulation                                        -3.38
                                                           (-1.63)
Regulation, average over 1980                               -4.23 ***
  and 1985                                                 (-2.92)
Youth unemployment rate,          -0.39 ***    -0.38 ***    -0.36 ***
  average over 1980 and 1985     (-2.84)      (-3.06)      (-3.18)
Change in GDP growth gap           0.18        -0.08         0.07
                                  (0.45)      (-0.25)       (0.27)
Change in population aged 0-14     0.38         0.39         0.57
                                  (0.84)       (1.04)       (1.50)
Tropical area                      5.66 *       6.35 **      9.16 ***
                                  (2.07)       (2.17)       (3.33)
Distance to coastline              4.02         2.67         6.44 *
                                  (1.28)       (0.68)       (1.95)
Ethnic fractionalization          -4.38        -3.70        -6.26
                                 (-0.81)      (-0.50)      (-1.15)
Change in war                     -8.12       -13.20 ***    -6.05
                                 (-1.20)      (-2.96)      (-1.40)
No. of observations               32           32           32
[R.sup.2]                          0.60         0.63         0.69
Standard error of regression       5.16         4.97         4.55
F-statistic                        3.68 ***     4.15 ***     5.42 ***

Ordinary least squares estimates. Change is defined as
average over 2000-2003 minus average over 1980 and 1985.
Heteroskedasticity-consistent t-statistics in parentheses
(White method). All regressions also contain a constant term.

* Statistically significant at 10%.

** Statistically significant at 5%.

*** Statistically significant at 1%.