The Elusive Quest for Growth: Economists' Adventures and Misadventures in the Tropics
The Elusive Quest for Growth: Economists' Adventures and Misadventures in the Tropics (ISBN: 9780262550420) was written by William Easterly in 2002.
A guiding principle of the book is that economic growth at the national level trickles down. Not necessarily at a desirable or fair level. Essentially, as an outsider, sometimes increasing a state's GDP by N% is a more feasible goal than, e.g., establishing an equitable social service.
Core Theory
Technology alone is insufficient as a predictor of growth.
- To explain variance, poor countries must have remarkably poor technology.
- Addressed again later: "Some countries even have negative productivity growth. ... I wouldn't argue that Costa Rica, Ecuador, Peru, and Syria had technological regress, but clearly other factors got in the way of technological progress. Technologically driven growth is anything but automatic." (p. 176).
Nonetheless, technology is the major causal factor for growth.
New knowledge is complimentary of old knowledge
- cumulative and multiplicative properties, rendering increasing returns
- Author does spend half a chapter exploring how new technology replaces ("destroys") old technology... There is clearly a complicated picture here.
Knowledge leaks. Because knowledge leaks, there are insufficient incentives (from a societal rational perspective) for technological investment in a free market.
- In a macro perspective, can look like poverty traps and 'rich getting richer'.
Individuals match in a free market based on technology/knowledge/skill.
- When individuals have complementary knowledge, they self sort into places of employment with similarity among employees.
- When individuals have substitutive knowledge, they self sort into places with dissimilarity.
- In a macro perspective, can look like brain drain and over-concentration of industries (e.g., Silicon Valley).
Bad governmental policy and corruption are causal factors for negative growth.
Only meaningful suggestion to improve outcomes is to make international development conditional on policy outcomes, not promises to enact policy, e.g...
- falling premia on black market exchange rates (as compared to legal market)
- stabilized inflation rates
- balanced government budgets
Reading Notes
Chapters 2, 3, and (to a lesser extent) 4 are an excellent review of the literature on production models and growth models, from Harrod-Domar on.
The remaining chapters of the first half devolve into disorganized, but still poignant, critiques of specific ideas.
Always remember: it's easy to pick out 'bad' policies in retrospect. It's easy to pick out 'bad' loans in retrospect.
There are so many issues with the critique of debt forgiveness. To each of his points...
- Correlation of debt and debt forgiveness
Of course! You must have debt to have it forgiven, and you must want to borrow more to care about debt forgiveness.
- Easy to pick bad loans when the period of study is 1989 to 1997, encompassing world-wide recessions in 1991-1992, the Mexican crisis in 1994, and the 1997 Asian finance crisis.
- Correlation of debt forgiveness and ratio of debt service to exports between 1979 and 1997
- Exports != ability to pay debt
When modeling exports based on the monetary market, outcomes will be driven by inflation more than anything, so I'm unsurprised to learn there is a correlation of inflation and new credit (either literal new credit or effective via debt forgiveness).
- Indeed, just a few pages later, the author notes that "HIPCs tend to have a more overvalued currency" (p. 130).
- Correlation of debt forgiveness and 'selling off assets'.
- For commodities, oil exports from 10 HIPCs that export oil between 1987 to 1996.
- obvious selection bias
- period of study includes the 1990 oil shock
- In summary, yes, unsurprisingly, poor countries increased oil sales when oil prices doubled.
- For state owned enterprises, revenues from privatization in HIPCs between 1988 to 1997.
- I'd argue this is impossible to model because measurement error in the fair market value of a state owned enterprise is going to be monstrous.
- For commodities, oil exports from 10 HIPCs that export oil between 1987 to 1996.
- Correlation of debt forgiveness and falling per capita GDP.
- As above; you only care about debt forgiveness if you want to borrow more; a country faced with a growing GDP doesn't need more debt.
There is a ton of false equivalence in the core theory.
- 'Quality of work' != knowledge/skill
- Experts make mistakes; experts can be incentivized to prioritize quantity over quality; experts can have dominating incentives; etc.
The urban-rural divide is not evidence of self-sorting.
- Laughable assertion that cities are inherently undesirable.
After establishing that property values/rents differ across it (as part of suggesting that cities are undesirable?), the author points to differential incomes among comparable individuals as evidence of self-sorting. It's tautology at this point.
- "If you think I'm saying something tautological, I'm not. The individual is too small to affect the average..." (p. 157). Well at least the author is self-aware!
Many people, but especially new entrants to the job market take whatever job they can. They don't have the luxury of selecting employers, much less of selecting coworkers. And new entrants, i.e. people who just finished investing in their knowledge/skills, ought to be a primary population of interest.
Author points to brain drain phenomena repeatedly, especially outward from India. I think there's meaningful doubt on that interpretation of population change.
The kicker is how 'luck' is presented as the only alternative plausible theory.
Review of the Literature and Replications
Discussion of Harrod-Domar model. In order to reach a target growth rate, a calculable amount of investment must be secured to fill a financial gap.
- Fit into other, contemporary theories.
- Fit into political and cultural fear of communism.
Rostow's stages of economic growth is a great example of both.
- Suggested that increasing investment from 5 to 10 percent of income was the causal step to starting self-sustaining growth.
Forced investment was credited with the rapid production growth rate seen in the USSR.
Work of Chenery and Strout as well.
- Successive aid injections were more effective.
The Domar model gave way to the Solow-Swan model in economics.
- To say that investment is causal to productivity begs the question of why investors do not cross borders to chase ROIs.
- Certainly shouldn't be necessary to subsidize international investment through NGOs.
- To say that investment is causal to productivity begs the question of why investors do not cross borders to chase ROIs.
- However, international finance never moved on from the model.
John Holsen developed a computerized minimum standard model (MSM) based on it for the World Bank. It was slightly revised (RMSM) within a couple years, but was not substantially changed for decades.
Similar models are used by the IMF and European Bank for Reconstruction And Development.
Author specifically discusses Ghanaian government and the Volta dam project
The author tests the Domar model with current data. It fits just 17 of 88 countries comparative data. Furthermore, the assumption that savings rates increase with aid was validated in just 6.
More fundamentally, the author tests the assumption that increased investment causes short-term growth. Essentially a replication of similar work by Blomström, Lipsey, and Zejan. Author uses 4-year periods, on the basis that international finance economists usually project 4 years forward in reports. They find just 4 countries show this property at any time:
History of using international finance to dictate economic, monetary, and fiscal policy.
- Projects financed on credit contribute to deficits.
- Expectation built into international development: recipients of such credit will sustain unsubsidized growth thereafter
- Contrary evidence: the repetition of loans
- In the 1980s, IMF and World Bank averaged 6 policy-conditional loans to each country in Africa, 5 to each country in Latin America, 4 to each country in Asia, etc. (p. 102).
- "The 24 former Communist economies were the recipients of 143 adjustment loans" (p. 106).
Cote d'Ivoire got 18 between 1980 and 1994.
Prior to 1980, World Bank only placed conditions on projects themselves. Ghana was the leading experiment in credit conditional on policy reforms.
- IMF has always placed conditions on loans.
- There is some defection from loan conditions, but indirect noncompliance is more common (e.g., reform then un-reform).
- Some donors do not care about compliance (e.g., care only about profitability of loan and preventing defaults on debt).
History of debt forgiveness initiatives.
Fifth session of UNCTAD (Manila, 1979) is credited with debt forgiveness of about $6 billion
The G7 conferences held in Venice (1987), Toronto (1988), Houston (1990), London (1991), and Naples (1994) coordinated refinancing terms (e.g., partial forgiveness, longer maturity, lower rates) that are then pushed through the Paris Club (as the, e.g., "London terms", "Naples terms", etc.)
World Bank's Special Program of Assistance (SPA) established in 1987
IMF's Enhanced Structural Adjustment Facility (ESAF)
The WorldBank and IMF jointly recognize a status of heavily indebted poor countries (HIPC) and established an HIPC Initiative in 1996 to forgive debt if good policies are in place
- By 1999, debt forgiveness of about $3.4 billion
- Expanded following the G7 conference in Cologne (1999)
Author posits why debt forgiveness does not solve unequal growth. But I've put all of that under Reading Notes for a reason...