View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

Home / Publications / Research / Economic Brief / 2024

Understanding the Lack of Skill Specialization in Peru
By Andrea Atencio De Leon, Munseob Lee and Claudia Macaluso

Economic Brief
Aug ust 2024, No. 24-29

Reprinted with permission from VoxDev: “Understanding the Lack of Skill
Specialisation in Peru.”

Jobs in Peru use a larger number of skills than comparable jobs in the U.S.
This lack of specialization is consistent with rms’ hiring of "toderos" (workers
with many skills, do-it-alls), given the high levels of worker reallocation.
Labor markets in poorer countries are characterized by higher worker reallocation rates.1
In richer economies, a dynamic labor market that seamlessly reallocates workers to jobs
goes hand in hand with robust productivity growth and sustained income growth.2 Instead,
in poorer countries, the brisk pace of reallocation is accompanied by high unemployment
and under-employment risk and lower returns to human capital.3
In our research, we explore why fast-reallocating labor markets do not yield productivity
enhancements.4 T o do so, we use detailed microdata to provide new evidence on disparities
in detailed job skills between a rich and a poorer country. While a preference for workers
who are generalists is a response to an economic environment with a high turnover rate,
the lack of specialization limits productivity growth.

Survey of Skills and Employers Recruiting Behavior
(SSERB) in Peru
In the U.S., the O*NET skill survey provides detailed information on the average
importance and level of large set cognitive and non-cognitive skills, as reported by
workers, managers, and occupational experts for each occupation. On the other hand,
there are no data sources to study the skill content of jobs in poor countries. T o remedy

/

the dearth of data, we therefore conducted a similar survey with a nationally
representative sample of workers and rms in urban Peru. Managers were asked to re ect
on the performance of a speci c employee’s occupational tasks and report the importance
of nine skills (cognitive, social, organization, writing, customer service, project
management, people management, nancial skills, and computer skills) on a 1-5 scale (just
like the O*NET ).

Workers in Peru Are Less Specialized Than in the U.S.
We compare occupational skill pro les in Peru and the U.S., and nd that occupational skill
pro les are substantially atter in Peru — di erent professional gures do “a little bit of
everything” instead of specializing in a limited set of core tasks. As an illustration, Figure 1
below compares the average importance of skills for one such occupation, accountants,
across the two countries. In the U.S., accountants specialize in organization and writing
skills, but not so much in project management or cognitive problem-solving skills. Instead,
in Peru accountants attribute more importance to every skill, including those that were
less important for U.S. accountants — hence the atter skill pro le.

Enlarge
/

In our research, we formally measure this in several ways and conclude that we nd similar
patterns in most occupations. Figure 2 below depicts this fact. Each point in the scatter plot
corresponds to an occupation average “coe cient of variation” (CV), which illustrates a
variable’s degree of dispersion around the mean: with little dispersion, the coe cient is
close to zero and the mean is a very good guess for any of the variable’s values. A higher
CV represents a more specialized skill pro le: when some skills are unimportant (value of
1) and other skills are extremely important (value of 5), a mean of 2.5 will be a poor
predictor of either. On the other hand, at, unspecialized skill pro les have the opposite
characterization: if all skills are somewhat important or important (a value of 2 or 3), a
guess of 2.5 is much better a prediction than in the specialized case.

Enlarge
T hus, the fact that occupational coe cients of variation are smaller in Peru reinforces the
idea that Peruvian occupations are performed without skill specialization. Or, in the words
of Eduardo Galeano, Peruvian workers are toderos: “In Caracas, they are called toderos (doit-alls), because they indeed do everything ; these marginalized workers live on occasional
jobs, nibbling work bit by bit: they are servers or servants, stone-cutters or occasional
masons, salespeople or street vendors, occasional electricians or plumbers or wall painters
or car attendants; simply labor, available for whatever comes.”5 While the term toderos
/

refers not only to marginalized workers nowadays, Galeano’s reference highlights its main
meaning: the versatility of Latin American workers –for which we nd evidence in our
paper.

Does Skill Flattening Come From Ine cient Hiring
Practices?
A lack of specialization may come from ine cient hiring technology, yielding fewer and
poorer matches between employers and employees than the same resources would
produce in the U.S. T o investigate this hypothesis we develop a survey of hiring methods
and yields, which we elded both in Peru and the U.S.6 We do not nd evidence of ine cient
hiring in Peru. Open jobs ll quickly, with an average vacancy duration of 9.1 days, and only
10% (1%) of vacancies remain un lled after two weeks (four months). We also nd very little
appreciable di erence in recruiting methods between Peru and the U.S. Our data con rms
that the Peruvian labor market is uid and that hiring procedures, unlike managerial ones
as documented by Nicholas Bloom and John Van Reenen in 2007, are not especially
unproductive.7

The Uncertainty of Labor Availability Explains Firms’
Hiring of Unspecialized Workers
Production tasks are often complementary. Michael Kremer in 1993 proposed the “O-ring
theory,” where production tasks must be executed pro ciently together for any of them to
be of value.8 T he name is a reference to the 1986 Challenger shuttle disaster; it exploded
because it was launched at a temperature that caused one of those components, the Orings, to malfunction. O-rings are small, mechanical gaskets, a seemingly insigni cant part
of a space shuttle. Yet, their malfunction led to disaster. Similarly, the O-ring theory of
production postulates that output can be manufactured only if all inputs (in this case,
skills) are present. As an illustration, one can think of the skills needed to turn on the light
in a manufacturing establishment. Mastery of any speci c task on the production line is of
little use if workers and equipment lies in the dark!
When there is this type of complementarity in production tasks, high worker reallocation
rates mean production losses. Employers are seldom sure what available labor they might
nd as they open their shops or come to the factory oor on any speci c date, as
idiosyncratic shocks a ect workers and their access to transportation, child or elderly care,
or their willingness or ability to work for pay. T hus, employers value workers who can
perform many tasks to minimize disruptions to production. In the terms of our previous
example, when employees are often absent or churn across jobs, employers need all
employees to have the skills to turn on the factory oor lights — then, it doesn’t matter if
someone is missing for production to happen.
/

In our research, we propose a stylized model that delivers a crucial insight consistent with
this intuition: hiring generalists (“toderos”) becomes more pro table when the separation
rate is high.

Conclusion: High Turnover Can Inhibit Specialization
We document that jobs in Peru use a larger number of skills than comparable jobs in the
U.S. T he lack of specialization is consistent with an environment in which skills are
complementary in production and a fast-paced labor market, with frequent separations
and hires, causing a preference for toderos (workers with many skills, do-it-alls). T ogether
with evidence from the literature, we nd evidence that this is the case.9 We conclude that a
brisk pace of job reallocation has the potential to signi cantly inhibit occupational skill
specialization, as we show is the case in Peru, potentially contributing to human capital and
productivity de cits.
Andrea Atencio De Leon is an economist in the Strategy and Policy Review Department at
the International Monetary Fund. Munseob Lee is an assistant professor at the University
of California San Diego. Claudia Macaluso is an economist in the Research Department at
the Federal Reserve Bank of Richmond.

1 See the 2023 paper “Labor Market Dynamics and Development” by Kevin Donovan, Will Jianyu

Lu and Todd Schollman.
2 See the 1993 paper “Job Turnover and Policy Evaluation: A General Equilibrium Analysis” by

Hugo Hopenhayn and Richard Rogerson; the 2001 book chapter “Aggregate Productivity
Growth: Lessons From Microeconomic Evidence” by Lucia Foster, John Haltiwanger and C.J.
Krizan; and the 2014 working paper “Labor Market Fluidity and Economic Performance” by
Steven Davis and John Haltiwanger.
3 See the 2024 paper “Unemployment and Development” by Ying Feng, David Lagakos and

James Rauch and the 2018 paper “Life Cycle Wage Growth Across Countries” by David Lagakos,
Benjamin Moll, Tommaso Porzio, Nancy Qian and Todd Schollman.
4 See our forthcoming paper “Does Turnover Inhibit Specialization? Evidence From a Skill Survey

in Peru.”
5 See the 1971 book Las Venas Abiertas de America Latina by Eduardo Galeano.
6 See the 2021 article “How Do Employers Recruit New Workers?” by Steven Davis, Claudia

Macaluso and Sonya Ravindranath Waddell.
7 See Bloom and Van Reenen’s 2007 paper “Measuring and Explaining Management Practices

Across Firms and Countries.”
8 See Kremer’s 1993 paper “The O-Ring Theory of Economic Development.”

/

9 See the aforementioned 2023 paper “Labor Market Dynamics and Development.”

To cite this Economic Brief, please use the following format: Atencio De Leon, Andrea; Lee,

Munseob; and Macaluso, Claudia. (August 2024) "Understanding the Lack of Skill
Specialization in Peru." Federal Reserve Bank of Richmond Economic Brief, No. 24-29.

T his article may be photocopied or reprinted in its entirety. Please credit the authors,
source, and the Federal Reserve Bank of Richmond and include the italicized statement
below.
Views expressed in this article are those of the authors and not necessarily those of the Federal
Reserve Bank of Richmond or the Federal Reserve System.

Topics
Employment and Labor Markets

Workforce Development

Subscribe to Economic Brief
Receive a noti cation when Economic Brief is posted online.
By submitting this form you agree to the Bank's Terms & Conditions and Privacy Notice.

Email Address

Subscribe

Contact Us
RC Balaban

/

(804) 697-8144

© 1997-2024 Federal Reserve Bank of Richmond

/