The Paradox
If you’re searching for the least controversial finding in
Economics, a strong contender is the observation that people with more
education earn higher wages (Pritchett, 2001). This finding holds virtually invariant of time and
national borders: from imperial Japan to modern day Addis Ababa, more educated
members of society have always tended to earn more money.
A similarly (though not as) well-established finding is that
these private returns tend to vastly outweigh the social returns to education
investment. In this context, social returns refers to the extent to which
education raises the aggregate productivity of the economy of as a whole.
The graph below (from Patrinos, 2004) illustrates this,
comparing the private and social return to different tiers of education from a
cross-section of countries around the world. Patrinos calculates a private return
to primary and secondary education of 26.6% and 17.0% respectively: meaning
that a primary school graduate can expect to earn 26.6% more than someone who
hasn’t attended school, and a secondary school graduate can expect to earn 17%
more than someone who has just completed primary education. Private returns
exceed social returns at each stage: a finding that gives rise to the Micro-Macro
Paradox of Education.
Source: Patrinos
(2014)
The paradox deepens when we consider the significant
positive externalities we tend to associate with education. Put simply, my
ability to read not only makes me more productive; it also makes you more productive (as you spend less
time explaining tasks to me). This implies that we should see a social return
to education that exceeds the sum of
its individual parts; instead, the opposite occurs.
The policy implications of this paradox are hard to
overstate. Education delivery is at the forefront of United Nations Millennium
Development Goals, yet these lacklustre aggregate effects make its prioritisation
over healthcare or infrastructure investments harder to justify. Of course,
there is more to life than productivity and education fulfils many functions
beyond this. Nonetheless, the failure of private returns to translate into
social returns hints at the presence of market failures that, if rectified,
could bring the returns closer together. To help identify these, two compelling
explanations are often put forward to explain the paradox: the piracy
explanation and the wasteful signal explanation.
The Piracy Explanation
The piracy explanation argues that education equips people
to succeed in privately remunerative but socially wasteful activities. If the
basic institutional framework is arranged such that piracy/rent seeking is more
preferable to chemical manufacturing (for instance), schooling may simple make
people more skilled in embezzlement and other anti-competitive, growth-sapping
activities. At a less extreme level, graduates may simply be funnelled into
bloated public sectors – into sinecures that pay well individually, but have
little to no effect on the productivity of the economy as a whole. These
instances help square the high micro returns with low macro returns: education
enables people to command a greater share of society’s resources, without
increasing the size of the pie for all. This explanation points to the need for
effective institutions to make privately remunerative activities more aligned
with those that are socially useful.
The Wasteful Signal Explanation
An alternative explanation states that in many places, the
quality of education is so poor that it equips people with very little
productivity-enhancing skills. There is a plethora of dire statistics to
support this view: in recent assessments in Ghana and Malawi, more than 4/5 of
students at the end of Grade 2 (7 years old) could not read ‘the’ or ‘cat’,
while only half of Grade 3 students in Nicaragua could correctly solve ‘5+6’
(World Development Report, 2018). These students probably attended schools
characterised by the same universally depressing hallmarks: overpopulated
classrooms, absent teachers and insufficient desks/learning materials. Against
this backdrop, it is easy to see how education investment generates such disappointing
aggregate results.
Of course, to complete this explanation, we must reconcile
an ineffective education system with its ability to generate high individual
returns. We can do this by introducing the concept of a ‘signal’. Suppose some
people are naturally more productive than others. In a world of perfect
information, these people earn higher wages. Now suppose that employers cannot identify
how productive workers are ex ante. In the presence of this information
asymmetry, education may be an effective means to ‘signal’ that you are a high
productivity worker – if you graduate from school with a string of high grades,
an employer may infer that you are more skilled than someone with no education,
even if education itself does nothing to
enhance your productivity.
When used this way, investment in education can lead to bad
equilibrium outcomes. Imagine a theatre in which people stand to get a better
view. This prompts others behind them to stand. If it is possible to stand,
doing so is a dominant strategy for any individual. Standing not only has no
benefits to society, it also increases the incentive for other individuals to
engage in uncomfortable standing. The end result is that everybody is worse
off, but no individual has an incentive to deviate – a concept known as coordination
failure.
This explanation offers two policy implications. First, if
people are using education as a signalling device, we can at least make the
signal less wasteful by improving the quality of education. Our understanding
of how to do so has greatly improved in recent years, thanks largely to the
popularisation of randomised control trials (to be discussed in a later post). Even
if we fail in this pursuit, if employers have access to alternative screening
technologies this should reduce the incentive for workers to invest in a
socially wasteful signal. Examples of other technologies include referral
systems, interviews and apprenticeship programmes – all of which allow
employers to better assess the ability of workers ex-ante.
[961 words]
References
This post draws heavily on the work of Lant Pritchett and
his seminal 2001 paper, which offers the best discussion of the education
paradox. Patrinos’ 2014 paper is (to my knowledge) the best paper in measuring
the returns to schooling.
- · Montengro & Patrinos (2014), ‘Comparable Estimates of Returns to Schooling Around the World’
- · Pritchett (2001), ‘Where Has All the Education Gone?’
