The Economics of Education: A Time-Series Reflection on UCL's Founding Vision

Introduction

Founded in 1826 as a secular alternative to the traditional university system, University College London was built on the idea that education should be accessible and the pioneer of progression. That principle was not only socially progressive but also carries a somewhat implicit economic claim: expanding access to education should improve economic outcomes through higher productivity and human capital formation. (1)

 

Education is widely viewed as a driver of economic growth through its effect on productivity and innovation. However, this relationship is not as straightforward to identify empirically. Economic growth is made of many variables and volatile while education evolves gradually, making its effects less detectable in macroeconomic data. This article examines whether changes in human capital have measurable effects on UK economic growth (1971-2019). Though initial specifications suggested a weak relationship, once underlying trends are accounted for, changes in human capital emerge as a significant predictor of growth, in consistency with human capital theory. (2)

​

Economic Framework

To analyse the relationship between education and economic growth, the primary framework utilised is the human capital theory developed by Gary Becker. Becker formalises education as an investment decision rather than a consumption decision. Individuals accumulate skills and knowledge that increase their productivity which increases economic output in aggregate. (3) This framework implies that improvements in human capital translate into higher economic growth through more efficient labour and greater innovation.

 

This mechanism, at the macroeconomic level, is embedded in the augmented Solow growth model where human capital enters alongside physical capital as a determinant of output. (4) A key implication of this framework is that the effects of education materialise gradually, reflecting the time required to acquire skills and use them in production. People get hard and soft skills required in jobs through education but need time to work on and improve them.

 

​​

​The slow evolution of human capital and short-run volatility of economic growth poses an empirical challenge in the time-series analysis. Appropriate specifications are needed to identify the relationship. So, short-run regressions may underestimate the true effect of education.

 

This perspective is consistent with John Stuart Mill, whose alma mater is UCL, who viewed education as a driver of long-term economic and social progress rather than an immediate policy instrument. (5) Therefore, the framework suggests that the relationship between education and growth is likely to be gradual.

​

Data and Methodology

The analysis uses annual UK data spanning from 1971 to 2019. The dependent variable, GDPGrowth, (6) measures real GDP growth and is obtained from the World Bank database. Human capital, HC, (7) is proxied using the Penn World Table human capital index which adjusts years of schooling by estimated returns to education. Education spending, EduSpending, (8) is government expenditure on education as a share of GDP alongside control variables Investment (9) (gross capital formation) and PopGrowth (10) (population growth), all three also sourced from the World Bank.

 

Given the time-series nature of the data, all variables are ordered chronologically and transformed into first differences where appropriate. This addresses the issue of trending variables and avoids spurious regression, a common problem in macroeconomic time-series analysis. The main specification therefore focuses on changes rather than levels.

 

The baseline model is estimated using OLS with heteroskedasticity-consistent standard errors:

​​​

​

The initial level-based specifications are not reported due to weak explanatory power and multicollinearity. The differenced specification provides a more reliable framework in identifying the short-run relationship between changes in human capital and economic growth.

​

Results and Discussion

The baseline regression indicates that changes in human capital are positively associated with GDP growth. The coefficient on ΔHC remains statistically significant which suggests that improvements in education translate into measurable increases in economic growth. This relationship only becomes visible once the data is expressed in first differences, rather than level-based interpretation, highlighting the importance of addressing underlying trends in time-series analysis. (11)

​

Investment exhibits a strong and highly significant positive effect on growth which can be tied back to the Solow growth framework. This provides a useful benchmark for the model as investment remains a robust predictor of economic expansion. However, changes in education spending are not statistically significant, implying that increases in government expenditure do not mechanically translate into higher growth. This suggests that the relationship develops through the accumulation of effective human capital rather than policy inputs by themselves.(12)

​

​

​

​

​

​

​​

​

​

​

​

​

 

Final Regression Table​

​

Initial level-based regressions suggested weak or inconsistent relationships between education and growth but these results were largely driven by trending variables and multicollinearity. The relationship is much easier to identify once differencing is applied. 

​

Diagnostic tests support the reliability of the specification. VIF levels are low indicating no multicollinearity. Breusch-Pagan tests also fail to detect heteroskedasticity. Residuals do deviate from normality but no systematic pattern is observed, likely reflecting macroeconomic shocks, so not posing a major limitation to the model.

​

The findings show a gradual interpretation of the relationship between education and growth. Though improvements in human capital do not generate immediate effects, they contribute to growth through cumulative productivity increases which is in line with the human capital theory. (13) Overall, it can be said that education matters, but the importance can only be understood when measured appropriately.

​

Conclusion

This article examines whether improvements in human capital are associated with economic growth in the UK using time-series data and a differenced regression framework. Though initial specifications suggest little relationship, the results become more significant when underlying trends are addressed. Changes in human capital emerge as a statistically significant predictor of growth, and education spending alone does not illustrate such an effect.

 

These findings point us toward a more nuanced interpretation of this relationship. The level of investment in education is not all that matters, but how effectively the investment translates into productivity over time. Growth seems to respond to the gradual accumulation of human capital rather than short-run policy changes.

 

This begs the question: if education’s economic effects are gradual and hard to detect, are current policy frameworks too focused on immediate outcomes rather than long-term capability building?

​

References:

1. University College London, “History,” UCL, accessed March 15, 2026, https://www.ucl.ac.uk/about/who-we-are/history.

2. Gary S. Becker, Human Capital: A Theoretical and Empirical Analysis, with Special Reference to Education, 3rd ed. (Chicago: University of Chicago Press, 1993), National Bureau of Economic Research, accessed March 15, 2026,https://www.nber.org/books-and-chapters/human-capital-theoretical-and-empirical-analysis-special-reference-education-third-edition.

3. Becker, Human Capital, https://www.nber.org/books-and-chapters/human-capital-theoretical-and-empirical-analysis-special-reference-education-third-edition.

4. N. Gregory Mankiw, David Romer, and David N. Weil, “A Contribution to the Empirics of Economic Growth,” Quarterly Journal of Economics 107, no. 2 (1992): 407–437, accessed March 15, 2026, https://academic.oup.com/qje/article-abstract/107/2/407/1838296.

5. John Stuart Mill, Principles of Political Economy, vol. 1, Book I, Chapter VII (1871), accessed March 15, 2026, https://en.wikisource.org/wiki/Principles_of_Political_Economy_(J.S._Mill,_1871),_vol._1/Book_I,_Chapter_VII.

6. World Bank, “GDP Growth (Annual %), United Kingdom,” World Bank Data, accessed March 16, 2026, https://data.worldbank.org/indicator/NY.GDP.MKTP.KD.ZG.

7. University of Groningen, “Penn World Table,” GGDC, accessed March 16, 2026, https://www.rug.nl/ggdc/productivity/pwt/.

8. World Bank, “Government Expenditure on Education (% of GDP), United Kingdom,” World Bank Data, accessed March 16, 2026, https://data.worldbank.org/indicator/SE.XPD.TOTL.GD.ZS.

9. World Bank, “Gross Capital Formation (% of GDP), United Kingdom,” World Bank Data, accessed March 16, 2026, https://data.worldbank.org/indicator/NE.GDI.TOTL.ZS.

10. World Bank, “Population Growth (Annual %), United Kingdom,” World Bank Data, accessed March 16, 2026, https://data.worldbank.org/indicator/SP.POP.GROW.

11. C. W. J. Granger and Paul Newbold, “Spurious Regressions in Econometrics,” Journal of Econometrics 2, no. 2 (1974): 111–120, accessed March 17, 2026, https://www.sciencedirect.com/science/article/abs/pii/0304407674900347.

12. Eric A. Hanushek and Ludger Woessmann, “The Economics of International Differences in Educational Achievement,” NBER Working Paper No. 15949 (2010), accessed March 17, 2026, https://www.nber.org/papers/w15949.

13. Becker, Human Capital, https://www.nber.org/books-and-chapters/human-capital-theoretical-and-empirical-analysis-special-reference-education-third-edition.