Relation between intelligence and unemployment at the individual and national level.

Lynn, Richard ; Zietsman, Garth

Introduction

In this paper we explore the relationship between intelligence and unemployment at the levels of individuals and nations. At the individual level, several studies have shown that there is an association between low intelligence and unemployment. Toppen (1971) reported that a sample of the unemployed in the United States had an average IQ of 81. Lynn, Hampson & Magee (1984) reported that a sample of the unemployed in Northern Ireland had an average IQ of 92. Herrnstein and Murray (1994) reported that in a sample in the United States, 14 per cent of those with IQs below 74 had been unemployed for one month or longer during the preceding year, and the percentages of the unemployed declined in successively higher IQ groups to 4 percent among those with IQs above 126.

The likely explanation for the association between low intelligence and unemployment is that individuals compete for jobs, and employers select those that they judge will be the most efficient. Sometimes employers select on the basis of intelligence tests. For instance, the US military tests applicants and normally only accept those with IQs above 92 (Department of Defense, 1998). More commonly, employers select on the basis of educational qualifications (as a proxy for intelligence plus the capacity for application). Employers are reasonable to use intelligence as a criterion for employability, since numerous studies have shown that intelligence is positively related to the efficiency of performance in the United States (Ghiselli, 1966; Hunter & Hunter, 1984); Schmidt & Hunter, 1998) and Europe (Salgardo, Anderson, Moscoso, et al., 2003). The effect of this is that in a competitive labor market those with low IQs find it difficult to obtain employment and are more likely to be unemployed.

A further factor is that United States and the United Kingdom, where the association between low intelligence and unemployment has been reported, have minimum wage legislation. The effect of minimum wage legislation is that employers are reluctant to employ those with low IQs and associated low skills at the required minimum wage. These and other economically developed nations also provide welfare benefits for the unemployed and the effect of these is that those with low IQs and associated low skills are able to survive as unemployed.

We are not able to make a prediction about whether the association between low intelligence and unemployment among individuals can be extended to nations. There are two countervailing forces. First, countries with high per capita incomes also tend to have high national IQs; the correlation between national IQ and per capita GDP 0.64 (Lynn & Vanhanen, 2006, p. 104). Economic theory predicts that countries with high GDP should have high rates of unemployment, as a result of the trend of corporations to outsource employment to countries with low GDPs, to gain the advantage of low labor costs. During recent decades there has been an increasing trend for corporations in economically developed high GDP/high IQ nations to move manufacturing and services (e.g. call centers) to poorer countries where labor costs are lower. This generates unemployment in high GDP/high IQ nations and employment in poorer countries. This would lead to a positive correlation between national IQs and rates of unemployment.

There is a countervailing force that national populations with high IQs should have the same advantages as individuals within nations in selling their products and services. National populations with higher IQs should have a competitive advantage because they can make and provide more cognitively demanding and higher value products and services (e.g. aircraft, computers, banking, etc.) that require high IQs, and that national populations with lower IQs are unable to make and provide. There is a strong demand for the products and services that high IQ populations provide, and this generates higher employment in high IQ nations, entailing a positive a positive correlation between national IQs and rates of employment. It is a matter for empirical investigation which of these two countervailing force is the stronger, and therefore whether the correlation between national IQs and rates of unemployment are positive or negative.

Method

To examine which of the two countervailing forces acting on the relationship between national IQ and rate of unemployment is the stronger, we have computed the correlation between these two variables. National IQs for 192 nations comprising all the nations in the world with populations greater than 40,000 are taken from Table 4.3 in Lynn & Vanhanen (2006). National data for unemployment are taken from the Central Intelligence Agency (CIA) Yearbook (2003 & 2008). For a few nations the CIA Yearbook gives the official unemployment figure and an estimate for underemployment (based on those working part time, etc). In these cases we have used the official estimate and ignored the estimate of underemployment. The general effect of this decision is to reduce the degree of unemployment of mainly low IQ countries and therefore underestimates the true size of the relationship between IQ and unemployment.

The CIA Yearbook figures are also not always for a single calendar year. For a number of nations the Yearbook gives the most recent estimate at the time of publication. Some of these are up to 5 years old. Taking this into account we have defined two periods encompassing a range of dates. The first period is from 1996 to 2002 (93.6% of the unemployment figures are within the range 1999 to 2002). The median year is 2001. The second period is from 2003 to 2009 (92.8% of the unemployment figures are within the range 2005 to 2008). The median year is 2008.

The first period (median year 2001) has unemployment data for 141 nations for which national IQ data exist. The median unemployment figure was 10.3% and the mean 14.3%. The standard deviation was 12.3 and first and third quartiles 5.4% and 18.25% respectively. The second period (median year 2008) has unemployment data for 128 nations for which IQ national data exist. The median unemployment figure was 6.8% and the mean 11.1%. The standard deviation was 13.894 and first and third quartiles 4% and 11.8% respectively. The average of the two periods yielded unemployment data for 107 nations for which national IQ data exists.

Results

The scatterplot of national IQ and unemployment suggested that the relationship is not linear, especially during the first period when international unemployment is lower, so we fitted a non-linear equation using least squares estimation of the parameters. Using the average unemployment for the two periods the equation is

% unemployment = 0.21*e**(-0.05*IQ + 8.45).

The correlation between the unemployment estimate based on this equation and national IQ unemployment is r = -0.66 (107 nations). This figure can be corrected for unreliability of both variables. The correlation between the unemployment figures in the two periods is r = 0.81. This is the reliability estimate of the average unemployment figure. The reliability of national IQs given by Lynn & Vanhanen (2006) is 0.94. Correcting for unreliability, the correlation between national IQ and unemployment is r = -0.756 and the unemployment variance explained by national IQ is 57.2%.

We have also examined the relationship between national GDP in and rates of unemployment. The correlation is negative (r= -0.38), i.e. counties with high GDP have lower rates of unemployment. To examine more closely the contributions of national IQs and GDP to rates of unemployment, we have run a multiple regression entering national IQs and GDP as independent variables. The result is that the beta coefficient for national IQ is -0.59 and the beta coefficient for GDP is -0.02. The GDP coefficient is effectively zero, showing that high GDP does not generate either higher or lower rates of unemployment, independently of high national IQ.

We have also examined the relation of economic freedom to rates of unemployment, controlling for national IQ. National differences in economic freedom were taken from the Economic Freedom of the World Index (Gwartney & Lawson, 2008). The results are that economic freedom independently explains (after accounting for the relationship of both with IQ) a further 12.9% of the variance in unemployment, with less economic freedom increasing rates of unemployment.

Discussion

The results resolve a problem for economic theory that that countries with high GDP could have high or low rates of unemployment. High of unemployment should be present as a result of the trend of corporations to outsource employment to countries with low GDPs to gain the advantage of low labor costs. Alternatively, counties with high GDP also have high IQs, and this enables them to produce cognitively demanding goods and services that cannot be produced by low IQ populations. We show that countries with high GDP tend to have low rates of unemployment (r= -0.38). We believe that the explanation for this is that countries with high GDP also have high national IQs. We show that the correlation (corrected for unreliability) between national IQ and unemployment is r = -0.756, and hence that low national IQ explains 57.2% of the variance in unemployment. We show further that when national IQ is controlled, national GDP has no effect on rates of unemployment. We propose the likely explanation for the negative correlation between national IQ and rates of unemployment is that national populations with high IQs have the same competitive advantages as individuals within nations in selling their products and services. This generates higher employment in high IQ nations, entailing a negative correlation between national IQs and rates of unemployment.

References

Central Intelligence Agency. Yearbook (2003/2008) Washington, D.C.: CIA. Department of Defense (1998) Population Representation in the Military Services.

Ghiselli, E.E. (1966) The Validity of Occupational Aptitude Tests. New York: Wiley. Gwartney, J.D. & Lawson, R.A. (2008) Economic Freedom of the World: 2008 Annual Report. www.freetheworld.com/release.html.

Herrnstein, R.J. & Murray, C. (1994) The Bell Curve: Intelligence and Class Structure in American Life. New York: Free Press.

Hunter, J.E. & Hunter, R.F. (1984) Validity and utility of alternative predictors of job performance. Psychological Bulletin, 96, 72-98.

Jensen, A.R. (1980) Bias in mental testing. New York: The Free Press.

Jensen, A.R. (1998) The g Factor. The Science of Mental Ability. Westport, CT: Praeger Publishers.

Lynn, R., Hampson, S., & Magee, M. (1984) Home background, intelligence, personality and education as predictors of unemployment in young people. Personality and Individual Differences, 5, 549-558.

Lynn, R. & Vanhanen, T. (2006) IQ and Global Inequality. Augusta, GA: Washington Summit Publishers.

Salgado, J.S., Anderson, N., Moscoso, S., Bertua, C., de Fruyt, F. & Rolland, J.P. (2003) A meta-analytic study of general mental ability validity for different occupations in the European community. Journal of Applied Psychology, 88, 1068-1081.

Schmidt, F.L., & Hunter, J.E. (1998) The validity and utility of selection methods in psychology: practical and theoretical implications of 85 years of research findings. Psychological Bulletin, 124, 262-274.

Toppen, J.T. (1971) Unemployment: Economic or psychological? Psychological Reports, 28, 111-122.

Richard Lynn, Address for correspondence: Lynnr540@aol.com

University of Ulster, Coleraine, Northern Ireland Garth Zietsman

Johannesburg, South Africa