MARKETS, EXTERNALITIES, AND THE DYNAMIC GAINS OF OPENNESS

• Published date: 01 August 2019
• DOI: http://doi.org/10.1111/iere.12381
• Date: 01 August 2019
• Author: Alexander Monge‐Naranjo

INTERNATIONAL ECONOMIC REVIEW

Vol. 60, No. 3, August 2019 DOI: 10.1111/iere.12381

MARKETS, EXTERNALITIES, AND THE DYNAMIC GAINS OF OPENNESS∗

BYALEXANDER MONGE-NARANJO 1

Federal Reserve Bank of St. Louis, U.S.A., Washington University in St. Louis, U.S.A.

Inflows of foreign knowledge are key for developing countries to catch up with the world technology frontier.

I construct a model to analyze the entry decisions of foreign firms and the incentives of domestic firms to

invest in their own know-how, given the exposure to foreign ideas and competition. The model embeds two

diffusion mechanisms typically considered separately in the literature: externalities and markets. I find that

openness allows developing countries to fully catch up only when market transactions dominate the diffusion of

ideas. Externalities are never enough to catch up and may lead to losses in income and welfare.

1. INTRODUCTION

Inflows of production-related knowledge have been widely regarded as the key for developing

countries to catch up with the world technology frontier.2Besides static gains,3the premise that

foreign ideas can be adopted by domestic firms has motivated many developing countries to

open up not just to trade, but also to foreign direct investment (FDI). The gains of catching

up with foreign knowledge can be tantamount to a transition from rags to riches, as supported

by the historical evidence on the diffusion of the Industrial Revolution to Europe, the United

States, and other Western countries, and, more recently, the successful adoption of Western

technology by Japan, Korea, and China. Economists have long emphasized two mechanisms

for the diffusion of knowledge: externalities and markets.4In this article, I build a model that

incorporates both forms of diffusion and use it to examine the impact of openness for the output

and welfare of a developing country.

I explicitly consider (a) the incentives of foreign firms to locate in a developing country—and

bring their know-how with them—and (b) the incentives of domestic firms to invest in their own

know-how. The model allows a clean, analytical characterization of the aggregate formation

of knowledge inside a country, given the equilibrium exposure to ideas and competition from

abroad and the dynamic behavior of foreign ideas inflows in the country. Characterizing the

equilibrium produces a number of simple but important implications for the literature on the

cross-country diffusion of ideas.

∗Manuscript received October 2017; revised September 2018.

1I am thankful to George Alessandria, Pol Antras, Gadi Barlevy, David Backus, Paco Buera, Ariel Burstein,

Berthold Herrendorf, Hugo Hopenhayn, Boyan Jovanovic, Patrick Kehoe, Manolis Galenianos, John Van Reenen,

Andr´

es Rodr´

ıguez-Clare, Esteban Rossi-Hansberg, Thierry Verdier, Jonathan Vogel, and multiple seminar partic-

ipants for comments and suggestions. Qiuhan Sun provided superb research assistance in the last stages of this

project. The views expressed here are exclusively those of the author and do not necessarily reect the opinion of the

FederalReserve Bank of St. Louis or the Federal Reserve System.Please address correspondence to: A. Monge-Naranjo,

Department of Economics, Federal Reserve Bank of St Louis, P. O. Box 442, St. Louis, MO 63166-0442. E-mail:

Alexander.MongaNaranjo@stls.frb.org.

2Among others, see Lucas (2009).

3See, for example, Antras et al. (2006), Burstein and Monge-Naranjo (2009), and Eeckhout and Jovanovic (2012).

The gains are even larger if productivities are not driven by skills but by nonrival factors, as in Ramondo (2014) and

McGrattan and Prescott (2009).

4See Romer (1986), Klenow (1998), Lucas (2002), Klenow and Rodriguez-Clare (2005), and Jones (2005) just to cite a

few of the many papers emphasizing externalities. See Prescott and Boyd (1987a, 1987b), Chari and Hopenhayn (1991),

Jovanovic and Nyarko (1995), and Boldrin and Levine (2010) for models in which knowledge is diffused by markets.

1131

C

(2019) by the Economics Department of the University of Pennsylvania and the Osaka University Institute of Social

and Economic Research Association

1132 MONGE-NARANJO

First, contrary to the emphasis in the vast empirical literature and in recent theoretical

models, externalities are neither necessary nor sufficient for the diffusion and adoption of

foreign ideas. Indeed, I show that it is only when markets fully internalize the costs and benefits

of knowledge transfers that countries end up catching up with the world leaders. Second,

openness to foreign know-how necessarily requires destructive, reallocative, and renewal forces

for firms and occupations inside the country. These forces are only fully unleashed by knowledge

transmitted by markets, and this mechanism is needed for the model to be consistent with salient

empirical observations on the impact of FDI on developing countries. Third, using a simple

quantitative extension of the model, I find that the dynamic gains of openness can be very large,

an order of magnitude higher than the static gains. This holds whether knowledge formation

is driven by markets or by externalities. Interestingly, the gains can be an increasing function

of the preponderance of markets in knowledge formation, a proposition I support throughout

the article.

The basic model is as follows: Entrepreneurs set up firms and production teams in which the

entrepreneur is the manager of a group of workers. In the quantitative model, firms are teams

of the entrepreneur, middle managers, and workers. As in Lucas (1978), the knowledge of the

entrepreneur determines the productivity of the team.5The environment is an overlapping-

generations (OLG) economy in which some of the young build up knowledge to set, manage,

and profit from firms when old. Knowledge, the engine of growth in the economy, has a dual

nature here. On one hand, it is a rival factor, the skills of an individual with a limited span-of-

control on production activities. On the other hand, knowledge may also be a nonrival factor,

the productive ideas that could be used by anybody in the country for building up their own

skills. I consider “closed” and “open” countries. In a closed country, only national entrepreneurs

can set up firms; in an open country, foreign firms are free to enter.6The entry of foreign firms

not only impacts the set of entrepreneurial skills implemented in a country but also the set of

productive ideas available to its young generation.

The ideas upon which a young entrepreneur builds up his know-how come from two sources:

(i) the specific know-how running the firm in which he is a worker and an apprentice and (ii)

the productive ideas implemented by the entire set of firms operating in the country. I assume

that the first source is fully internalized by market transactions. The second source is a pure

externality; it would be highly unrealistic to expect a market arrangement to internalize it. In

this way, the model encompasses, as special cases, two common—but often conflicting—views

of the accumulation and diffusion of knowledge. In one extreme, the young individual’s own

firm is the only source of ideas, as in Prescott and Boyd (1987a, 1987b), Chari and Hopenhayn

(1991), and Jovanovic and Nyarko (1995) and similar to Boldrin and Levine (2010). In the

other extreme, the productive ideas implemented by each firm are uniformly exposed to all the

young in the country. Variants of such externalities have a dominant presence in the growth

literature (e.g., Romer, 1986; Klenow, 1998; Jones, 2005) and trade and growth (e.g., Stokey,

1991; Young, 1991). Notably, by including at least some degree of markets in the diffusion of

knowledge, the model can be consistent with two key empirical observations on the impact

of FDI in developing countries. First, openness to FDI can push preexisting domestic firms to

reduce their productivity, as documented by Aitken and Harrison (1999), Xu (2000), and Alfaro

et al. (2007). Second, entry of firms from developed countries can generate the emergence of

new sectors of domestic firms in developing countries, as described by Rhee and Belot (1990).7

5A long line in the literature links the productivity of firms to the quality of their management, for example, Kaldor

(1934), Lucas (1978), Oi (1983), Rosen (1982), Prescott and Visscher (1980), Garicano (2000), and Bloom and Van

Reenen (2007).

6The emphasis on the cross-border reallocation of management conforms with the observation that multinational

firms heavily rely on home expatriates—and home trained individuals—to manage their operations, especially in

developing countries (see Chapters 5 and 6 of UNCTAD, 1994).

7For support of internal diffusion at the industry level, see Klepper (2001, 2002, 2007) for the car industry and

Agarwal et al. (2004), Franco and Filson (2006), and Franco (2005) for the rigid disk drive industry.

MARKETS,EXTERNALITIES,AND THE DYNAMIC GAINS OF OPENNESS 1133

The preponderance of markets vis-`

a-vis externalities is governed by a parameter in the model.

The knowledge externalities are determined by the set of skills operating in the country at large

and, therefore, driven by the past investments of older entrepreneurs and the entry of foreign

ones. The market (or internal) element for the transfer of knowledge is embedded in labor

market relationships, apprenticeships that internalize the costs, and returns of the learning

opportunities that each firm offers to its workers relative to those offered by all other firms.

On the basis of available learning opportunities, each young entrepreneur builds up his

own skills, foreseeing the set of skills, domestic and foreign, against which he will be compet-

ing. The endogenous formation of skills leads to nontrivial dynamics even for a closed econ-

omy. I provide simple parameter conditions upon which the accumulation of entrepreneurial

knowledge is an engine of sustained growth and the country exhibits a balanced growth path

(BGP). I also show that the convergence to the BGP exhibits an interesting form of collapsing

heterogeneity.

Openness to foreign entrepreneurs impacts the accumulation of skills of the host country in

three ways. First, foreign firms enhance the exposure to ideas of the domestic young working

directly for them. Second, foreign firms may have positive externalities (spillovers) on the

set of ideas circulating in the country, which benefits all the domestic, future entrepreneurs,

including those working for domestic firms. Although these two are positive effects, a third

one is detrimental: Foreign entrepreneurs bid up the cost of labor in the country for all future

periods, reducing the returns and, therefore, the incentives of domestic entrepreneurs to invest

in know-how. I show that in the presence of externalities, openness to foreign firms can even

slow down the formation of domestic knowledge and can even possibly reduce welfare.

Open economies may exhibit an interesting vintage structure for the population of domestic

firms. With less-than-perfect internalization, the domestic entrepreneurs who build up their

skills working for foreign firms do not fully catch up with their foreign counterparts. Moreover,

the young workers working for them will lag further behind next period as entrepreneurs. Then,

endogenously, each vintage will fall below the previous vintages in size and productivity. This

equilibrium structure is similar to that of Chari and Hopenhayn (1991), but with two important

additional aspects: First, the productivity level of each vintage is endogenous. Second, the

set of productive ideas circulating in the country is also endogenous, determined by domestic

investments and foreign entry.

The model implies large—potentially huge—dynamic gains from openness for countries far

below the world knowledge frontier.8Openness would enable poor, unproductive countries to

build up skills on the basis of the more advanced knowledge of developed countries. In present

value terms, the enhanced exposure to ideas more than compensates the negative effect of a

higher price for (unskilled) labor and the negative impact on the current cohort of domestic

entrepreneurs. Interestingly, openness can lead to leapfrogging among developing countries if

a complementarity between domestic and foreign sources of ideas is not too strong in a sense

explained below. In this case, when two developing countries open, the initially poorer one may

surpass and temporarily remain ahead while both countries eventually converge to the same

position in the BGP.

The gains from openness are even stronger if occupation choices are introduced. As in

Antras et al. (2006), Burstein and Monge-Naranjo (2009), and more forcefully in Eeckhout and

Jovanovic (2012), the static gains of openness are larger when individuals can choose between

managerial and labor occupations. With endogenous skill formation, occupation choices can

enhance the gains of openness even further. Not only can they change the structure of the

BGPs toward more productive ones, but occupation choices can also redirect an open economy

away from a laggard (interior) BGP and toward fully catching up with developed countries.

Occupation choices can also accelerate the convergence. Introducing occupation choices leads

to a different and starker form of leapfrogging: After openness, a very poor country may end

8This is quite the opposite from the gains of openness to trade in Stokey (1991) and Young (1991).