Print Reference This Reddit This

Outsourcing and innovative activity

Literature Critique

Mariam Metskhovrishvili

Supervisor: Adolfo Cristobal Campoamor

International School of Economics at Tbilisi State University

Tbilisi, 2010

1. Introduction

Before the 20th century, communication and transportation between countries was too costly and took much time. If today we can travel around 100 miles in 2 hours, at that time the same distance might took about 2-3 days. Because of this problem, goods and services were produced within country, in general. Companies were doing everything themselves, manufacturing from raw materials to final goods.

Today technologies are improving. Because of Internet, mobile phone, fax, and other modern achievements, communication and transportation costs are perceptibly low. These technologies made delivering costs of some kinds of tasks and services relatively easier and cheaper. The term “outsourcing” (or offshoring) which is more and more commonly used nowadays is defined as a business strategy utilized by a company to delegate a certain task to another firm and let that firm do the task on behalf of the company. It should be mentioned that outsourcing and offshoring are not the same; Outsourcing stands for the distribution of the tasks and processes to the other companies without taking account weather those companies are domestic or foreign. Offshoring stands for distribution of the tasks and processes to any foreign company, does not matter if that company is external or branch of the firm. Hence, outsourcing may include tasks performed between foreign and also domestic companies, while offshore include tasks performed only internationally.

Since the middle of 20th century, the number of goods and services outsourced has increased significantly. Usually, firms outsource with the intention of reducing the cost of doing business, obtaining more information to increase product quality, to gain access to world-class capabilities, to improve company’s focus, and etc. But firms still have to decide whether outsource or produce at home. Many economists have done and continue to work about consequences of outsourcing (Feenstra and Hanson, 2001; Grossman and Rossi-Hansberg, 2006; and etc.), but it is also interesting to investigate how firms choose between outsourcing and home-making. In this article I will review some papers that address decision making issue at firm level, connection between the frequency of technological progress and outsourcing intensity. There seems to be some confusion as to whether international trade is a relevant explanation (face to face with skill-based technological changes) to account for rising inequality.

2. Literature Critique

2.1 Technological changes and outsourcing

First, I want to review the paper by Bartel, Lach, and Sicherman, “outsourcing and technological change” (2009). This study gives an explanation for the question of many discussions: What are the reasons of increase in outsourcing during the last few decades? The analysis is conducted at firm (or plant) level. The authors focused on different point of view, instead of study nature of technologies, they studied the probability of technological changes.

The authors investigate how probability of changes in technology affects firms’ decision of outsourcing. This seems to me an interesting question for investigation. Firms mostly outsource in order to decrease costs, for example making a call-centre in less developed country where wages are lower; technological changes do the same, for example, invention of paper and plastic based packages which are cheaper than their glass counterparts and hence, lowers the costs. Nevertheless, there are some exceptions from this general ideas to which I will return soon.

Bartel, Lach, and Sicherman constructed a dynamic model in which goods are produced using labor and innovative technology. Firms can either outsource or make at home. The latter choice requires some initial expenses which include cost of purchasing equipment, cost of training, and things that are needed to use new technology. The idea of the paper is that with the increase number of technological changes, magnitude of outsourcing will decrease and that larger firms have lower incentive to outsource than smaller firms. This happens because when technologies increase more rapidly firms have to purchase new equipment in relatively smaller periods of time, which will be harmful for production. This makes outsourcing more attractive.

The authors assume that if a firm decides to make at home, it will make home always with new technologies. However, it is not always profitable to produce using the newest technology (Clayton M. Christensen and Joseph L. Bower, 1995). An example of such unnecessary new invention was the Dvorak Simplified Keyboard (“DSK”) with optimally located letters and punctuation signs. These keyboards failed to replace ordinary QWERTY keyboards in massive production because people were already accustomed to those QWERTY keyboards and replacement of all old keyboards with these new ones was costly and, what is the most important, unnecessary.

To support the main point of the model, authors used Spanish data (Encuesta sobre Estrategias Empresariales) of manufacturing firms between 1990 and 2002. This is an annual survey of 3,195 firms. They studied whether manufacturers were outsourcing finished products or part of production, or not, and if this was the case, they indicated value of outsourcing in production.

To analyze and investigate the central idea of the model, authors needed a proxy variable for technological changes. They used number of patents (or Research & Development expenditure) as a proxy, which they took from NBER Patent Citations Data File which is shown by Hall, Jaffe and Trajtenberg (2001). Since, particular patents, as usual, are used by manufacturers producing in other industry, researchers had to transform this data from the quantity of patents from each industry into the quantity of patents in industry in which they are used. This proxy variable seems to be a good substitute for technological change because today all new inventions are being patented, so they seem to be correlated and this could be an indicator of new technologies. However, there are list of works pointing out that broadening number of patents does not lead, or at least has no direct impact on increasing number of innovations (Sakakibara and Branstetter, 2001; Kremer, 1997; Schankerman, 1998). Bartel, Lach, and Sicherman also considered components of previous research, related to outsourcing. They included average labor cost, capacity utilization during the year, one dummy variable indicating if market size of firm increased or decreased during the year, second dummy variable indicating the age of firm, another dummy variable which is equal to one if company is producing standardized products, and export propensity. This empirical analysis confirmed the main result of the model, i.e. increasing technological changes increases outsourcing.

Although, one, from my point of view, important variable is not included in this regression; this is cultural differences in the host and home country, because if there is a large gap between these cultures, it will be more difficult to communicate and this will cause some inconveniences. For example, problems that may occur regarding language; the outsourcing of call centers and costumer care centers has mainly concentrated in India, due to its large English speaking workforce. This could not be done, for example, in China.

Overall, the paper is well structured. It is written in very explicit manner and it is easy to understand.

2.2 Reverse causality

The second paper that I will review is by Holger Gorg and Aoife Hanley, “Services Outsourcing and Innovations: An Empirical Investigation” (2009). This paper examines the reverse causality of the Bartel, Lach, and Sicherman paper. It examines dependence of innovations on outsourcing of services at firm level, i.e. how expanding outsourcing influences technological changes. The purpose of the paper is to analyze this question empirically, based on previously studied theoretical literature.

The paper is based on models introduced by Glass and Saggi (2001) and Leahy and Montagna (2008). The model of Glass and Saggi is a dynamic model that is based on idea that companies outsource services when prices of inputs are relatively low abroad. The purpose of Glass and Saggi’s paper is to check the statement that international outsourcing leads to decrease in welfare of workers and a lower relative wages in outsourcing country. They conduct a North-South product cycle model, which predicts that outsourcing lowers relative wages in North (more industrialized country). But they also showed that this negative effect is outbalanced by the reduction in costs, which will lead to higher profits, which in turn leads to more investment in R&D. Here outsourcing affects innovations indirectly. However, it may happen that outsourcing enables firm to readjust to more innovative technology. In this case, outsourcing primarily affects technological changes. The model of Glass and Saggi examines only international outsourcing. Gorg and Hanley investigate this prediction for both, international and domestic outsourcing of services. For this, they use the model by Leahy and Montagna (2008). This is a four step model. In the first step two companies have options either to outsource domestically or produce in-house. In the second step, if firm decides to make in-house it will face some fixed cost related with expanded boards of the company. On the other hand, if the firm decides to outsource it will avoid this fixed cost; in this situation fixed cost is undertaken by supplier. The amount of this investment cost determines quality of intermediate input and it is chosen by supplier after the agreement is done. In the third step, outsourcing firm and supplier agree about price of intermediate. The authors assume that firm has time to agree only with one supplier. And in the final step the firms begin to compete in Cournot competition. The idea of the model is that investment by the supplier is made after agreement and supplier has an incentive to invest less in quality. Outsourcing firm will face relatively higher marginal costs. Also, as competitor knows that this company will face higher costs, it also invests less in quality and hence, increases its output. And this increase in rivals output affects firm’s profits negatively. So, Leahy and Montagna show that outsourcing may lead to higher marginal costs and lower profits. I think this model is not quite realistic because here companies agree on price after choosing one as a supplier. In the real world prices play an important role in making decision whether outsource or not, so, companies bargain on prices before making an agreement. Also, the assumption that a company has time to negotiate only with one supplier does not demonstrate the real world situation. In real world company can negotiate with several suppliers and then choose the best one.

Gorg and Hanley study these two models empirically. They used firm level data for companies of the Republic of Ireland. This data is Annual Business Survey of Economic Impacts (ABSEI), from 2002 to 2004, which is an annual survey of firms with not less than 10 workers. The data contains variables such as purchase of services, overall spending on R&D, total sales, work-force, exports, spending on trainings domestically, spending on wages and total purchases, and also, an information if a company purchases intermediates via internet, and if yes, what is its ratio compared to total sales. The data contains many valuable components; it quite answers the question that authors set forth. The presentation of it is provided clearly and convincingly. But still, from my point of view, this is very short-time data for investigating effects and consequences of outsourcing and R&D. Even though the paper studies outsourcing of services, this data is still too short-time for observing changes in innovations and making conclusions. The reason why they choose this data is that it is the most recent data they could use.

In their regressions Gorg and Hanley consider also variables from previous researches. The authors constructed two regressions. The first regression shows dependence of international outsourcing of services and innovation activity. The variables that are included in this regression are: outsourcing defined as imported (domestically sourced) services divided by sales, employment which is measure of the firm size, dummy variable which shows whether firm outsources or not, investment in skill-upgrading, plant size, profitability, international and domestic outsourcing of materials, and also, dummies to control for any sector specific time varying effects. Gorg and Hanley also run a regression to examine the link between profitability and international and domestic outsourcing of services. This regression also includes variable containing characteristics such as: firm size, foreign ownership, training and export dummies, market share, and industry sales. Gorg and Hanley begin by running these two regressions separately, treating outsourcing as exogenous, and then jointly treating outsourcing as endogenous.

There was a problem with described regression. As Bartel, Lach, and Sicherman show technological change is somehow cause of expanding outsourcing, and hence, is connected with companies’ spending in R&D, then this raises problem of endogeneity. But Gorg and Hanley consider this complication and use instrumental variables for outsourcing which are introduced by Ruane and Sutherland (2005). These variables are twice lagged growth tares of international and domestic outsourcing of services and materials intensities and a dummy which shows if a company is exporting to the UK and the value of these sales.

The first regression proves the model prediction that international outsourcing of services affects R&D activity, as well as domestic outsourcing, but in the latter case the effect is smaller. The second regression shows that international outsourcing and profits are positively depended but domestic outsourcing is not. This negative dependence of domestic outsourcing and profitability is shown by Leahy and Montagna (2008), also. This is very surprising result. If only international outsourcing is profitable, this may imply that only large companies profit from outsourcing, because small companies are less likely to outsource internationally. This result provokes interesting questions are those (internationally) outsourcing firms the same ones that introduce the innovations? Do the other firms (that outsource domestically) simply survive to the wave of innovations, generated by large firms? These issues are studied by Veugelers and Cassiman (1998), which will be reviewed in the next section.


It would be also interesting to investigate whether technological change, which is affected by outsourcing (according to Bartel, Lach, and Sicherman), is somehow depended on firm size. Reinhilde Veugelers and Bruno Cassiman in their paper – “Make and buy in innovation strategies: evidence from Belgian manufacturing firms” – investigate empirically how firms decide to source innovations. They used the Eurostat Community Innovation Survey to conduct analysis on Belgian firms.

The authors first study the determinants that have an influence on making decision whether or not to innovate. In their empirical model researchers include commonly recognized explanatory variables like firm size and measures of technological opportunity; as well as variables that measure obstacles of innovation such as cost and risk obstacle (OBSTcost), lack of information (OBSTinfo), ‘no need for innovation’ (OBSTneed), lack of opportunities (OBSTopport), resistance against change within the firm (OBSTresist), and ease of imitation of innovation (OBSTimit). They include firm size through two dummies: sizeS (which is 1 if the firm has less than 50 employees) and sizeL (which is 1 if the firm has more than 500 employees). The researchers conduct split regression for both values of sizeS and sizeL to study marginal effects. The results show that marginal effects of other independent variables are much more pronounced in explaining innovation by small firms than it is for large firms. Also, industry dummies become more significant.

The regression show that coefficient of export intensity is highly significant, which is more or less intuitive. The coefficient of OBSTcost occurs with positive sign and significant, this means that high costs and high risks do not prevent firms from innovating. This result is very surprising and is rejected by Bartel, Lach, and Sicherman’s paper where high costs of innovation prompt firms to outsource. Similar effect has the lack of technological information (OBSTinfo). OBSTopport has the expected negative sign, but appears insignificant in the decision whether or not to innovate. More important in this decision is OBSTneed and OBSTresist, which have expected negative signs.

The second part of the paper focuses on the sourcing decision for innovative firms. Firms have three options of sourcing technology. The first option is to BUY technology from external source. The second option is to develop new technology, the MAKE strategy. And the third option is a combination of these two options, i.e. MAKE and BUY strategy. Veugelers and Cassiman saw that the percentage of the firms that accept this latter decision is significantly high than the percentage of the firms that accept only MAKE or BUY strategy. The authors study what factors determine this combined strategy and report results which were the most significant. The regression shows that large firms are more likely to innovate than small firms.

3. Conclusion

The paper of Bartel, Lach, and Sicherman shows a causal connection between the frequency of technological changes and outsourcing intensity. Technological progress requires firms to outsource in order to restructure their production and facilitate the adoption of innovations. I think that this could generate a virtuous circle, in which innovation generates outsourcing, which itself widens the market for the adoption of new innovations, which themselves call for further outsourcing, and etc. On the other hand, Gorg and Hanley explore the reverse causality (from outsourcing to innovation) where, in principle, both positive and negative connections are possible, since (according to Leahy and Montagna) outsourcing could reduce the profitability of firms in a predatory (Cournot competition) environment.

There are many debates about concepts of outsourcing and home making. However, my opinion is that well managed outsourcing is always beneficial for a company. Of course, the organization should have sufficient management skills and the ability to adapt new behaviors and processes to successfully manage an external part of their business. The management should be done in order to maximize profits and at the same time to minimize riskiness. Well managed outsourcing increase the total performance of the company as empirical work suggests. “Do what you do best,” says Fabio Rosati, chief executive officer of Elance, “You need to acknowledge that you cannot do everything well, and sometimes your company will need help.”

Need help with your literature review?

Our qualified researchers are here to help. Click on the button below to find out more:

Literature Review Service

Related Content

In addition to the example literature review above we also have a range of free study materials to help you with your own dissertation: