Intellectual Property Protection, Product Life-cycle Lengths, and the Boundaries of Multinational Corporations

doi:10.21203/rs.3.rs-3877554/v1

Download PDF

Research Article

Intellectual Property Protection, Product Life-cycle Lengths, and the Boundaries of Multinational Corporations

https://doi.org/10.21203/rs.3.rs-3877554/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

This paper presents a theoretical model based on transaction cost analysis and the resource-based view. It establishes that the corporate boundaries of multinational corporations are determined by the length of the product life cycle and the level of intellectual property protection in the host country. The empirical results based on the panel data of U.S. multinational corporations' import trade of 208 types of products in 18 countries from 2002 to 2013 validate the extrapolation results of the theoretical model. The study finds that the most significant determinants of multinational corporations' boundaries are the length of the product life cycle and the level of IPR protection in the host country.

Boundaries of the multinational firm

Intellectual property protection

Product life-cycle lengths

The establishment and management of boundaries for multinational corporations (MNCs), as significant players in global commerce, have far-reaching implications for the global economic system. Deciding how to set the boundaries of MNCs is a crucial decision for operators in the international trading system. Economists and management scientists have extensively researched the factors that determine the boundaries of MNCs. Most explanations are based on transaction cost analysis or resource-based view. However, these two explanations are driven by different logics: cost minimisation and revenue maximisation, respectively (Cuervo-Cazurra et al., 2018). This paper integrates transaction cost analysis and the resource base view to explain how intellectual property rights (IPR) protection and product life-cycle lengths affect the boundaries of MNCs. The model setup and derivation are presented based on a summary of existing literature. MNCs are significant players in US trade. The study uses US MNCs' trade data to validate the theory that U.S. overseas market penetration and sourcing are highly selective, and the main types of import sourcing (related-party and unrelated-party trade) vary by market (Co, 2010). The paper examines the inter-firm trade ratio as a measure of the propensity of MNCs to cross corporate borders.

When reviewing theories on the study of firm boundaries, the most commonly used are transaction cost analysis and the resource-based view (Brouthers and Hennart, 2007). These two different lines of thought have sparked a debate in academia as to whether the main driver behind a firm's competitive advantage is the minimization of transaction costs or the exploitation of resources to maximize revenues. The debate on the study of firm boundaries continues to pit cost minimization against revenue maximization. Several scholars are currently working on integrating transaction cost analysis and resource-based concepts (Silverman, 1999; Foss and Foss, 2005; Cuervo-Cazurra et al., 2018). The challenge of integration arises from the assumptions made. The transaction cost analysis assumes that firms have developed their own sources of advantage and therefore need to focus on minimizing operating costs. On the other hand, the resource-based view assumes that the transaction problem can be solved and focuses on ensuring that advantages and revenues are maximized. This paper aims to integrate transaction cost analysis and the resource-based view to construct a theoretical model for MNCs’ boundary choice.

Transaction cost analysis is a widely used theory in the study of international trade patterns. The transaction cost analysis framework assumes that decision making is influenced by three factors: asset specificity, uncertainty, and frequency (Williamson, 1985). Most of these studies use asset specificity as the central explanatory variable. Some studies based on transaction cost analysis have found that high asset specificity is associated with high control patterns (Gatignon and Anderson, 1988; Erramilli and Rao, 1993; Hennart and Larimo, 1998; Makino and Neupert, 2000; Brouthers and Brouthers, 2003; Brouthers et al., 2003). Integration is chosen for knowledge assets when asset specificity is high (Brouthers and Hennart, 2007). However, it has been found that the opposite conclusion is true in other studies (Delios and Beamish, 1999; Palenzuela and Bobilo, 1999; Grossman and Helpman, 2003). Some studies suggest that there is no significant relationship between asset specificity and the choice of international trade patterns (Hennart, 1991; Kim and Hwang, 1992; Taylor et al., 1998). The firm's R&D or advertising intensity (the sum of R&D and advertising expenditures divided by total sales) is a common measure of asset-specificity. Other studies have used perceived measures of asset-specific investments, including physical (service) asset specificity (Joskow, 1988; Klein and Roth, 1990; Erramilli and Rao, 1993; Klein et al., 1993; Brouthers and Brouthers, 2003; Brouthers et al., 2003;Wang et al., 2019), technology asset specificity (Taylor et al., 1998; Palenzuela and Bobilom, 1999), human asset specificity (Dyer, 1996; Handfield and Bechtel, 2002; Brouthers and Brouthers, 2003; Brouthers et al., 2003; Shakir and Ahmed, 2023), site asset specificity (Handfield and Bechtel, 2002; Williamson, 2007; Parker et al., 2019), and specialised asset specificity (Brouthers and Brouthers, 2003; Brouthers et al., 2003). One innovation of this paper is the use of the product life cycle length as a measure of asset specificity.

The second factor is uncertainty, which comprises of internal behaviour and external market idiosyncrasies. Internal uncertainty makes it difficult to verify performance ex post, while external uncertainty makes it difficult to specify all possible contingent events in the contract in advance. Previous research employs various measures to assess external uncertainty, including the Eurocurrency Country Risk Index (Delios and Beamish, 1999), Frost and Sullivan Country Risk Guide (Contractor and Kundu, 1998), industry growth (Kim and Hwang, 1992; Makino and Neupert, 2000; Luo, 2001), industry concentration (Kim and Hwang, 1992; Makino and Neupert, 2000), market size (Gomes-Casseres, 1989; Contractor and Kundu, 1998), perceived indicators of volatility and diversity in target markets (Klein et al., 1990; Luo, 2001), political and economic stability (Kim and Hwang, 1992; Luo, 2001; Brouthers, 2002; Brouthers and Brouthers, 2003; Brouthers et al., 2003) and perceived market potential (Brouthers, 2002). According to Williamson's model, uncertainty becomes problematic only when combined with asset specificity or higher switching costs. Uncertainty often leads to high levels of asset-specific integration (Leiblein and Miller, 2003). The prevalence of offshoring has increased due to better contractual regimes in host countries (Antràs and Helpman, 2006).The patent law of the host country is used as an indicator for measuring external uncertainty, based on the length of the product life cycle as a measure of asset specificity. This is one of the innovations of this paper. This paper examines the impact of product life cycle length (asset specificity) and patent law (uncertainty) on changes in the boundaries of the MNCs.

Resource-based view, including knowledge-based theory and organizational capability theory, suggest that firms develop unique resources that they can exploit in foreign markets or use foreign markets as a source for acquiring or developing new resource-based advantages (Madhok, 1997; Tsang, 2000; Luo, 2002). Firms can gain foundational advantages by developing or acquiring a set of firm-specific resources and capabilities that are valuable, rare, not fully imitable, and for which there are no universally available substitutes (Barney, 1991). Researchers have attempted to identify, measure, and test various resource-based advantages. For instance, Ekeledo and Sivakumar (2004) used multiple perceptual scales to measure firm-specific resources, including proprietary technology, tacit knowledge, corporate reputation, and international business experience.

Integrating transaction cost analysis and resource-based view in the analysis of MNCs’ boundaries refines and extends both approaches. The initial research argued that firm boundaries are driven by minimizing transaction costs, while later analyses proposed that firm boundaries are driven by developing and utilizing resources to maximize value creation and capture. To improve the discussion, we could consider (a) the link between asset specificity and uncertainty under transaction cost analysis, (b) the characteristics of relationships between headquarters and subsidiaries, as well as those between headquarters and unrelated-party outsourcing entities, from the perspective of the resource-based view, and (c) the differences in the institutional environments of home and host countries in the context of MNCs. The nature of economic systems is not independent of the environment, as they are based on certain social foundations that vary according to the local context (Meyer et al., 2011).

This paper analyses the influence of product life cycle length and patent law on the boundaries of MNCs. The study is based on transaction cost analysis and the resource-based view. The theoretical model is extrapolated and the results are validated using a real dataset. It is found that changes in the length of the product life cycle and patent law can account for the changes in the boundaries of MNCs between 2002 and 2013.

The article is structured as follows: section 2 presents the derivation of the theoretical model, section 3 provides the data and descriptive statistics, section 4 outlines the empirical model, section 5 presents the empirical results, section 6 conducts the robustness test, and the last section presents the conclusion.

2.1 Setup

There are two countries in the world, South (S) and North (N) (Antras and Helpman, 2004; Bilir, 2014). According to transaction cost analysis and the resource base view, it is assumed that the R&D component of the MNCs are located in the home country, while the production component is established in the host country through both affiliates and unrelated-party firms. Time is continuous and products are continuously differentiated. Firms produce a variety of products continuously and sell them to consumers in the South and North countries, which are symmetric countries with a monopolistically competitive market of size 1. MNCs incur a one-time fixed cost that is specific to the product for its development and gain exclusive rights over the product until it is either imitated or eliminated. The rate of product obsolescence is determined by idiosyncratic technological developments and is exogenous to individual firms. Therefore, the length of the life cycle T varies from product to product, but the length of the life cycle of a similar product i is the same. Once a product type reaches T years of maturity, it becomes obsolete and loses economic value for consumers. The associated intellectual property rights also become obsolete.

The rate of new product entry is constant and exogenous, equal to the rate of category obsolescence. This means that new products immediately replace obsolete ones, keeping the total number of products in each category constant. It is assumed that the total number of all products i is 1. It is assumed that the distribution of the life cycle length of the product maturity at any point in time is uniform with a density ${\psi }_{i}\left(t\right)=1/{T}_{i}$.

2.2 Production

MNCs typically have their headquarters located in the North, where they conduct R&D for specific product technologies. However, they may transfer production to subsidiaries in the South at no cost. MNCs carry out R&D in their home country and produce in the host country through subsidiaries or unrelated-party companies.The firm earns flow-through profits from production in the North ${\pi }^{N}$ and the South ${\pi }^{S}$ as a monopolist. It is assumed that economic conditions favour production in the South ${\pi }^{S}>{\pi }^{N}$. These conditions may include differences in regulatory costs, trade costs, and factor prices between the South and North. The objective of profit maximisation, where each firm earns revenue ${r}^{c}=\rho {\pi }^{c}$ when producing in country c, $\rho >1$ is the elasticity of demand faced by all firms. Therefore, firms earn higher revenues when they produce in Southern countries ${r}^{S}>{r}^{N}$.

Imitation can undermine the monopoly power of MNCs. To produce a product, imitators require access to specific information (Arora, 1996; Cohen, Nelson and Walsh, 2000). While MNCs conducting R&D protect proprietary information, they must disclose the information necessary for production (both proprietary and implicit) to their employees. This paper assumes that when production is outsourced to a unrelated party overseas, imitators gain access to undisclosed information about a particular product at a reduced cost. Additionally, the cost of trade secrets obtained across borders is high, which economically constrains imitators to produce locally.

When imitators enter a market, it can result in a loss of profits for the MNCs conducting the R&D; therefore, MNCs are cautious about their location. In the presence of imitators, MNCs conducting R&D will only receive a portion of the previously established per-period profits, which is dependent on the quality of local patent enforcement ${\lambda }_{c}$. With reference to the Bilir (2014) setting, ${\lambda }_{c}$ represents the probability or proportion of patents being enforced in country c at any given point in time. Assume that patents are fully protected in the North ${\lambda }_{N}=1$, but not in the South $0<{\lambda }_{S}<1$. Imitation products only appear in the South where patents are not fully protected. It is important to note that imitators are a direct result of inadequate patent protection.

2.3 Imitation

There are potential imitators in the South, but it is uncertain how long it will take for them to successfully imitate $w=\left[0,\stackrel{-}{w}\right]$. The success rate follows a Poisson distribution and is constant, while w is uniformly distributed (Grossman and Helpman, 1991; Glass and Saggi, 2002; Bilir, 2014). Imitation is a low-cost option, and outsourcing production to unrelated-party overseas companies immediately increases the risk of imitation.

2.4 The Product Cycle

MNCs choose locations with the aim of maximising profits, considering the actions of imitators in the southern region. In particular, MNCs relocate production to the South to achieve their profit goals and determine the optimal product maturity ${t}_{i}^{\ast }\in \left[0,{T}_{i}\right]$ for industry i to manufacture in the host country.

$${E}_{w}\left[{\varPi }_{i}\left(t\right)\right]=2{\pi }^{N}t+2{\pi }^{S}{E}_{w}\left[\underset{}{\text{min}}\left\{{T}_{i}-t,w\right\}\right]+\left(1+{\lambda }_{S}\right){\pi }^{S}{E}_{w}\left[\underset{}{\text{max}}\left\{0,{T}_{i}-t-w\right\}\right]$$

Optimizing over the initial offshoring maturity t, it is apparent that

$${\tau }^{\ast }\left({\lambda }_{S}\right)\equiv {T}_{i}-{t}^{\ast }=\frac{{\pi }^{S}-{\pi }^{N}}{\left(1-{\lambda }_{S}\right)}2\stackrel{-}{w}$$

depends on ${\lambda }_{S}$,innovators’ ${\pi }^{N}$ and ${\pi }^{S}$,and initators’ $\stackrel{-}{w}$.Notice that firms manufacture in the South earlier in the product life cycle when Southern patents are better protected:$\frac{\partial {\tau }^{\ast }\left({\lambda }_{S}\right)}{\partial {\lambda }_{S}}>0$.

Proposition 1

When the lifecycle length of product i,, the product is manufactured in the South. When the lifecycle length of product i,, the product is initially manufactured in the North, it gradually approaches the phase-out time, at which point it migrates to the South until it is phased out. The initial offshore production time cut-off increases with the degree of patent protection in the South: .

Proposition 1

states that product i is produced in the North at time and in the South at time. This means that the inter-firm trade ratio of products i in the South at any give can be measured as:

$${S}_{i}\left({\lambda }_{S}\right)\equiv {\int }_{{t}_{i}^{\ast }\left({\lambda }_{S}\right)}^{{T}_{i}}{\psi }_{i}\left(t\right)dt=\frac{{T}_{i}-{t}_{i}^{\ast }\left({\lambda }_{S}\right)}{{T}_{i}}=\underset{}{\text{min}}\left\{1,\frac{{\tau }^{\ast }\left({\lambda }_{S}\right)}{{T}_{i}}\right\}$$

${T}_{i}$ exhibits a weakly decreasing trend, with longer life cycle products being manufactured more in the North than in the South. The production quantity of product i in the South decreases as the product life cycle lengthens: $\partial {S}_{i}\left({\lambda }_{S}\right)/\partial {T}_{i}\le 0$.

Figure 2 shows the inter-firm trade ratio of product i in the South ${S}_{i}\left({\lambda }_{S}\right)$ (vertical axis) plotted against its life-cycle length ${T}_{i}$ (horizontal axis) ranges between ${T}_{min}<{\tau }^{\ast }\left({\lambda }_{S}\right)<{T}_{max}$. Two different levels of patent protection in the South, ${\lambda }_{S}$ (solid lines) and ${\lambda }_{S}^{{\prime }}$ (dashed lines).

The distribution of revenues earned in the South across products i is also determined by the initial offshoring maturity ${\tau }^{\ast }\left({\lambda }_{S}\right)$,

$${R}_{i}\left({\lambda }_{S}\right)={\int }_{\underset{}{\text{max}}\left\{0,{T}_{i}-{\tau }^{\ast }\left({\lambda }_{S}\right)\right\}}^{{T}_{i}}\left(2{r}^{S}\left(1-p\left(t\right)\right)+\left(1+{\lambda }_{S}\right){r}^{S}p\left(t\right)\right){\psi }_{i}\left(t\right)dt$$

The given formula calculates the revenues earned in the South, where ${\psi }_{i}\left(t\right)=1/{T}_{i}$ is the density of product maturity and $p\left(t\right)$ is the probability of a product with maturity t being imitated are the variables. It is important to note that a certain number of products are imitated at any given time, $p\left(t\right)=\frac{t}{\stackrel{-}{w}}$. This means that the revenue of the product in the southern region is affected.

$${R}_{i}\left({\lambda }_{S}\right)=2{r}^{S}\left(1-\frac{{T}_{i}}{2\stackrel{-}{w}}\right)+\left(1+{\lambda }_{S}\right){r}^{S}\frac{{T}_{i}}{2\stackrel{-}{w}}, {T}_{i}<{\tau }^{\ast }\left({\lambda }_{S}\right)$$

$${R}_{i}\left({\lambda }_{S}\right)=2{r}^{S}\left(\frac{{\tau }^{\ast }\left({\lambda }_{S}\right)}{{T}_{j}}-\frac{{{\tau }^{\ast }\left({\lambda }_{S}\right)}^{2}}{2\stackrel{-}{w}{T}_{j}}\right)+\left(1+{\lambda }_{S}\right){r}^{S}\frac{{T}_{i}}{2\stackrel{-}{w}}, {T}_{i}\ge {\tau }^{\ast }\left({\lambda }_{S}\right)$$

2.5 Cross-Product Response to IPR Protection

Suppose that the South enacts a policy change that increases the level of patent protection from ${\lambda }_{S}$ to ${\lambda }_{S}^{{\prime }}$, ${\lambda }_{S}^{{\prime }}>{\lambda }_{S}$. According to Eq. (2), this change would lead to firms offshoring production at an earlier stage of the product life cycle. Eq. (3) implies that the inter-industry trade ratio in the South is based on the difference in the degree of patent protection in the country ${\lambda }_{S}$, which is a segmented function based on ${T}_{i}$.

${S}_{i}\left({\lambda }_{S}^{{\prime }}\right)-{S}_{i}\left({\lambda }_{S}\right)=0$, ${T}_{i}<{\tau }^{\ast }\left({\lambda }_{S}\right)$

${S}_{i}\left({\lambda }_{S}^{{\prime }}\right)-{S}_{i}\left({\lambda }_{S}\right)=\frac{{T}_{i}-{\tau }^{\ast }\left({\lambda }_{S}\right)}{{T}_{i}}$, ${\tau }^{\ast }\left({\lambda }_{S}\right)\le {T}_{i}\le {\tau }^{\ast }\left({\lambda }_{S}^{{\prime }}\right)$

${S}_{i}\left({\lambda }_{S}^{{\prime }}\right)-{S}_{i}\left({\lambda }_{S}\right)=\frac{{\tau }^{\ast }\left({\lambda }_{S}^{{\prime }}\right)-{\tau }^{\ast }\left({\lambda }_{S}\right)}{{T}_{i}}$, ${T}_{i}>{\tau }^{\ast }\left({\lambda }_{S}^{{\prime }}\right)$ (5)

The generalised marginal effect ${S}_{i}\left({\lambda }_{S}^{{\prime }}\right)-{S}_{i}\left({\lambda }_{S}\right)$ is plotted in the middle section of Fig. 2. It represents a change in the inter-firm trade ratio of product i in the South immediately after the enhancement of patent protection. MNCs do not respond to enhanced patent protection if the life cycle of product i is shorter than the initial unrelated-party sourcing date, ${T}_{i}\le {\tau }^{\ast }\left({\lambda }_{S}\right)$. When the patent protection in the South is ${\lambda }_{S}$, the entire life cycle of product i is produced in the South, ${T}_{i}\le {\tau }^{\ast }\left({\lambda }_{S}\right)$; if the intensity of patent protection in the South is high, ${\lambda }_{S}^{{\prime }}$, product i continues to be produced there. However, for products with longer life cycles, ${T}_{i}>{\tau }^{\ast }\left({\lambda }_{S}\right)$, the MNCs outsource the marginal production chain to a unrelated party in the South for the remaining years before obsolescence, ${\tau }^{\ast }\left({\lambda }_{S}\right)\le {T}_{i}\le {\tau }^{\ast }\left({\lambda }_{S}^{{\prime }}\right)$. ${S}_{i}\left({\lambda }_{S}^{{\prime }}\right)-{S}_{i}\left({\lambda }_{S}\right)$ is a non-monotonic function depends on ${T}_{i}$. The inter-firm trade ratio in the South increases non-monotonically, ${S}_{i}\left({\lambda }_{S}^{{\prime }}\right)-{S}_{i}\left({\lambda }_{S}\right)$, with the degree of patent protection for product i after the patent reform from ${\lambda }_{S}$ to ${\lambda }_{S}^{{\prime }}$. The increase is zero for ${T}_{i}<{\tau }^{\ast }\left({\lambda }_{S}\right)$, then increases for ${\tau }^{\ast }\left({\lambda }_{S}\right)\le {T}_{i}\le {\tau }^{\ast }\left({\lambda }_{S}^{{\prime }}\right)$, and finally decreases for ${T}_{i}>{\tau }^{\ast }\left({\lambda }_{S}^{{\prime }}\right)$. The point of greatest impact is at ${T}_{i}={\tau }^{\ast }\left({\lambda }_{S}^{{\prime }}\right)$.

Revenues earned by MNCs are impacted by enhanced protection of intellectual property rights both because of newly shifted manufacturing and existing imitated varieties capture a larger share of Southern sales under the stronger patent regime. Based on equations (4) and (5), the change in product i revenues earned offshore following a Southern enhanced protection of intellectual property rights from ${\lambda }_{S}$ to ${\lambda }_{S}^{{\prime }}$ depends on ${T}_{i}$.

The patent reform in South has impacted the revenues of MNCs. This is due to the shift of South production from subsidiaries to unrelated parties and the emergence of South imitations as unrelated-party outsourcing begins to take up a part of the MNCs' earnings. Equations (4) and (5) indicate that the change in revenues affected by offshore production depends on ${T}_{i}$, which is affected by the change in patent protection in South from enhancement from ${\lambda }_{S}$ to ${\lambda }_{S}^{{\prime }}$.

${R}_{i}\left({\lambda }_{S}^{{\prime }}\right)-{R}_{i}\left({\lambda }_{S}\right)={r}^{S}\left({\lambda }_{S}^{{\prime }}-{\lambda }_{S}\right)\frac{{T}_{i}}{2\stackrel{-}{w}}$, ${T}_{i}<{\tau }^{\ast }\left({\lambda }_{S}\right)$

${R}_{i}\left({\lambda }_{S}^{{\prime }}\right)-{R}_{i}\left({\lambda }_{S}\right)={r}^{S}\left(2\frac{{T}_{j}-{\tau }^{\ast }\left({\lambda }_{S}\right)}{{T}_{j}}+\left({\lambda }_{S}^{{\prime }}-{\lambda }_{S}\right)\frac{{\left[{\tau }^{\ast }\left({\lambda }_{S}\right)\right]}^{2}}{2\stackrel{-}{m}{T}_{i}}\right)$, ${\tau }^{\ast }\left({\lambda }_{S}\right)\le {T}_{i}\le {\tau }^{\ast }\left({\lambda }_{S}^{{\prime }}\right)$

${R}_{i}\left({\lambda }_{S}^{{\prime }}\right)-{R}_{i}\left({\lambda }_{S}\right)={r}^{S}\left(2\frac{{\tau }^{\ast }\left({\lambda }_{S}^{{\prime }}\right)-{\tau }^{\ast }\left({\lambda }_{S}\right)}{{T}_{j}}+\left({\lambda }_{S}^{{\prime }}-{\lambda }_{S}\right)\frac{{\left[{\tau }^{\ast }\left({\lambda }_{S}\right)\right]}^{2}}{2\stackrel{-}{m}{T}_{i}}\right)$, ${T}_{i}>{\tau }^{\ast }\left({\lambda }_{S}^{{\prime }}\right)$ (6)

The function at the bottom of Fig. 2 is ${R}_{i}\left({\lambda }_{S}^{{\prime }}\right)-{R}_{i}\left({\lambda }_{S}\right)$ a non-monotonic function based on ${T}_{i}$ that increases for ${T}_{i}\le {\tau }^{\ast }\left({\lambda }_{S}^{{\prime }}\right)$ and decreases for ${T}_{i}>{\tau }^{\ast }\left({\lambda }_{S}^{{\prime }}\right)$, the point of greatest response to IPR protection reform at ${T}_{i}={\tau }^{\ast }\left({\lambda }_{S}^{{\prime }}\right)$.

3.1 Product Life-Cycle Lengths

To measure the length of a product's life cycle ${T}_{i}$, it is to associate the product with proprietary information such as patents and implicit intellectual property rights that are necessary for production but lose their economic value when the product becomes obsolete (Bilir, 2014). The view implies that proprietary information must be separated across products, meaning that the information needed to create a new product must differ from that required for any previous product. Therefore, the main distinction between industries in the model is not the time lag between versions of a product that embody the same innovative idea, but rather the economic lifecycle of the innovative idea itself. This lifecycle may span consecutive versions of the product's lifecycle (Klepper, 1996).

The length of a product's life cycle is determined by the economic durability of the embedded technology. This paper uses United States patent citation data from 1976 to 2006, which includes 6,832 products from 122 countries. The duration for which a patent is cited by subsequent patents is an indicator of the market longevity of the patented technology. A long average 'pre-citation lag', which is the time between the grant date of the cited patent and the subsequent citation, indicates that the technology has persistent relevance to future innovation. To construct the product life cycle length index ${\widehat{T}}_{i}$, we used the methodology of Hall et al. (2001) and Bilir (2014) and mapped from SIC 3-digit code to NAICS 6-digit code. The product life-cycle length index is constructed without any restrictions on the patent categories in which patents are cited. This allows patented technologies to be relevant not only to future innovations in their own category but also in other patent categories.

Figure 3 shows a bar chart of the Product Life Cycle Length Index, highlighting the differences in life cycle length among various products. Table 1 presents the product name, life cycle length, and capital intensity for the top five, median, and bottom five product life cycle lengths. Table 1 shows that products with shorter life cycles have higher capital intensity. This relationship is further supported by the scatter plot in Fig. 4, which has a correlation coefficient of -0.072.

Table 1

Product Life-cycle Lengths and Capital Intensity
Name	Rank	SIC 3-Digit Code	Product Life-cycle Lengths	Capital Intensity
Shortest product life cycle
Electronics machinery	1	383	6.73	3.592
Watches, clocks, clockwork operated devices	2	387	7.37	5.691
Computer and office equipment	3	357	8.38	6.891
Agricultural chemicals	4	287	8.69	7.072
Electronic components and accessories	5	367	8.83	4.566
Intermediate-length product life cycles
Miscellaneous industrial and commercial	19	359	9.68	2.275
Miscellaneous chemical products	20	289	9.73	3.775
Surgical, medical, dental instruments, and supplies	21	384	9.75	1.452
Farm and garden machinery and equipment	22	352	9.78	2.418
Household appliances	23	363	9.78	2.592
Longest product life cycles
Fabricated structural metal products	33	344	10.25	1.182
Cutlery, handtools, and general hardware	34	342	10.41	2.787
Screw machine products, bolts, nuts, screws	35	345	10.42	1.718
Metal cans and shipping containers	36	341	10.63	3.320
Heating equipment, except electric	37	343	10.89	2.187
Notes: This table presents the top five, median, and bottom five product names, life cycle lengths, and capital intensities sorted by product life cycle length. Product life-cycle lengths were measured in years and calculated using patent citation data.

3.2 Intellectual Property Protection in Different Countries

This paper utilises Park's (2008) indicator to measure the strength of IPR protection across different countries. The indicator, which is comprehensive and widely applicable, assesses IPR protection in five dimensions: coverage, membership of international patent agreements, provisions for loss of patent, enforcement mechanisms, and duration of protection. Each dimension is scored between 0 and 1 based on objective criteria to determine whether the current patent law meets specific standards. The Patent Protection Index is the sum of these five dimensions, ranging from 0 to 5. Higher values indicate stronger IPR protection. These indices are available at five-year intervals for 122 countries from 1980 to 2005. In this paper, the year of each benchmarking survey is matched to the closest available index year.

3.3 Inter-firm Trade Ratio

This paper utilises US MNCs trade data from 2002–2013 for 118 countries import data for 208 products based on the NAICS 6-digit classification. Table 2 presents summary statistics for the aforementioned data.

Table 2

Summary Statistics
Variables	Obs.	Mean	S. D.	Min	Max.
IFR	229,089	0.194	0.309	0	1
T	229,089	9.523	0.725	6.681	13.500
IPR	229,089	3.447	0.835	0.200	4.800
Capital Intensity	229,089	3.172	2.636	0.989	28.929
GDP per capita	229,089	18697.02	17824.27	0	95012.67

Based on Fig. 2 in the second part of the theory, the safeguarding of host country IPR protection through offshore production activities of MNEs is a non-monotonic function of the product life cycle's length. This paper uses inter-firm trade ratios to characterise the corporate boundaries of MNCs (Antràs, 2003; Antràs and Helpman, 2004; Antràs and Helpman, 2006; Nunn and Trefler, 2008). The baseline regression equation of this paper is based on the results of the theoretical derivation.

$${IFR}_{cit}={\beta }_{0}+{\beta }_{1}\bullet {IPR}_{ct}\times {T}_{i}+{\beta }_{2}\bullet {IPR}_{ct}\times {T}_{i}^{2}+{\eta }_{ct}+{\eta }_{i}+{ϵ}_{cit}$$

where ${IFR}_{cit}$ is the inter-firm trade ratio of U.S. MNCs in country c for product i in year t. ${IPR}_{ct}$ is an index of patent protection in country c in year t. ${T}_{i}$ denotes the life cycle length of product i. The main coefficients of interest in this paper are ${\beta }_{1}$ and ${\beta }_{2}$, which shows the impact of different product life-cycle lengths and host country IPR protection on differences in MNCs' offshore production activities. In order to make the empirical model consistent with the theoretical model, the estimated coefficients of the empirical model must satisfy the following conditions: ${\beta }_{1}>0$, ${\beta }_{2}<0$, ${\beta }_{1}\bullet {T}_{i}+{\beta }_{2}\bullet {T}_{i}^{2}\ge 0$. The country-level control variable ${\eta }_{ct}$ controls for GDP, Log GDPpc×T and $log GDPpc\times {T}^{2}$, while the product-level control variable ${\eta }_{i}$ controls for capital intensity, $IPR\times Capital intensity$ and $IPR\times {Capital intensity}^{2}$. The error term includes any omitted factors that affect the offshore production patterns of MNCs. The results are clustered at the country level across all reports, which is robust due to the possibility of measurement error in the host country patent protection index.

5.1 Baseline Results

Table 3 presents the main regression results, which provide favourable support for the above theoretical results. Column 1 shows that stronger IPR protection leads to a higher rate of MNCs choosing unrelated-party outsourcing in the production chain in host countries. Column 2 shows that host countries with stronger IPR protection have a higher rate of choosing unrelated-party outsourcing for products with longer life cycles. In column 3, the empirical model includes the interaction between ${IPR}_{ct}$ and ${T}_{i}^{2}$ based on the results of the theoretical model. The empirical results show that the coefficient of IPR×T is 0.133 > 0, while the coefficient of $IPR\times {T}^{2}$ is -0.129 < 0, consistent with the theoretical model. The impact of IPR protection in the host country follows a non-monotonic trend with respect to the length of the product life cycle. The effect is most pronounced in industries with moderately long product life cycles. Column 4 includes $IPR\times Capital intensity$ and $IPR\times {Capital intensity}^{2}$ in the model from column 3 to differentiate the impact of product life cycle length from that of product idiosyncrasies. Additionally, column 4 introduces Log GDPpc×T and $log GDPpc\times {T}^{2}$ to the model from column 3 to better distinguish the effect of IPR protection from that of overall economic development. Column 6 includes all control variables. The empirical regression results are consistent with the theoretical notation, indicating that the addition of control variables has a minimal impact on the main coefficients of interest ${\beta }_{1}$ and ${\beta }_{2}$ in this paper.

Table 3

Main Result
		Inter-firm Trade Ratio
	(1)	(2)	(3)	(4)	(5)	(6)
IPR	0.108*** (0.017)
IPR×T		-0.029 (0.021)	0.133*** (0.029)	0.133*** (0.029)	0.098*** (0.025)	0.102*** (0.028)
$IPR\times {T}^{2}$			-0.129*** (0.022)	-0.129*** (0.022)	-0.095*** (0.023)	-0.099*** (0.023)
$IPR\times Capital intensity$				0.000 (0.002)		0.001 (0.002)
$IPR\times {Capital intensity}^{2}$				-0.000 (0.000)		-0.000 (0.000)
GDPpc	0.028** (0.011)
Log GDPpc×T					0.019* (0.011)	0.018 (0.011)
$log GDPpc\times {T}^{2}$					-0.001* (0.001)	-0.001* (0.001)
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes
Country-year FE	No	Yes	Yes	Yes	Yes	Yes
Obs.	229,089	229,089	229,089	229,089	229,089	229,089
$Pesudo {R}^{2}$	0.204	0.332	0.333	0.333	0.333	0.333
Notes: Standard errors in brackets clustered at country level. *Significant at the 1 percent level. Significant at the 5 percent level. *Significant at the 10 percent level.

5.2 Flexible Estimation

To evaluate the influence of IPR protection in the host country on inter-firm trade ratio, we estimate coefficients that vary flexibly over the product life cycle's length ${T}_{i}$. Table 4 presents the estimates based on Eq. (8), where the product life cycle length is divided into quartiles (${Q}_{1}^{T}$,${Q}_{2}^{T}$,${Q}_{3}^{T}$,${Q}_{4}^{T}$,${Q}_{5}^{T}$ and ${Q}_{6}^{T}$). Dummy variables corresponding to the last five quintiles intersect with the degree of IPR in the host country, with interactions as shown in Eq. (8).

$${{IF}R}_{{cit}}=\beta +\sum _{k=2}^{6}{\beta }_{k}\bullet {IPR}_{ct}\times {1}_{{T}_{i}\in {Q}_{k}^{T}}+{\eta }_{ct}+{\eta }_{i}+{ϵ}_{cit}$$

so that $\beta$ and fixed effects ${\eta }_{ct}$ capture the effect of IPR on short-life-cycle sectors in the bottom fifth of the ${T}_{i}$ distribution. An advantage of this approach is that the differential effect of IPR protection, as reflected by the coefficients ${\beta }_{2}$, ${\beta }_{3}$, ${\beta }_{4}$, ${\beta }_{5}$ and ${\beta }_{6}$, is unrestricted across fifths of the ${T}_{i}$ distribution. A consistent pattern emerges from the estimates in Table 4: the fourth and fifth sextile coefficients tend to be negative and smaller than the second or third sextile coefficients. This pattern indicates that shrink in inter-firm trade ratio is concentrated within posterior-${T}_{i}$ industries. This pattern provides additional evidence consistent with the non-monotonicity implications of the theory.

Thus, the combination of $\beta$ and fixed effects ${\eta }_{ct}$ explains that host country ${IPR}_{ct}$ impact on short life cycle lengths ${T}_{i}$ in the bottom quartile of the distribution. This approach allows for coefficients ${\beta }_{2}$, ${\beta }_{3}$, ${\beta }_{4}$, ${\beta }_{5}$ and ${\beta }_{6}$ to reflect the differential impact of IPR protection in different quartiles of the product life cycle length ${T}_{i}$ distribution. The coefficients in Table 4 tend to be positive for the fourth and fifth quintiles. The data indicates that inter-firm trade has increased in the middle and late products of the life-cycle length quartiles. Additionally, the expansion of unrelated-party outsourcing by MNCs in offshore production segments has been concentrated in products with medium- to long-term life-cycle lengths. These findings provide further evidence consistent with the non-monotonicity of the results of the theoretical model.

Table 4

Host-country IPR and Inter-firm Trade Ratio Flexible Estimation
	Inter-firm Trade Ratio
	(1)	(2)	(3)	(4)
$IPR\times {T}_{2}$	−0.002 (0.005)	−0.001 (0.004)	−0.001 (0.004)	−0.000 (0.004)
$IPR\times {T}_{3}$	−0.001 (0.004)	−0.001 (0.004)	−0.000 (0.004)	0.001 (0.004)
$IPR\times {T}_{4}$	0.003 (0.005)	0.003 (0.005)	0.005 (0.006)	0.005 (0.006)
$IPR\times {T}_{5}$	0.004 (0.005)	0.005 (0.005)	0.007 (0.006)	0.007 (0.005)
$IPR\times {T}_{6}$	−0.011* (0.006)	−0.010** (0.005)	−0.004* (0.006)	−0.004 (0.005)
$IPR\times Capital intensity$		0.002 (0.002)		0.002 (0.002)
$IPR\times {Capital intensity}^{2}$		−0.000** (0.000)		−0.000** (0.000)
Log GDPpc×T			0.018 (0.012)	0.018 (0.012)
$log GDPpc\times {T}^{2}$			−0.001* (0.001)	−0.001* (0.001)
Industry FE	Yes	Yes	Yes	Yes
Country-year FE	Yes	Yes	Yes	Yes
Obs.	229,089	229,089	229,089	229,089
${R}^{2}$	0.332	0.332	0.332	0.333
Notes: Standard errors in brackets clustered at country level. *Significant at the 1 percent level. Significant at the 5 percent level. *Significant at the 10 percent level.

6.1 Tobit Regression

To ensure the robustness of the regression results, the econometric model was replaced with Tobit regressions based on theoretical model extrapolation. Table 5 reports the impact of host country IPR protection and product life-cycle length on inter-firm trade based on Tobit regressions. The results obtained through Tobit regressions are consistent with those obtained through OLS regression, indicating the robustness of the regression results.

Table 5

Tobit Regression
		Inter-firm Trade Ratio
	(1)	(2)	(3)	(4)	(5)	(6)
IPR	0.140*** (0.036)
IPR×T		-0.013 (0.013)	0.111** (0.044)	0.204*** (0.062)	0.080 (0.054)	0.154** (0.069)
$IPR\times {T}^{2}$			-0.108** (0.045)	-0.178*** (0.054)	-0.080 (0.056)	-0.136** (0.064)
$IPR\times Capital intensity$				-0.010* (0.005)		-0.007 (0.005)
$IPR\times {Capital intensity}^{2}$				0.000 (0.000)		0.000 (0.000)
GDPpc	0.000** (0.000)
Log GDPpc×T					0.091** (0.038)	0.077** (0.039)
$log GDPpc\times {T}^{2}$					-0.004** (0.002)	-0.004** (0.002)
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes
Country-year FE	No	Yes	Yes	Yes	Yes	Yes
Obs.	136,918	139,596	139,596	139,596	136,918	136,918
${R}^{2}$	0.110	0.230	0.230	0.231	0.231	0.231
Notes: Standard errors in brackets clustered at country level. *Significant at the 1 percent level. Significant at the 5 percent level. *Significant at the 10 percent level.

6.2 Heterogeneity Analysis

Table 6 presents estimates based on whether the host country's IPR protection is above or below its mean (3.447). The results align with expectations, indicating that the IPR protection sensitivity mechanism is more effective in countries with high IP protection. Columns 1 to 5 pertain to the high IPR protection sample. The regression results indicate significantly positive coefficients for IPR×T and significantly negative coefficients for $IPR\times {T}^{2}$, which are consistent with the main regression results. However, in the low IPR protection sample, the direction of the coefficients of the regression results is not significant, although it is consistent with the main regression results. The data suggests that the mechanism for protecting intellectual property rights is more effective in countries with stronger IPR protection.

Table 7 distinguishes between short, medium, and long product life cycle lengths based on the trichotomy of the product life cycle length. The results indicate that inter-firm trade ratio is more sensitive in samples with longer product life cycles. Specifically, the coefficients of IPR×T are significantly positive and those of $IPR\times {T}^{2}$ are significantly negative for the sample with longer product life cycles in columns 7 to 10. This is consistent with the results of the main regression. The study shows that long life-cycle products tend to have a higher rate of unrelated-party outsourcing in the host country when they are far from the phase-out deadline. The rate of outsourcing decreases after reaching a peak and gradually approaches the phase-out date. However, for short and medium product life cycles, the coefficients show the opposite sign to that of long product life cycles and are not significant. The reason for the opposite sign may be that the risk of imitation for products with shorter and medium life cycles is relatively lower than that of products with longer life cycles. As a result, products with shorter and medium life cycles tend to outsource to unrelated parties at the initial stage, which is in line with the expected theoretical results. The results indicate that the rate of unrelated-party outsourcing in the host country is explained by both host country IPR protection and product life-cycle length. The relationship between these factors is non-linear, and the rate of host country unrelated-party outsourcing is more sensitive for products with long life-cycle lengths.

Table 6

Heterogeneity of IPR
	High Patent Protection					Low Patent Protection
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)	(10)
IPR×T	−0.917** (0.079)	0.916 (0.593)	1.377** (0.599)	1.202* (0.670)	1.659** (0.652)	−0.021 (0.041)	0.625 (0.387)	0.525 (0.400)	0.673* (0.397)	0.565 (0.415)
$IPR\times {T}^{2}$		−0.547* (0.280)	−0.722** (0.280)	−0.657** (0.317)	−0.831*** (0.311)		−0.321* (0.181)	−0.285 (0.185)	−0.335* (0.184)	−0.296 (0.191)
$IPR\times Capital intensity$			0.016** (0.008)		0.016* (0.008)			−0.002 (0.002)		−0.002 (0.002)
$IPR\times {Capital intensity}^{2}$			−0.001** (0.000)		−0.001** (0.000)			−0.000 (0.000)		−0.000 (0.000)
Log GDPpc×T				−0.015 (0.010)	−0.015 (0.010)				−0.003 (0.008)	−0.002 (0.008)
$log GDPpc\times {T}^{2}$				0.001 (0.000)	0.001 (0.000)				0.000 (0.000)	0.000 (0.000)
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Country-year FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Obs.	113,179	113,179	113,179	113,179	113,179	115,910	115,910	115,910	115,910	115,910
${R}^{2}$	0.272	0.272	0.272	0.272	0.272	0.262	0.262	0.262	0.262	0.262
Notes: Standard errors in brackets clustered at country level.*Significant at the 1 percent level; Significant at the 5 percent level; *Significant at the 10 percent level.

Table 7

Heterogeneity in Product Life Cycle Length
	Short-length Product Life Cycle			Intermediate-length Product Life Cycle			Long-length Product Life Cycle
	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
IPR×T	−0.081 (0.060)	−0.083 (0.073)	−0.154* (0.085)	−0.240 (0.227)	−0.191 (0.224)	−0.069 (0.248)	0.222*** (0.033)	0.212*** (0.035)	0.186*** (0.037)
$IPR\times {T}^{2}$	0.088 (0.067)	0.093 (0.076)	0.173* (0.089)	0.250 (0.238)	0.174 (0.227)	0.047 (0.257)	−0.215*** (0.033)	−0.209*** (0.032)	−0.184*** (0.033)
$IPR\times Capital intensity$		0.000 (0.003)	0.000 (0.003)		0.013 (0.011)	0.012 (0.011)		0.003 (0.006)	0.003 (0.006)
$IPR\times {Capital intensity}^{2}$		−0.000 (0.000)	−0.000 (0.000)		−0.001 (0.002)	−0.001 (0.002)		−0.000 (0.001)	−0.000 (0.001)
Log GDPpc×T			0.019 (0.035)			−3.799 (1.783)			0.009 (0.043)
$log GDPpc\times {T}^{2}$			−0.001 (0.002)			0.200 (0.093)			−0.001 (0.002)
Industry FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Country-year FE	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Obs.	76,363	76,363	76,363	76,363	76,363	76,363	76,363	76,363	76,363
${R}^{2}$	0.322	0.323	0.323	0.363	0.363	0.363	0.343	0.343	0.343
Notes: Standard errors in brackets clustered at country level.*Significant at the 1 percent level; Significant at the 5 percent level; *Significant at the 10 percent level.

This paper explores the sensitivity of MNCs to host country IPR protection regimes from both theoretical and empirical perspectives. The firm's global production location decision model applies two dimensions, namely product life cycle length and the degree of patent protection in the host country, to explain MNCs' inter-firm trade.

The panel data of import trade for 208 types of products in 18 countries by U.S. MNCs from 2002 to 2013 strongly support the empirical results. The rate of unrelated-party outsourcing by MNCs in host countries, as a measure of changes in MNCs' corporate boundaries, can be explained by exploiting the interaction between patent law and product life cycle length. The research indicates that variations in the duration of the life cycle of products play a significant role in determining the sensitivity of MNCs to IPR protection in host countries. Additionally, it establishes a causal link between IPR protection and inter-firm trade of MNCs.

The paper's heterogeneity analyses demonstrate that the sensitivity of the IPR protection mechanism is more effective in countries with a high degree of IPR protection. Additionally, the sensitivity to IPR protection in the host country is concentrated in products with a longer life cycle. The results derived from the theoretical model are consistent with reality.

Finally, this paper's theoretical and empirical results support the notion that the length of a product's life cycle determines the degree of dependence of MNCs on IPR protection in host countries. It is crucial to note that the sophistication of IPR protection in host countries is particularly important for products with longer life cycles from a policy perspective.

Antràs, P. (2003). Firms, contracts, and trade structure. The Quarterly Journal of Economics, 118(4), 1375–1418.
Antras, P., & Helpman, E. (2004). Global sourcing. Journal of political Economy, 112(3), 552–580.
Antràs, P., & Helpman, E. (2006). Contractual frictions and global sourcing.
Argyres, N. S., & Zenger, T. R. (2012). Capabilities, transaction costs, and firm boundaries. Organization Science, 23(6), 1643–1657.
Arora, A. (1996). Contracting for tacit knowledge: the provision of technical services in technology licensing contracts. Journal of Development Economics, 50(2), 233–256.
Bilir, L. K. (2014). Patent laws, product life-cycle lengths, and multinational activity. American Economic Review, 104(7), 1979–2013.
Brouthers, K. D. (2002). Institutional, cultural and transaction cost influences on entry mode choice and performance. Journal of international business studies, 33, 203–221.
Brouthers, K. D., & Brouthers, L. E. (2003). Why service and manufacturing entry mode choices differ: The influence of transaction cost factors, risk and trust. Journal of management studies, 40(5), 1179–1204.
Brouthers, K. D., & Hennart, J. F. (2007). Boundaries of the firm: Insights from international entry mode research. Journal of management, 33(3), 395–425.
Brouthers, K. D., Brouthers, L. E., & Werner, S. (2003). Transaction cost-enhanced entry mode choices and firm performance. Strategic management journal, 24(12), 1239–1248.
Co, C. Y. (2010). Intra-and inter-firm US trade. International Review of Economics & Finance, 19(2), 260–277.
Cohen, W. M., Nelson, R., & Walsh, J. P. (2000). Protecting their intellectual assets: Appropriability conditions and why US manufacturing firms patent (or not).
Contractor, F. J., & Kundu, S. K. (1998). Modal choice in a world of alliances: Analyzing organizational forms in the international hotel sector. Journal of international business studies, 29, 325–356.
Cuervo-Cazurra, A., Mudambi, R., & Pedersen, T. (2018). The boundaries of the firm in global strategy. Global Strategy Journal, 8(2), 211–219.
Delios, A., & Beamish, P. W. (1999). Ownership strategy of Japanese firms: Transactional, institutional, and experience influences. Strategic management journal, 20(10), 915–933.
Dyer, J. H. (1996). Does governance matter? Keiretsu alliances and asset specificity as sources of Japanese competitive advantage. Organization science, 7(6), 649–666.
Erramilli, M. K., & Rao, C. P. (1993). Service firms’ international entry-mode choice: A modified transaction-cost analysis approach. Journal of marketing, 57(3), 19–38.
Foss, K., & Foss, N. J. (2005). Resources and transaction costs: How property rights economics furthers the resource-based view. Strategic Management Journal, 26(6), 541–553.
Gatignon, H., & Anderson, E. (1988). The multinational corporation's degree of control over foreign subsidiaries: An empirical test of a transaction cost explanation. The Journal of Law Economics and Organization, 4(2), 305–336.
Geyskens, I., Steenkamp, J. B. E., & Kumar, N. (2006). Make, buy, or ally: A transaction cost theory meta-analysis. Academy of management journal, 49(3), 519–543.
Glass, A. J., & Saggi, K. (2002). Licensing versus direct investment: implications for economic growth. Journal of International Economics, 56(1), 131–153.
Gomes-Casseres, B. (1989). Ownership structures of foreign subsidiaries: Theory and evidence. Journal of Economic Behavior & Organization, 11(1), 1–25.
Grossman, G. M., & Helpman, E. (2003). Outsourcing versus FDI in industry equilibrium. Journal of the European Economic Association, 1(2–3), 317–327.
Hall, B. H., Jaffe, A. B., & Trajtenberg, M. (2001). The NBER patent citation data file. Lessons, insights and methodological tools.
Handfield, R. B., & Bechtel, C. (2002). The role of trust and relationship structure in improving supply chain responsiveness. Industrial marketing management, 31(4), 367–382.
Hennart, J. F. (1991). The transaction costs theory of joint ventures: An empirical study of Japanese subsidiaries in the United States. Management science, 37(4), 483–497.
Hennart, J. F., & Larimo, J. (1998). The impact of culture on the strategy of multinational enterprises: does national origin affect ownership decisions? Journal of international business studies, 29, 515–538.
Holmström, B., & Roberts, J. (1998). The boundaries of the firm revisited. Journal of Economic perspectives, 12(4), 73–94.
Joskow, P. L. (1988). Asset specificity and the structure of vertical relationships: empirical evidence. The Journal of Law Economics and Organization, 4(1), 95–117.
Kim, W. C., & Hwang, P. (1992). Global strategy and multinationals' entry mode choice. Journal of international business studies, 23, 29–53.
Klein, B., Crawford, R. G., & Alchian, A. A. (1993). In L. Wrin, & H. Wijkander (Eds.), Contracts and incentives: the role of contract terms in assuring performance. Contract Economics.
Klein, S., Frazier, G. L., & Roth, V. J. (1990). A transaction cost analysis model of channel integration in international markets. Journal of Marketing research, 27(2), 196–208.
Klepper, S. (1996). Entry, exit, growth, and innovation over the product life cycle (pp. 562–583). The American economic review.
Leiblein, M. J., & Miller, D. J. (2003). An empirical examination of transaction-and firm‐level influences on the vertical boundaries of the firm. Strategic management journal, 24(9), 839–859.
Luo, Y. (2001). Determinants of entry in an emerging economy: A multilevel approach. Journal of management Studies, 38(3), 443–472.
Makino, S., & Neupert, K. E. (2000). National culture, transaction costs, and the choice between joint venture and wholly owned subsidiary. Journal of International Business Studies, 31, 705–713.
Meyer, K. E., Mudambi, R., & Narula, R. (2011). Multinational enterprises and local contexts: The opportunities and challenges of multiple embeddedness. Journal of management studies, 48(2), 235–252.
Nunn, N., & Trefler, D. (2008). The boundaries of the multinational firm: an empirical analysis. The organization of firms in a global economy, 55–83.
Palenzuela, V. A., & Bobillo, A. M. (1999). Transaction costs and bargaining power: Entry mode choice in foreign markets. Multinational Business Review, 7(1), 62.
Palenzuela, V. A., & Bobillo, A. M. (1999). Transaction costs and bargaining power: Entry mode choice in foreign markets. Multinational Business Review, 7(1), 62.
Park, W. G. (2008). International patent protection: 1960–2005. Research policy, 37(4), 761–766.
Parker, G. G., Tan, B., & Kazan, O. (2019). Electric power industry: Operational and public policy challenges and opportunities. Production and Operations Management, 28(11), 2738–2777.
Parker, G. G., Tan, B., & Kazan, O. (2019). Electric power industry: Operational and public policy challenges and opportunities. Production and Operations Management, 28(11), 2738–2777.
Shakir, A., & Ahmed, S. (2023). Are Asset Specificity and Legal Contracts Really Helpful in Enhancing the Level of Trust?-A Review. Journal of Management Sciences, 10(2), 67–79.
Silverman, B. S. (1999). Technological resources and the direction of corporate diversification: Toward an integration of the resource-based view and transaction cost economics. Management science, 45(8), 1109–1124.
Taylor, C. R., Zou, S., & Osland, G. E. (1998). A transaction cost perspective on foreign market entry strategies of US and Japanese firms. Thunderbird International Business Review, 40(4), 389–412.
Wang, L., Jiang, F., Li, J., Motohashi, K., & Zheng, X. (2019). The contingent effects of asset specificity, contract specificity, and trust on offshore relationship performance. Journal of Business Research, 99, 338–349.
Williamson, O. E. (2007). The economic institutions of capitalism. Firms, markets, relational contracting (pp. 61–75). Gabler.
Williamson, O. (1985). The Economic Institutions of Capitalism. Free Press.

Download PDF

Version 1

posted

You are reading this latest preprint version

Intellectual Property Protection, Product Life-cycle Lengths, and the Boundaries of Multinational Corporations

Status:

Version 1

Abstract

Figures

1 Introduction

2 Theory Model

2.1 Setup

2.2 Production

2.3 Imitation

2.4 The Product Cycle

2.5 Cross-Product Response to IPR Protection

3 Data and Descriptive Statistics

3.1 Product Life-Cycle Lengths

3.2 Intellectual Property Protection in Different Countries

3.3 Inter-firm Trade Ratio

4 Econometric Framework

5 Results

5.1 Baseline Results

5.2 Flexible Estimation

6 Robustness Checks and Heterogeneity Analysis

6.1 Tobit Regression

6.2 Heterogeneity Analysis

7 Conclusions

References

Status:

Version 1

	Inter-firm Trade Ratio
	(1)	(2)	(3)	(4)
\(IPR\times {T}_{2}\)	−0.002 (0.005)	−0.001 (0.004)	−0.001 (0.004)	−0.000 (0.004)
\(IPR\times {T}_{3}\)	−0.001 (0.004)	−0.001 (0.004)	−0.000 (0.004)	0.001 (0.004)
\(IPR\times {T}_{4}\)	0.003 (0.005)	0.003 (0.005)	0.005 (0.006)	0.005 (0.006)
\(IPR\times {T}_{5}\)	0.004 (0.005)	0.005 (0.005)	0.007 (0.006)	0.007 (0.005)
\(IPR\times {T}_{6}\)	−0.011* (0.006)	−0.010** (0.005)	−0.004* (0.006)	−0.004 (0.005)
\(IPR\times Capital intensity\)		0.002 (0.002)		0.002 (0.002)
\(IPR\times {Capital intensity}^{2}\)		−0.000** (0.000)		−0.000** (0.000)
Log GDPpc×T			0.018 (0.012)	0.018 (0.012)
\(log GDPpc\times {T}^{2}\)			−0.001* (0.001)	−0.001* (0.001)
Industry FE	Yes	Yes	Yes	Yes
Country-year FE	Yes	Yes	Yes	Yes
Obs.	229,089	229,089	229,089	229,089
\({R}^{2}\)	0.332	0.332	0.332	0.333
Notes: Standard errors in brackets clustered at country level. *Significant at the 1 percent level. Significant at the 5 percent level. *Significant at the 10 percent level.