Foreign competition and acquisitions

Accepted Manuscript Foreign competition and acquisitions

Shweta Srinivasan PII: DOI: Reference:

S0929-1199(18)30378-X https://doi.org/10.1016/j.jcorpfin.2019.07.007 CORFIN 1484

To appear in:

Journal of Corporate Finance

Received date: Revised date: Accepted date:

28 May 2018 22 May 2019 18 July 2019

Please cite this article as: S. Srinivasan, Foreign competition and acquisitions, Journal of Corporate Finance, https://doi.org/10.1016/j.jcorpfin.2019.07.007

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT This paper is based on my dissertation at the University of Arizona. I am extremely grateful to Sandy Klasa (dissertation chair), Lubomir Litov, Ryan Williams and Satheesh Aradhyula for their guidance and many helpful discussions. I also thank Murali Jagannathan, Cihan Uzmanoglu, Matthew Serfling, Douglas Fairhurst, Vince Intintoli, seminar participants at Binghamton University and the University of Victoria, and two anonymous referees for their helpful comments. Foreign Competition and Acquisitions

T

Shweta Srinivasan* [email protected]

IP

School of Management, Binghamton University, Binghamton, NY 13902,

CR

Abstract

I provide evidence on the impact of foreign competition on firms’ propensities to engage in

US

mergers and acquisitions. Using import tariff reductions as exogenous shocks that increase foreign competition, I find that affected firms are more likely to make acquisitions following

AN

tariff reductions. Cross-sectional tests show that this association is more pronounced for single segment firms, capital intensive firms, firms with higher profit margins, and firms with better growth opportunities, which suggests that this association is stronger for firms that are affected

M

by increased competition to a greater extent, and firms that stand to gain more from acquisitions

ED

when faced with increased competition. Moreover, the positive relation between acquisition likelihood and tariff cuts is more pronounced for financially unconstrained firms, and during

PT

times of high capital liquidity, which implies that it is easier for firms with greater access to capital to respond to increases in foreign competition by making acquisitions. Finally, I find

CE

some evidence that the acquisitions made in response to tariff decreases are associated with better firm profitability ratios in the following year, indicating that firms respond to increased

AC

competition by making acquisitions to improve their operational efficiency. Keywords: foreign competition; mergers; acquisitions; international trade

1. Introduction Prior work identifies several motives for corporate acquisitions. For example, Harford (1999) finds that cash rich firms suffering from agency problems are more likely to make

ACCEPTED MANUSCRIPT acquisitions, while Malmendier and Tate (2005) argue that mergers are a manifestation of CEO overconfidence. Shleifer and Vishny (2003), and Rhodes-Kropf and Viswanathan (2004) develop models that predict that acquisition activity is driven by managers taking advantage of market overvaluations of their firms.

Extant literature also cites synergies as a motivation for

T

acquisitions (Rhodes-Kropf and Robinson (2008), Hoberg and Phillips (2010)). Mitchell and

IP

Mulherin (1996) argue that merger waves are a result of firms responding to industry-level

CR

shocks like deregulation (Becher (2000), Becher and Campbell (2005)), foreign competition, and financing innovations. Harford (2005) further examines this question and finds that economic,

US

regulatory, and technological shocks in an industry initiate merger waves, but only if firms have

AN

sufficient liquidity to undertake restructuring activities.

In this paper, I examine how acquisition activity within an industry changes in response

M

to changes in foreign competition. There is evidence in the industrial trade literature that foreign

ED

competition reduces price-cost margins (Katics and Petersen (1994), and Harrison (1994)), increases productivity (Pavcnik (2002)) and results in asset reallocation (Bertrand, Schoar and

PT

Thesmar (2007)). Several papers also examine the profitability of cross-border mergers in the

CE

face of trade liberalization (Long and Vousden (1995), Benchekroun and Ray Chaudhuri (2006)), but little is known about the domestic acquisition activity of U.S. firms in response to increased

AC

foreign competition. Increased foreign competition forces domestic firms to improve efficiency (Walter and Gray (1983) and Claessens, Demirgüç-Kunt and Huizinga (2001)). Given that firms often use acquisitions to respond to industry level shocks (Mitchell and Mulherin (1996), Mulherin and Boone (2000), and Andrade and Stafford (2004)), I predict that domestic firms make more acquisitions in response to increased competition from foreign firms.

ACCEPTED MANUSCRIPT Similar to Fresard (2010), Xu (2012) and Fresard and Valta (2016), I measure increases in foreign competition using data on decreases in import tariffs for the period 1998 to 2014. Prior research uses import penetration ratio as a measure of foreign competition, which suffers from concerns of potential reverse causality1 . An advantage of using import tariffs to measure changes

T

in industry competition structure is that a reduction in tariffs is an exogenous shock with respect

IP

to firms’ decisions to merge. Using changes in import tariffs affords a quasi-natural experiment

CR

that establishes a causal relation between foreign competition and acquisition activity within an industry. I find that compared to the average firm, U.S. firms affected by tariff cuts are 19% to

US

27% more likely to make acquisitions. This result is consistent with the hypothesis that firms

AN

increase their acquisition activity when faced with greater foreign competition.

M

I further examine the type of deals that drive the increased acquisition activity following tariff cuts. One possibility is that firms consolidate within the industry when competition

ED

increases. On the other hand, firms could respond to increased competition by diversifying into

PT

other industries when their primary industry gets highly competitive. I find that acquisitions in response to tariff cuts are concentrated within the same industry, and firms are not more likely to

CE

make diversifying acquisitions when their industries face tariff cuts.

AC

To rule out potential reverse causality, I examine the timing of the relation between likelihood of making acquisitions and increased foreign competition. It could be argued that following consolidation within an industry, the government reduces tariffs to mitigate the impact of such reduced competition. In other words, acquisition activity could be driving the tariff cuts. If reverse causality were an issue, the likelihood of making an acquisition should increase in the years prior to tariff reduction. I find no significant increase in firms’ propensities to make 1

Import penetration ratio is defined as the ratio of imports to the sum of imports and domestic product.

ACCEPTED MANUSCRIPT acquisitions during the year preceding or following a tariff cut. I do, however, document a positive relation between tariff decreases and the likelihood of acquisitions by firms in the affected industries in the years there are tariff cuts. As an additional robustness test, I assign acquirers to random firm-years and find no impact of tariff cuts on acquisition activity. I also

T

find that the positive and significant association between acquisition propensities and tariff cuts

IP

are robust to using alternative model and sample specifications. These results suggest that the

CR

positive association between likelihood of acquisitions and tariff cuts is unlikely to be driven by

US

reverse causality or by random chance.

I further investigate the cross-sectional variation in acquisition likelihood with respect to

AN

several firm characteristics. One way firms respond to increased threats of foreign competition is

M

by increasing product diversification (Bao and Chen (2013)). To the extent that diversification mitigates the impact of competition faced by any one segment of a firm, single segment firms

ED

should be affected by increased competition to a greater extent. Consistent with this argument, I

PT

find that the positive association between tariff cuts and acquisition likelihood is more

CE

pronounced and statistically significant for single segment firms. Extant literature cites reduced operating costs and/or capital investments as a potential

AC

source of post-merger gains (Bradley, Desai and Kim (1988), Devos, Kadapakkam and Krishnamurthy (2009)). These gains are more valuable for firms with higher capital investments, since these firms will be able to eliminate duplication of capital investments following acquisitions. If firms indeed respond to increased foreign competition by restructuring to improve production efficiencies, then firms with a higher proportion of capital investments should be more likely to make acquisitions in response to tariff decreases. Consistent with this argument, I find that the positive association between acquisition likelihood and tariff decrease is

ACCEPTED MANUSCRIPT more pronounced for firms with a higher capital-to-labor ratio. Overall, these findings show that firms that stand to gain more by acquiring are the ones more likely to make acquisitions in response to increased foreign competition. Prior work argues that industry shocks create incentives for transfer of assets to more

IP

T

productive uses. During times of industry expansion, more productive firms tend to acquire

CR

assets from less productive firms since a more productive firm will gain more value from the assets it controls (e.g., Maksimovic and Phillips (2001)). In other words, firms with higher

US

operating efficiencies will be more likely to make acquisitions in response to increased foreign competition. Consistent with this conjecture, I find that within industries that face tariff cuts,

AN

firms with higher profit margins are more likely to make acquisitions.

M

I also examine the possibility that low growth opportunities are driving both tariff cuts as

ED

well as increased acquisition activity. It is possible that the U.S. government relaxes tariff protection for industries with low growth opportunities. Firms in such low growth industries will

PT

also be more likely to acquire other targets to compensate for lack of internal growth

CE

opportunities. I find no evidence for this argument. In fact, I find that within industries that face tariff cuts, the increase in acquisition likelihood is more pronounced for firms with higher growth

AC

opportunities. This is again consistent with the argument that firms that have more potential for synergistic gains from acquisitions are more likely to respond to increased foreign competition through this channel. However, firms need to have the wherewithal to be able to respond to increases in competition. I next examine how financial constraints affect the association between increase in foreign competition and firms’ propensities to engage in mergers and acquisitions. I find that the

ACCEPTED MANUSCRIPT positive association between acquisition likelihood and tariff cuts is more pronounced and statistically significant for firms with more cash, lower leverage, and better z-scores, implying that financially unconstrained firms are better able to respond to increased foreign competition by making acquisitions. I also find that within industries facing tariff cuts, firms with bond

T

ratings are more likely to make acquisitions. Following Harford (2005), I use the Commercial

IP

and Industrial Loan (C&I) spread as an indicator of market liquidity and find that the increase in

CR

probability of acquisition in response to tariff cuts is only significant when the C&I spread is

US

lower. Overall, these results suggest that availability of sufficient capital is an important determinant of the acquisition activity in the face of increased competition.

AN

To further investigate the positive association between acquisition likelihood and foreign

M

competition, I examine the impact of firms’ response to foreign competition on shareholders’ wealth. If increased takeover activity is an optimal response by firms to increased foreign

ED

competition, then the announcement of such acquisitions should be viewed favorably by market

PT

participants. However, literature on merger announcement returns has generally found little to no evidence of abnormal returns to bidder shareholders. Consistent with prior literature, I find that

CE

the announcement returns for acquisitions in response to tariff cuts are generally insignificant for

AC

bidder shareholders. However, looking at longer CAR windows, I find some evidence of positive shareholder wealth creation for firms that acquire in response to tariff cuts.

The evidence thus far indicates that firms merge to enable them to respond to increase in foreign competition. I next examine the post-acquisition performance of such firms that acquire in response to tariff cuts and find some evidence that firms that acquire in the years they are

ACCEPTED MANUSCRIPT affected by tariff cuts have better ROAs and profitability ratios in the following year, especially for firms that make within- industry acquisitions. These results, taken together with the results above, suggest that mergers and acquisitions are an optimal response to increased competition, and are primarily undertaken by firms that are

IP

T

affected more by the increased competition, and by firms that have the wherewithal to respond

CR

through this channel.

The contribution of this paper is threefold. First, I contribute to increasing our current

US

understanding of the motives for firms to make acquisitions. The empirical evidence in Mitchell

AN

and Mulherin (1996) as well as Harford (2005) suggests that merger waves are driven by industry shocks like deregulation and financing innovations. Whereas research in this area has

M

extensively examined the association between M&A activity and deregulation as an industry

ED

shock, this paper fills the gap in literature by showing that firms make acquisitions in response to increased foreign competition. This study also provides empirical evidence that firms that

PT

acquire in response to increased foreign competition do so to achieve synergies and efficiency

CE

gains. This finding is consistent with Rhodes-Kropf and Robinson (2008) and Hoberg and Phillips (2010) who find that firms merge to exploit synergies.

AC

Second, I contribute to the growing literature on how competition affects corporate decisions. Prior studies examine the relation between competition and leverage (Chevalier (1995), Phillips (1995), Xu (2012)), cash holdings (Haushalter, Klasa and Maxwell (2007)) as well as risk management policies. This paper provides insights on how increase in foreign competition affects firms’ propensities to engage in acquisitions.

ACCEPTED MANUSCRIPT Finally, using import tariffs as a measure of foreign competition affords a quasi-natural experiment setting that allows one to make causal inferences about the results presented herein. Recent studies use reduction in import duties as an exogenous shock to industry competitive structure to address questions pertaining to various corporate policies like capital structure and

T

cash holdings. However, to my knowledge, this paper is the first to closely examine the impact

CR

IP

of increased foreign competition on the likelihood of firms making acquisitions. The remainder of the paper is organized as follows. Section 2 provides an overview of the

US

institutional setting in the U.S. with regard to international trade. Section 3 discusses the existing literature and hypothesis development, while Section 4 presents the data and methodology.

AN

Empirical results are discussed in Section 5 and Section 6 concludes.

M

2. Institutional Details

ED

Tariffs impose additional costs on foreign manufacturers, thereby increasing the price of

PT

imported goods relative to domestically produced ones. Economists favoring free trade have long since argued that the costs of tariffs outweigh the benefits. Tariffs imposed to protect domestic

CE

producers lead to increased prices, which in turn lead to reduced demand and therefore reduced

AC

production. In the long run, this can result in job losses and reduced productivity in the economy. On the other hand, governments levy import tariffs to protect nascent as well as mature domestic industries from foreign competition2 . Import tariffs can also help discourage foreign producers from “dumping” their goods in the domestic markets. Finally, tariffs generate revenue for the government imposing them. Anderson and van Wincoop (2004) estimate that tariffs and 2

Governments can also impose non-tariff barriers like import quotas to discourage foreign exporters, or offer subsidies to domestic producers. However, the ‘tariffication’ process as part of the 1995 Uruguay round converted all non-tariff trade barriers to tariffs. Absolute import quotas were either converted to import tariffs or tariff based quotas.

ACCEPTED MANUSCRIPT nontariff barriers for industrial goods to be about 7.7% of overall trade costs3 . According to the World Bank’s World Integrated Trade System (WITS) website, import of goods and services was about 15.45% of U.S. GDP in 2015, and the average tariff rate across all products and countries for the same year was 2.87%.4

IP

T

Barriers in international trade have largely been governed by agreements under General

CR

Agreements in Tariffs and Trade (GATT)/World Trade Organization (WTO). The GATT was a multilateral agreement, first signed in 1947, established with the purpose of regulating

US

international trade and reducing trade barriers. The Uruguay round of GATT in 1995 is the most notable for eliminating absolute quotas and import bans in the agricultural sector, and for the

AN

establishment of WTO. The WTO is a set of agreements that outlines principles of international

M

trade, procedures for settlement of disputes, and requires member countries to formulate trade policies in accordance with the agreement5 . The principles of GATT still form the basis of the

ED

WTO agreements.

PT

The United States International Trade Commission (USITC) is responsible for maintaining

CE

U.S. commitments under the GATT/WTO. The U.S. has been a member of GATT and later the WTO since 1948. Thus, any tariff reductions made by the U.S. Government in my sample would

AC

be in accordance with the GATT/WTO agreements. This implies that the tariff cuts in my sample are indeed exogenous industry-level shocks to domestic firms, albeit non-random6 . This affords a quasi-natural experiment setting to study the causal relationship between tariff reductions and firms’ acquisition propensities.

3

The authors also note that direct trade policy related costs are very difficult to find and use. For more details, see https://wits.worldbank.org/CountryProfile/en/Country/USA/Year/2015 5 For details on WTO agreements, see http://www.wto.org/english/thewto_e/whatis_e/tif_e/utw_chap2_e.pdf 6 The WTO agreements lay out a roadmap for phased tariff reductions by member countries. The approximate time and magnitude of tariff reductions can be anticipated, making these events non-random. 4

ACCEPTED MANUSCRIPT Currently, 164 countries are members of the WTO. WTO members accord the status of Most-Favored Nation (MFN) to each other. This implies that members of the WTO cannot treat any other member country any less favorably that the others with respect to trade barriers, with some exceptions. Specific member countries can negotiate bilateral or multilateral regional trade

T

agreements (RTAs), subject to certain rules. As of June 2016, all member countries have an RTA

IP

in force. This means that any tariff increase or decrease by a member country on imports from

CR

another member country will be part of an agreement that includes a reciprocal tariff change by the other member country. In this paper, I consider the ad-valorem import duty rates charged by

US

the U.S. aggregated across all industries and MFN countries. This allows me to use a single duty

AN

rate that is the same across all import countries in the sample regardless of the level of imports

M

from each country.

Another important aspect of the GATT/WTO agreements is “bound” tariffs – tariff rates that

ED

are committed under the agreements and are difficult to increase7 . An important outcome of the

PT

Uruguay Round was the significant increase in the number of imports that were bound. The WTO website states that developed countries increased the number of imports with bound tariffs

CE

from 78% to 99%, while developing nations increased their imports with bound tariffs from 21%

AC

to 73%. Thus, the general trend in international trade has been towards a gradual relaxation of trade barriers. Nonetheless, the median percentage change in import tariff rate in my sample is zero.

7

In other words, tariff bounds are ceilings above which members cannot apply tariffs. However, such bounds can be changed after negotiations with the trading partners and countries can still increase tariffs where it is deemed necessary and reasonable, although the negotiations could mean compensating the trading partners for loss of trade. For example, in 2009 U.S. increased tariffs on passenger and light truck tires imported from China. See http://online.wsj.com/news/articles/SB125271824237605479 for full article.

ACCEPTED MANUSCRIPT 3. Related Literature and Hypothesis Development The literature offers several reasons for why firms merge. Harford (1999) finds evidence that cash rich firms suffering from agency problems are more likely to make value decreasing acquisitions. Malmendier and Tate (2005) argue that overconfident CEOs tend to undertake

T

value destroying mergers as they overestimate their ability to generate returns. They also find

IP

stronger negative reactions to takeover bids. Masulis, Wang and Xie (2007) find that firms with

CR

more antitakeover provisions experience significantly lower abnormal returns during the

US

announcement period, which is consistent with the argument that managers less subject to the disciplinary actions of the market for corporate control tend to make value decreasing

AN

acquisitions. Shleifer and Vishny (2003) and Rhodes-Kropf and Viswanathan (2004) favor the

ED

advantage of overvalued markets.

M

managerial timing view in that they argue that merger waves occur because managers take

On the other hand, Jensen and Ruback (1983) review the evidence on the market for

PT

corporate control and find that in general, corporate acquisitions are value increasing activities where the target firm shareholders gain and the bidder firm shareholders do not lose. Mitchell

CE

and Lehn (1990) find evidence in favor of the disciplinary action of the market for corporate

AC

control. These authors find that firms that make bad acquisitions are more likely to be subsequently taken over. Mitchell and Mulherin (1996) argue that merger waves are a result of firms responding to industry level shocks. Harford (2005) further examines this question and finds that economic, regulatory as well as technological shocks in an industry initiate merger waves, but only if firms have sufficient liquidity to undertake restructuring activities. Prior work also documents synergistic gains as a motive for mergers (Bradley, Desai and Kim (1983, 1988), Hoberg and Phillips (2010)). Another vastly documented reason for mergers

ACCEPTED MANUSCRIPT between two firms is to obtain market power. For example, Kim and Singal (1993) study airline mergers and find that efficiency gains on airfares are more than offset by the price increases due to increased market power. Other related papers that examine mergers include research that looks at the factors that

IP

T

affect the probability of a firm receiving a takeover bid (Palepu (1986), Ambrose and Megginson

CR

(1992)), papers that examine cross border mergers (Moeller and Schlingemann (2005), Nocke and Yeaple (2007)), and theoretical works that provide a framework for the rules of approving

US

merger deals (Farrell and Shapiro (1990) and Barros and Cabral (1994)). Additionally, Ahmad, De Bodt, and Harford (2017)) examine global merger waves and find that merger waves in a

AN

country can be explained by M&A activity in the trading partner countries.

M

Yet another strand of literature examines the effect of industry competition on corporate

ED

policies. For example, Xu (2012) finds a negative relationship between import competition and leverage ratios. Grullon and Michaely (2007) show that product market competition incentivizes

PT

managers to be more efficient and have a positive effect on corporate payout policies while

CE

Hoberg, Phillips and Prabhala (2013) find that firms facing competitive threats adopt more conservative payout policies.

AC

In this paper, I study the impact of foreign competition on the acquisition activity of firms. To survive industry shocks, firms typically use several modes of restructuring including downsizing, layoffs, diversification as well as acquisitions (Kang and Shivdasani (1997)). Given that firms use acquisitions to respond to industry level shocks (Mitchell and Mulherin (1996), Mulherin and Boone (2000), and Andrade and Stafford (2004)), it follows that domestic firms will be more likely to make acquisitions to respond to increased competition from foreign firms.

ACCEPTED MANUSCRIPT A closely related paper is by Breinlich (2008), who examines the M&A activity in Canada and the U.S. following Canada-U.S. Free Trade Agreement (CUSFTA). The author finds that the trade liberalization caused an increase in M&A activity in the Canadian market, but had little impact on the acquisition activities of U.S. companies. In this paper, I use tariff changes

T

over a longer sample period and provide evidence that such increases in foreign competition do

IP

indeed have a significant impact on the acquisition activities of U.S. firms. This paper also

CR

conducts a deeper analysis of the cross sectional variations in the acquisition activities of

US

domestic firms affected by tariff cuts.

Given that firms can respond to increased competition in multiple ways, in this paper I

AN

examine whether acquisitions are a significant channel through which U.S. firms facing

M

increased foreign competition respond. Ultimately, whether tariff cuts affect the domestic M&A

ED

activity is an empirical question. Thus, the main hypothesis stated in alternative form is: H1: Firms’ propensities to make acquisitions increase in response to greater foreign

PT

competition.

CE

I use decreases in import tariffs to measure increases in foreign competition to empirically test this hypothesis. I predict a positive association between the likelihood of making

AC

acquisitions and decreases in import tariffs. If tariff cuts do indeed increase foreign competition, then firms that are affected more by such tariff cuts should be more likely to acquire. Further, if these acquisitions in response to increased foreign competition are driven by motives of efficiency gains, then we should observe that firms more likely to gain from operating efficiencies post-merger are the ones making these acquisitions. Therefore, I predict that single-

ACCEPTED MANUSCRIPT segment firms and firms with higher capital-to-labor ratio, higher gross margins, and higher growth opportunities are more likely to acquire when faced with increased foreign competition. In the face of greater foreign competition, only firms that have the resources and ability to respond through acquisitions should be observed to be doing so. Thus, I predict that the

CR

financially constrained firms and in times of low capital liquidity.

IP

T

positive association between acquisition likelihood and tariff cuts will be less pronounced for

Finally, to the extent that such acquisitions in response to increased foreign competition

US

are optimal responses by firms, I predict that these announcements are received favorably by

AN

shareholders, and that firms that acquire in response to tariff cuts perform better in the year

4.1.

Sample Construction

ED

4. Sample and Methodology

M

following the acquisitions.

PT

Data on acquisitions are obtained from SDC. Consistent with prior literature, I keep only

CE

completed merger deals. I drop observations where the bidder held more than 50% stake in the target company prior to the acquisition, or holds less than 100% after the merger. To ensure that

AC

the results are not driven by a large number of transactions with insignificant deal values, I also drop acquisitions with deal values of less than $1 million. Similar to Harford (1999) I define the dependent variable as a binary variable that takes a value of one if there is an acquisition in that year by a given firm. Data on accounting variables are obtained from Compustat and data on import tariffs is described in the following section. Finally, I drop industry-years with less than

ACCEPTED MANUSCRIPT three firms8 . The resulting sample consists of 15,803 observations with 1,925 acquisitions spanning 26 different industries at the 3- digit NAICS level9 . Table 1 presents the descriptive statistics. The mean probability of acquisition likelihood for my sample is 12.22%. Around 22% of firm-years in the sample face a tariff cut of 2% or more, 11.6% firm-years in the sample

T

experience a tariff cut of 20% or more, whereas the average firm faces tariff cuts of 30% or more

IP

in 8.1% of the sample years.

CR

Figure 1 shows the number of tariff cuts of 2%, 20%, 30%, or more, across all industries for each year in the sample period. There were more tariff cuts between 1998 and 2005 than in

US

the years after 2005. This could be because of multilateral agreements under WTO where several

AN

countries agreed to phase out tariffs for industries like agriculture and apparel by 2005. Figure 2 shows the trends in acquisition activities as well as percentage changes in import tariffs by year.

M

The graph shows that increases in acquisition activities across all industries generally coincide

ED

with years when percentage changes in ad valorem import duties are negative. This is consistent with Mitchell and Mulherin (1996) and Harford (2005) that firms respond to exogenous industry

Measurement of increase in foreign competition

CE

4.2.

PT

shocks through mergers and acquisitions.

AC

I measure increases in foreign competition within an industry using decreases in import tariffs. I gather data on import tariff at the 6-digit NAICS level from the United States International Trade Commission’s (USITC) website. The website allows interactive data download at the industry level and by country of import. Since this study uses decreases in

8

The results and analyses in this paper are robust to not excluding these industry-years as well as excluding industry-years with less than 10 firms. 9 Two NAICS industries, 111 and 112 (crop and animal production, respectively) only have three and one firm year observations in the sample, respectively. In regression models, I drop these four observations because they don’t sufficiently represent these two industries. All results are robust to including these observations.

ACCEPTED MANUSCRIPT import tariffs to measure increases in foreign competition within an industry, I need tariff data at the industry-year level only. I gather data on collected duties and dutiable import values by industry-year, aggregated across all MFN countries and products within the NAICS industry. Next, I calculate the change in the import tariffs faced by an industry each year as the percentage

T

change in the ad-valorem duty rate from the prior year. The independent variable of interest is a

IP

binary variable that takes a value of one for industry-years that face a significant tariff cut, and

CR

zero otherwise. I define the binary variables for three cutoffs: Decdummy1 takes a value of one for a tariff reduction of 2% or more (this is approximately the 27th percentile value), Decdummy2

US

takes value of one for tariff cuts of 20% or more (approximately the 16th percentile value), and

AN

Decdummy3 defines an increase in foreign competition as a tariff decrease of 30% or more (approximately the 12th percentile value). For ease of exposition, I define the tariff cut variables

M

at thresholds that are closest to the 25th , 15th and 10th percentile values and yet are round

ED

numbers. For this reason, I define Decdummy1, Decdummy2, and Decdummy3 as tariff cuts of 2% or more, 20% or more, and 30% or more, respectively. If a tariff cut is followed by an

PT

equivalent increase or more within the next two years, I set the tariff cut dummy variables to

CE

zero. This is similar to Fresard (2010) and is done to ensure that the empirical tests are not simply picking up spurious association between acquisition activities of firms and temporary or

AC

ad-hoc cuts in tariff rates. In timing and cross-sectional tests, I use the largest tariff cut threshold (Decdummy3) to measure increases in foreign competition. The tariff cuts are measured as a percentage reduction in import tariffs from the previous year, and not as changes in percentage points. For example, consider industry 334210 – Telephone Apparatus. The industry faced a reduction in import duties from 3.52% in 2007 to 2.93% in 2008. This is a percentage change of -16.70% in the ad-valorem import duty rate in

ACCEPTED MANUSCRIPT 2008. Hence for this industry in 2008, the variable Decdummy1 would take a value of one whereas Decdummy2 and Decdummy3 would each take a value of zero. Fresard (2010), Valta (2012), as well as Fresard and Valta (2016) use tariff data compiled by Feenstra, Romalis and Schott (2002). Xu (2011) uses the same database to compute the ad

T

valorem equivalent of MFN tariff rates from 1989 to 2001. This database is available from

CR

IP

NBER’s website. However, the data from the USITC website allows one to extend the sample period to more recent years. From this website, I obtain tariff data from 1997 to 2014 for 465 industries,

spanning the agriculture, mining, manufacturing, and

US

different 6-digit NAICS

AN

publishing industries.

Notably, USITC uses SIC as the basis of industry classification prior to 1997 and NAICS

M

classification 1997 onwards. To use a consistent industry classification throughout, I start my

ED

sample in 1997. Since the increase in foreign competition is defined as the percentage change in duty rates, the final sample starts at 1998. Figure 3 shows the evolution of average ad-valorem

PT

import tariff rates by year. Table 2 provides a breakup of average tariff changes by industries at

CE

the 3-digit NAICS level, along with the average import tariffs faced by these industries during the sample period. For the overall sample, the average industry faces an import tariff rate of

AC

3.3%. Firm-years in ten of the 26 3-digit industries in the sample experience a tariff cut from the previous year, while the remaining industries experience tariff increases. Interestingly, roughly 56% of the sample firm-years belong to three industries – Computer and electronic product manufacturing, chemical manufacturing, and publishing industries (except internet). Two of these three industries experience tariff increases on average. The average change in tariff across all industries in the sample was a 7% increase, whereas the median firm faced no change in tariffs.

ACCEPTED MANUSCRIPT Since I use tariff data that is different from what prior literature uses, I first check whether the tariff cuts in my sample are indeed exogenous shocks to the industries. I follow Fresard (2010) and regress variables that measure tariff cuts on industry level factors. Table 3 shows that tariff cuts as measured by the tariff decrease dummy Decdummy3, changes in ad

T

valorem duty rates, as well as percentage changes in ad valorem duty rates are largely

IP

independent of median industry cash holdings, ROA, size, leverage, and market-to-book value of

CR

assets. More importantly, the variables of tariff cuts are independent of the number of firms in an industry as well as the number of acquisitions within an industry. This is consistent with the

US

finding in the other papers using NBER data mentioned above that increases in foreign

AN

competition as measured by decreases in import tariffs are exogenous shocks to the industries.

M

Overall, the univariate analyses suggest that tariff cuts represent exogenous shocks to an industry, and that acquisition activity increases in the years that U.S. firms face significant tariff

Model

PT

4.3.

ED

cuts.

CE

To test the main hypothesis in a multivariate setting, I use a difference-in-difference model to estimate the change in a firm’s acquisition propensity in response to an exogenous

AC

shock to its industry’s competitive environment. The treatment sample comprises of firms operating in industries that face tariff cuts as defined in Section 4.2 in year t. If a firm does not belong to an industry that faces a tariff cut in year t, it belongs to the control sample that year. Since an industry can face tariff cuts in multiple years and of different magnitudes, the sample of treatment vs. control firms changes every year. However, as mentioned in Section 4.2, the tariff cut variables take a value of one only for industry-years that face significant tariff cuts and are not followed by an equivalent increase or more over the next two years. This ensures that the

ACCEPTED MANUSCRIPT results in the paper are not driven by ad-hoc tariff cuts that can be reversed in the next couple of years, but rather by long-term changes to the competitive environment of firms in the treatment industries. To test the main hypothesis, I use the following model:

is a matrix of control variables measured in year t-1, and Decdummy(j) are the

tariff decrease dummies as defined in section 4.2.

US

otherwise.

is a binary variable that takes a value of one for firms that acquire in year t, 0

CR

Where

IP

T

()

AN

I use a linear probability model10 to test my hypotheses, and control for firm and industry level factors recognized by prior studies as determinants of acquisition likelihood. Firm level

M

control variables include size, return on assets, level of cash holdings, R&D expense, debt-to-

ED

equity ratio, market-to-book ratio of assets, return volatility, and an indicator for whether the

PT

firm suffered a net loss in that year. Industry level variables include average industry ROA, cash holdings, leverage, and the number of firms in the industry. To minimize endogeneity concerns,

CE

all control variables are lagged by one year. I include firm and year fixed effects in my tests to

AC

control for unobserved time invariant firm-specific characteristics and for variations across years. The final sample consists of an unbalanced panel of 15,799 firm-years after excluding four firmyears in crop and animal production industries. I exclude these observations as they do not sufficiently represent the two industries. The actual number of observations vary across regressions due to perfect predictability of acquisition probabilities for some firms.

Standard

errors are corrected for heteroskedasticity and clustering at the 3-digit NAICS industry level. 10

Using a conditional logit model yields qualitatively similar results. The linear probability model allows the use of firm-year fixed effects rather than using dummies for each firm, which significantly increases the number of variables in Stata.

ACCEPTED MANUSCRIPT 5. Empirical Results This section discusses the empirical results from testing the hypothesis that firms are more likely to make acquisitions in response to increased foreign competition. If firms do indeed make acquisitions in response to a change in industry competition structure, then a decrease in

T

import tariffs should be associated with an increase in acquisition likelihood. Consistent with my

IP

hypothesis, Table 4 indicates that the intensification of industry competition, as measured by

CR

decreases in import tariffs, is associated with a 2.3% to 3.3% increase in the likelihood of firms

US

within that industry making an acquisition. In terms of economic significance, this finding implies that a 2% decrease in import tariffs results in firms within that industry being

AN

approximately 18.8% more likely to make an acquisition compared to the average firm in the sample (0.023/0.122). Models 6 and 9 indicate that a 20% and a 30% decrease in import tariffs

M

are associated with firms being 19.7% and 27% more likely than the average firm to acquire,

ED

respectively. The likelihood of making acquisitions monotonically increases with the magnitude

PT

of tariff cuts, indicating that the increase in firms’ propensities to make acquisitions can indeed be attributed to intensifying industry competition. Overall, these results are consistent with the

CE

hypothesis that firms are more likely to make acquisitions in response to increased competition.

separately.

AC

I repeat the analyses in Table 4 for same-industry and diversifying acquisitions It is possible that firms consolidate within an industry when competition increases.

This could be because firms would like to maintain their competitive standing within the industry and when faced with the possibility of competition from a larger number of firms, they counteract this effect by acquiring other firms within the industry. This is consistent with the prediction that firms will be more likely to make same-industry acquisitions in the years their industries face tariff cuts. On the other hand, firms could respond to increased foreign

ACCEPTED MANUSCRIPT competition by making diversifying acquisitions (Bao and Chen (2013)). This is consistent with the argument that when faced with increased foreign competition, domestic firms will be more likely to diversify into other markets to offset the effects of high competition in the industries they primarily operate in.

Table 5 shows that the positive association between acquisition

T

likelihoods and tariff cuts is statistically significant when the bidder and target firms are in the

IP

same 2-digit industry codes, whereas firms are not significantly more likely to make diversifying

CR

acquisitions when faced with tariff cuts. These findings indicate that firms are more likely to

US

consolidate within their industries, and not significantly more likely to diversify, when faced with increased foreign competition.

Reverse Causality

AN

5.1.

M

One potential concern is that the above tests could suffer from reverse causality. For example, mergers between a few firms within an industry could result in a reduction in

ED

competition, in turn inducing the USITC to reduce import tariffs to allow foreign firms to

PT

compete in the domestic market. I test the association between acquisition likelihood and the timing of the tariff cuts for evidence of causal relation. If reverse causality were an issue, the

CE

likelihood of making an acquisition should increase in the year prior to tariff reductions. Panel A

AC

of Table 6 presents results of this analysis. Decdummy3t+1 is an indicator variable that takes a value of one for industries that face a tariff cut of 30% or more in the following year. A significant positive coefficient on this variable would imply a higher likelihood of acquisitions by domestic firms in the year prior to the tariff cuts. Decdummy3t is an indicator variable that takes a value of one for the years of tariff reductions. Finally, Decdummy3t-1 takes a value of one for the year immediately following tariff decreases.

ACCEPTED MANUSCRIPT In columns 1 and 3 of Panel A, I find that the probability of acquisitions for the average firm is not significantly higher in the year prior to or the year after the tariff reductions. All three models in this table indicate that firms’ propensities to make acquisitions increase significantly in the year of the tariff cuts. The results in Panel A of Table 6 can also be viewed as a test of the

T

parallel trends assumption of the difference-in-difference model. The parallel trends assumption

IP

in the context of this paper would state that in the absence of treatment (tariff cuts), there should

CR

be no difference between acquisition activities of firms that face tariff cuts and firms that don’t. In other words, one should not expect any significant impact of tariff cuts in year t on firms’

US

acquisition propensities in years t-1 and t+1. This is consistent with the results in Panel A.

AN

However, the results in this table only provide evidence that the acquisition propensities in years t-1 and t+1 are not significantly different from zero. To truly support the parallel trends

M

assumption, one needs to establish whether the acquisition propensities in the years before and

ED

after the treatment event are significantly different from each other. Figure 4 plots the trend in the coefficient estimates from regressing acquisition likelihood on the dummy variables that take

PT

a value of one for two years before the treatment events through two years after the treatment

CE

events. Treatment event is defined as a percentage reduction in import duties of 30% or more (Decdummy3). The figure shows an absence of any significant trend in the acquisition

AC

propensities in the two years before and two years after tariff cuts. The graph in Figure 4 does not look like a traditional graph testing the parallel trends assumption. It must be noted here that the even though the tariff cuts in the sample are permanent (i.e., they are not followed by an equivalent tariff increase or more in the next two years) the dummy variables measuring tariff cuts take a value of one only for the years of significant reductions in tariffs, and zero in the years afterwards. This is different from a standard difference-in-difference model with a single

ACCEPTED MANUSCRIPT treatment event with pre- and post-sample periods. If an industry faces another tariff cut some years later that is not followed by an equivalent tariff increase or more, the tariff cut dummy variables, depending on the magnitude of the tariff cut, take a value of one again for that year. Figures A1 through A3 in the Appendix present the trends in the average number of

IP

T

acquisitions by treatment and control firms in year before, the year of, and the year after tariff

CR

cuts of 30% or more respectively. Again, these figures show an absence of any significant difference in the acquisition activities of treatment firms relative to control firms in the year

US

before and the year after tariff cuts.

AN

As an additional robustness test, I assign acquisitions to random firm-years. If the positive and significant association between acquisition propensities and tariff cuts is indeed

M

causal, one should expect to find no significant relation between tariff cuts and such randomly

ED

assigned acquisitions. Panel B of Table 6 shows that there is no significant association between tariff cuts and randomly assigned acquisitions, indicating that the results in Table 4 are unlikely

PT

to be driven by random chance.

CE

Overall, the findings in Table 6 that firms’ propensities to merge increase only in the

AC

years they are affected by tariff cuts suggest that the association is unlikely to be driven by reverse causality or by random chance.

5.2.

Additional robustness tests This section presents the results from robustness checks and alternative specifications of

the main results. Table 7 presents results from repeating the analyses in models 3, 6, and 9 of Table 4 after including firm and industry-year fixed effects to additionally account for timevarying industry shocks. I find that the main result that firms facing tariff cuts are more likely to

ACCEPTED MANUSCRIPT make acquisitions in the year they face the tariff cuts persists even after additionally controlling for industry-year fixed effects. In models 4 through 6 of Table 7, I also empirically examine the main hypothesis using a conditional logit model and find results consistent with the main results presented in Table 4.

IP

T

Overall, Table 7 shows that the main finding of this paper that firms are more likely to make

CR

acquisitions in response to increased foreign competition is robust to alternative regression specifications.

US

Since three industries account for 56% of the sample firm-years, one concern could be

AN

that the findings in the paper are driven primarily by these three industries. In Panels A, B, and C of Table 8, I repeat the analyses in Table 4 after excluding each of these three industries. Panel A

M

excludes the chemical manufacturing industry (NAICS 3-digit industry 325), Panel B excludes

ED

firms in the computer and electronic product manufacturing industry (NAICS 3-digit industry 334), and Panel C excludes firms in the publishing industry, except internet (NAICS 3-digit

PT

industry 511). Panel D repeats the analysis for the subsample of firms in only these three

CE

industries. Overall, the results in Table 8 show that the positive association between acquisition propensities and increased foreign competition is robust to focusing on or excluding the

AC

industries with the largest representations in the sample. Additionally, I also reexamine the main results without excluding industries with less than three firms. This is done to address the potential concern that dropping industries with less than three firms could artificially truncate the sample. Table 9 shows that the main results are robust to including industries with less than three firms.

ACCEPTED MANUSCRIPT Overall, Tables 7 through 9 show that the main finding in the paper that firms are more likely to acquire in response to increased foreign competition is robust to alternate model as well as sample specifications. Finally, to further mitigate the concern that the positive association between acquisition

IP

T

propensities and increased foreign competition are driven by a large number of deals with

CR

smaller deal values, I also reexamine the main results in Table 4 after excluding firm-years with deal values less than $10 million and $50 million, respectively. These results are presented in

US

table A1 in the Appendix. I find that the results are qualitatively similar to those in Table 4. However, it must be noted that these tests lose power as one excludes deals smaller than $10

AN

million and $50 million, respectively. Specifically, the mean probability of acquisition likelihood

M

for the main sample is 12.22%. This proportion drops to 10.2% when I drop deals less than $10 million in size. When I exclude deals less than $50 million in value, on average only 6.41% of

ED

firm-years have an acquisition. Thus, excluding deals smaller than $50 million in size has the

PT

effect of excluding approximately half the acquisition deals in the original sample, which could

5.3.

CE

potentially be causing the low power of the robustness tests.

Likelihood of acquisitions and efficiency gains

AC

In this section I explore the motives for firms to acquire in response to increased foreign competition by examining the cross-sectional variations in acquisition probabilities. Bao and Chen (2013) use news on foreign investment to measure threat of foreign competition and find that one way firms respond to such threats is by product diversification. Diversification allows a firm to extract profits in other markets that might offset the reduced profits in the current market due to increased competitive pressures. To the extent that tariff cuts increase industry competition, one should not expect multi segment firms to be affected by such increases in

ACCEPTED MANUSCRIPT competition since these firms can refocus on their more profitable segments. Consistent with this argument, Columns 1 and 2 of Table 10 shows that the positive and significant relation between acquisition likelihood and tariff cuts is more pronounced and statistically significant for single segment firms.

IP

T

Next, I investigate the synergy effects of these acquisitions. Bradley, Desai and Kim

CR

(1988) argue that merging firms gain from economies of scale as well as redeployment of assets to more valuable uses. Devos, Kadapakkam and Krishnamurthy (2009) provide examples of

US

improvements in operating performance post-merger. For example, they cite reduced operating costs and/or capital investments as a potential source of post-merger gains. These gains are more

AN

valuable to firms with higher capital investments, since these firms will be able to eliminate

M

duplication of capital investments following acquisitions. If firms indeed respond to increased foreign competition by restructuring to improve production efficiencies, then firms with higher

ED

proportions of capital costs relative to labor costs should be more likely to make acquisitions in

PT

response to tariff decreases. In Table 10, I show that within industries that face tariff cuts, firms

make acquisitions.

CE

with higher than median capital-to-labor ratios in the year prior to tariff cuts are more likely to

AC

Drawing on theories of firm scope, Maksimovic and Phillips (2001) predict that in times of industry expansion, more productive firms tend to acquire assets from less productive firms since a more productive firm will gain more value from the assets it controls. Extending their argument, firms with higher ex-ante operating efficiencies will be better positioned, and therefore more likely, to make acquisitions in response to increased foreign competition. I use gross margin as a measure of operating efficiency, following Piotroski (2000) and Piotroski and So (2012). Consistent with the above prediction, I find that within industries that face tariff cuts,

ACCEPTED MANUSCRIPT firms with higher gross margins the year before tariff reductions are more likely to make acquisitions. Table 10 also reports that firms with higher than median net profit margins tend to respond to tariff cuts through acquisitions, whereas low profit margin firms don’t. Finally,

I examine

the possibility that firms in industries with different growth

IP

T

opportunities differ in their acquisition propensities while simultaneously facing different tariff

CR

structures. For example, the USITC might decide to stop protecting an industry with lower growth opportunities and thus reduce tariffs. Simultaneously, firms in such low-growth

US

industries would be more likely to pursue growth through mergers in order to survive. To the extent that low growth opportunities are driving increased acquisition activity as well as tariff

AN

cuts, the positive relation between acquisition propensities and tariff cuts should be significant

M

only for the subsample of firms with low growth opportunities. The last two columns in Table 10 show no evidence of this. The positive association between firms’ propensities to acquire and

ED

increased competition is only significant for the subsample of firms with high market-to-book

PT

value of assets in the previous year. This rules out the concern that low growth opportunities are driving the relation between acquisition likelihood and increased foreign competition. In fact, the

CE

finding is consistent with the argument that when faced with increased competition, firms with

AC

more potential synergistic gains are more likely to respond through acquisitions. The differences between the coefficients on Decdummy3 across the sample splits are statistically significant for subsample splits based on capital-to-labor ratio, operating profit margin, and single vs. multiple segments. Overall, these results suggest that when faced with increased foreign competition, the firms that stand to gain more by acquiring are the ones that are more likely to make acquisitions. The findings are consistent with the argument that firms use

ACCEPTED MANUSCRIPT mergers and acquisitions as a medium to respond to increases in foreign competition by attempting to improve operating efficiencies.

5.4.

Financial constraints and likelihood of acquisitions I next examine the association between tariff cuts and the probability of acquisitions

IP

T

under conditions that affect access to capital. Firms need to have the wherewithal to be able to

CR

respond to the increased competition. When faced with increased competition, financially constrained firms will probably not be in a position to react by acquiring other firms. On the

US

other hand, firms with better access to capital will be able to respond to changes in industry competition through the M&A channel. Consistent with this hypothesis, Table 11 shows that the

AN

positive association between tariff cuts and acquisition propensities is statistically significant for

M

firms with higher cash holdings, lower book leverage, and better z-scores than the median firm in the year before the acquisitions. I also find that the association between tariff cuts and increased

ED

probability of acquisitions is statistically significant for firms with bond ratings. Surprisingly, I

PT

find that both firms with and without investment grade ratings are more likely to make acquisitions in the years they experience tariff cuts. However, the magnitude of the probability of

CE

acquisitions is higher for firms with investment grade ratings.

AC

Finally, following Harford (2005), I use the C&I spread as an indicator of market liquidity. Harford (2005) argues that merger waves are clustered in periods of low transaction costs and high capital liquidity. Thus, firms should be able to respond to increased foreign competition through acquisitions more easily in periods of high capital liquidity. The last two columns in Table 11 show that the probability of making acquisitions in response to tariff cuts is higher and statistically significant for the years when the C&I spread is below the sample median. The differences between the coefficients for Decdummy3 are statistically significant for

ACCEPTED MANUSCRIPT subsamples splits based on book leverage, z-score, and C&I spread. Taken together, the results in Table 11 are consistent with the prediction that the increased acquisition activity in response to increased foreign competition is more pronounced for firms that have access to sufficient capital to make such acquisitions. These findings are also consistent with availability of capital being an

IP

T

important consideration in firms’ response to increased foreign competition.

CR

The fact that the positive relation between acquisition propensities and tariff cuts is statistically significant in only certain subsamples also lends support to the argument that such

US

association is unlikely to be driven by omitted variables. For an omitted variable to explain the main results in Table 4, the variable will have to be uncorrelated with all control variables while

AN

being correlated with the firm characteristics for which the relation between tariff cuts and

Announcement returns and post-acquisition performance

ED

5.5.

M

acquisition propensity are stronger (Agrawal and Matsa (2013)).

In this section, I analyze whether the increased acquisition activity associated with tariff

PT

decrease is indeed an optimal response by firms. If shareholders anticipate that the acquisitions

CE

in response to increased competition are an appropriate response by firms, then the announcement of such acquisitions should be associated with a positive shareholder reaction. To

AC

test this conjecture, I examine the cumulative abnormal returns in the one-day, three-day, and five-day windows surrounding acquisition announcements. Literature abounds on studies using announcement returns to gauge shareholders’ reaction to merger announcements 11 . However, the evidence on value creation for bidder firm shareholders is mixed. For example, Dodd (1980) and Bouwman, Fuller, and Nain (2009) find negative abnormal returns to acquirer shareholders,

11

See Jensen and Ruback (1983), Jarrell, Brickley and Netter (1988) and Andrade, Mitchell and Stafford (2001) for a review of these studies.

ACCEPTED MANUSCRIPT whereas Asquith (1983) and Asquith, Bruner, and Mullins (1983) document positive abnormal returns to bidder shareholders. Andrade et al. (2001) summarize that on average the abnormal returns to acquirer shareholders are not statistically different from zero. I compute daily abnormal returns in excess of the Fama-French three factor model to examine the bidder

T

shareholders’ reaction to mergers announced in response to tariff cuts 12 . The average cumulative

IP

abnormal return (CAR) for bidder firm shareholders over the [-1,+1] window is 1.02% while the

CR

average CAR for bidders that acquire in the years of tariff cuts is 0.98%. Although these return numbers are statistically different from zero, they are not significantly different from each other.

US

Next, I regress the CARs on Decdummy3 to further examine the impact of announcing

AN

acquisitions in response to increased foreign competition. These results are presented in Table 12. Columns 1, 3, and 5 present results of univariate analyses while columns 2,4, and 6 include

M

control variables consistent with prior studies. I find that acquisition announcements in response

ED

to tariff decreases are not associated with significant abnormal returns to acquirers in the one to

PT

five days surrounding the announcement13 .

Overall, Table 12 doesn’t provide any evidence that acquisitions made in response to

CE

increased foreign competition are seen as favorable responses by acquirer shareholders.

AC

However, these results are consistent with prior literature that finds that the abnormal returns for acquirer shareholders are not different from zero. Next, I extend the CAR windows to 60 and 120 days after acquisition announcements to examine whether firms that acquire in response to tariff cuts generate positive shareholder wealth in the longer run. Table 13 shows the longer-term impact of announcing acquisitions in response 12

Using abnormal returns in excess of the market model yield similar results for all three announcement windows. In unreported tables, I also examine the CARs to bidder shareholders over the [-2, +2] window and find no significant market reaction to acquisitions announced in the years that firms are affected by tariff cuts. 13

ACCEPTED MANUSCRIPT to increased foreign competition. Panel A presents results from using abnormal returns in excess of the three-factor model and shows that the CARs to firms that announce acquisitions in response to tariff cuts are not significantly different from zero. On the other hand, Panel B uses abnormal returns in excess of the market model and shows that shareholders react positively in

T

the 60 and 120 days after firms announce acquisitions in the years they face tariff cuts. Overall, I

IP

find limited evidence of shareholder value creation by firms that acquire in response to increased

CR

foreign competition.

US

Since the CAR results are ambiguous about whether acquisitions in response to tariff cuts are viewed favorably by acquirer shareholders, I next look at the acquirer firms’ accounting

AN

performance in the year following the acquisitions. If firms optimally respond to increased

M

competition by acquiring targets that improve their competitive standing within the industry, this should translate into better operating performance by these firms. In Panel A of Table 14, I

ED

regress acquirer ROA and net profit margins on the interactions between Decdummy3 and the

PT

dummy variables for all acquisitions (Acq), same industry acquisitions, and conglomerate acquisitions. In models 1 and 4, I find that the net profit margins in year t+1 for firms that

CE

acquire in response to tariff cuts are significantly higher than for firms that acquire in other

AC

years. However, the ROA for firms that acquire in response to tariff cuts are not statistically different from ROA of firms acquiring in other years. This could mean that when faced with increased foreign competition, the average firm acquires to maintain or improve its pre-tariff cut performance and profitability. Interestingly, on separating the same industry acquisitions from the diversifying mergers, I find that firms making same industry acquisitions in the years they face tariff cuts perform significantly better the following year than firms that make same industry acquisitions in the other years. On the other hand, firms that make diversifying acquisitions in

ACCEPTED MANUSCRIPT the years they face increased competition don’t significantly improve their profit margins in the following year compared to firms that make diversifying mergers in the other years. In fact, the ROA and profitability in the following year for such firms in slightly negative, albeit statistically insignificant. In Panel B of Table 14, I regress industry-adjusted ROA and net profit margins on

T

the interaction between Decdummy3 and the acquisition dummies and find results consistent with

CR

IP

those in Panel A.

Taken together, these tests show that firms use acquisitions to respond to increased

US

foreign competition and that while such acquisitions are not associated with bidder abnormal returns in the days surrounding the acquisition announcements, these deals might allow firms to

AN

maintain or improve their accounting performance in the year following such acquisitions.

M

Finally, when firms respond to increased competition by consolidating within their industries,

ED

they are able to improve their profitability in the year following such acquisitions.

PT

6. Conclusion

I examine whether firms are more likely to make acquisitions in response to changes in

CE

industry competitive structure. Using decreases in import tariffs as exogenous shocks that

AC

increase foreign competition, I find that firms within an industry are more likely to make acquisitions in response to intensification of foreign competition. The positive and significant association between tariff cuts and increased propensities to acquire is driven by same industry acquisitions rather than diversifying mergers. Timing and falsification tests show that the significant positive association between tariff cuts and increased probability of acquisitions holds only in the years of the tariff decreases and does not hold when acquisitions are assigned to random firm-years, suggesting that the results

ACCEPTED MANUSCRIPT are unlikely to be driven by reverse causality. Cross sectional tests show that firms that stand to gain more by acquiring or are affected by tariff cuts to a greater extent are more likely to make acquisitions in response to increased competition. Specifically, the likelihood of making acquisitions in response to tariff cuts is higher for single segment firms and firms with higher

T

capital-to-labor ratios, higher operating and net profit margins, and higher growth opportunities.

IP

I also find that the positive relation between tariff decreases and likelihood of acquisitions is

CR

stronger for financially unconstrained firms and in times of higher capital liquidity, suggesting

US

that access to capital is an important factor that drives M&A activity following increased competition.

AN

Tests of post-acquisition performance show that when firms acquire in response to tariff

M

cuts, they are able to improve their profitability in the year after the acquisitions. Such improvements in profitability ratios are more pronounced and significant for firms making within

ED

industry acquisitions in the years they face tariff cuts. These findings are consistent with the

PT

argument that firms undertake corporate mergers to improve efficiency in order to enable them to

CE

face the increased competition.

The evidence has implications for our understanding of the M&A literature. Researchers

AC

are divided in their opinion of whether mergers are an optimal outcome of firms seeking to achieve efficiency gains or a result of managers taking advantage of market inefficiencies. This paper documents evidence that acquisitions provide a medium for firms to restructure in response to industry-level shocks. Moreover, these restructuring activities seem to be driven by firms seeking to obtain efficiency gains that allow them to compete against other firms in their industry and to survive industry shocks.

ACCEPTED MANUSCRIPT Declarations of interest: none

Referecnes Acharya, Viral V., Ramin P. Baghai, and Krishnamurthy V. Subramanian. 2014. "Wrongful

IP

T

Discharge Laws and Innovation." The Review of Financial Studies 27: 301-346.

CR

Agrawal, Ashwini K., and David A. Matsa. 2013. "Labor unemployment risk and corporate

US

financing decisions." Journal of Financial Economics 108: 449-470. Ahmad, M.Farooq, Eric de Bodt, and Jarrad Harford. 2018. "International trade and the

AN

propagation of merger waves." Working Paper.

M

Ambrose, Brent W., and William L. Megginson. 1992. "The role of asset structure, ownership

ED

structure, and takeiver defenses in deterining acquisition likelihood." The Journal of Financial and Quantitative Analysis 27: 575-589.

PT

Anderson, James E., and Eric van Wincoop. 2004. "Trade Costs." Journal of Economic

CE

Literature 42: 691-751.

AC

Andrade, Gregor, and Erik Stafford. 2004. "Investigating the economic role of mergers." Journal of Corporate Finance 10: 1-36. Andrade, Gregor, Mark Mitchell, and Erik Stafford. 2001. "New Evidence and Perspectives on Mergers." Journal of Economic Perspectives 15: 103-120. Asquith, Paul. 1983. "Merger Bids, Uncertainty, and Stockholde Returns." Journal of Financial Economics 11: 51-83.

ACCEPTED MANUSCRIPT Asquith, Paul, Robert F. Buner, and David W. Mullins Jr. 1983. "The Gains to Bidding Firms from Merger." Journal of Financial Economics 11: 121-139. Barros, Pedro P., and Luis Cabral. 1994. "Merger policy in open economics." European

T

Economic Review 38: 1041-1055.

IP

Becher, David A. 2000. "The valuation effects of bank mergers." Journal of Corporate Finance

CR

6: 189-214.

US

Becher, David A., and Terry L. Campbell. 2005. "Interstate banking deregulation and the changing nature of bank mergers." Journal of Financial Research 28: 1-20.

AN

Benchekroun, Hassan, and Amrita Ray Chaudhuri. 2006. "Trade liberalization and the

M

profitability of mergers: A global analysis." Review of International Economics 14: 941-

ED

957.

Bertrand, Marianne, Antoinette Schoar, and David Thesmar. 2007. "Banking Deregulation and

PT

Industry Structure: Evidence from the French Banking Reforms of 1985." The Journal of

CE

Finance 62: 597-628.

AC

Bertrand, Olivier, and Habib Zitouna. 2006. "Trade Liberalization and Industrial Restructuring: The Role of Cross-Border Mergers and Acquisitions." Journal of Economics and Management Strategy 15: 479-515. Bertrand, Olivier, Jean-Louis Mucchielli, and Habib Zitouna. 2007. "Location Choices of Multinational Firms: The Case of Mergers and Acquisitions." Journal of Economic Integration 22: 181-209.

ACCEPTED MANUSCRIPT Blonigen, Bruce A., and Christopher T. Taylor. 2000. "R&D Intensity and Acquisitions in HighTechnology Industries: Evidence from the US Electronic and Electrical Equipment Industries." Journal of Industrial Economics 48: 47-70. Bouwman, Christa H.S., Kathleen Fuller, and Amrita S. Nain. 2009. "Market Valuation and

IP

T

Acquisition Quality: Empirical Evidence." The Review of Financial Studies 22: 633-679.

CR

Breinlich, Holger. 2008. "Trade liberalization and industrial restructuringthrough mergers and

US

acquisitions." Journal of International Economics 76: 254-266. Chevalier, Judith A. 1995. "Do LBO Supermarkets Charge More? An Empirical Analysis of the

AN

Effects of LBOs on Supermarket Pricing." The Journal of Finance 50: 1095-1112.

M

Claessens, Stijn, Asli Demirguc-Kunt, and Harry Huizinga. 2001. "How does foreign entry affect

ED

domestic banking markets." Journal of Banking and Finance 25: 891-911. Devos, Erik, Palani-Rajan Kadapakkam, and Krishnamurthy Srinivasan. 2009. "How Do

PT

Mergers Create Value? A Comparison of Taxes, Market Power, and Efficiency

AC

1211.

CE

Improvements as Explanations for Synergies." The Review of Financial Studies 22: 1179-

Dodd, Peter. 1980. "Merger proposals, management discretion and stockholder wealth." Journal of Financial Economics 8: 105-137. Farrell, Joseph, and Carl Shapiro. 1990. "Horizontal mergers: An equilibrium analysis." The American Economic Review 80: 107-126. Feenstra, Robert C. 1996. "U.S. Imports, 1972-1994: Data and Concordances." NBER Working Paper # 5515.

ACCEPTED MANUSCRIPT Fresard, Laurent. 2010. "Financial Strength and Product Market Behavior: The Real Effects of Corporate Cash Holdings." The Journal of Finance 65: 1097-1122. Fresard, Laurent, and Philip Valta. 2016. "How Does Corporate Investment Respond to

T

Increased Entry Threat?" The Review of Corporate Finance Studies 5: 1-35.

IP

Grullon, Gustavo, and Ron Michaely. 2007. "Corporate Payout Policy and Product Market

CR

Competition." Working Paper.

US

Hall, Bronwyn H., Adam B. Jaffe, and Manuel Trajtenberg. 2001. "The NBER Patent Citations Data File: Lessons, Insights and Methodological Tools." NBER Working Paper No. 8498.

AN

Harford, Jarrad. 1999. "Corporate Cash Reserves and Acquisitio ns." The Journal of Finance 54:

M

1969-1997.

ED

Harford, Jarrad. 2005. "What drives merger waves?" Journal of Financial Economics 77: 529-

PT

560.

Harrison, Ann E. 1994. "Productivity, imperfect competition and trade reform." Journal of

CE

International Economics 36: 53-73.

AC

Hoberg, Gerard, and Gordon Phillips. 2010. "Product market synergies and comeptition in mergers and acquisitions: A text-based analysis." The Review of Financial Studies 23: 3773-3811.

Horn, Henrik, and Lars Persson. 2001. "The equilibrium ownership of an international oligopoly." Journal of International Economics 53: 307-333.

ACCEPTED MANUSCRIPT Jensen, Michael C, and Richard S Ruback. 1983. "The market for corporate control: The scientific evidence." Journal of Financial Economics 11: 5-50. Kang, Jun-Koo, and Anil Shivdasani. 1997. "Corporate restructuring during performance

T

declines in Japan." Journal of Financial Economics 46: 29-65.

IP

Katics, Michelle M., and Bruce C. Petersen. 1994. "The effect of rising import competition on

CR

market power: A panel data study of US manufacturing." The Journal of Industrial

US

Economics 42: 277-286.

Kim, E. Han, and Vijay Singal. 1993. "Mergers and market power: Evidence from the airline

AN

industry." The American Economic Review 83: 549-569.

M

Long, Ngo Van, and Neil Vousden. 1995. "The effects of trade liberalization on cost-reducing

ED

horizontal mergers." Review of International Economics 3: 141-155. Maksimovic, Vojislav, and Gordon Phillips. 2001. "The Market for Corporate Assets: Who

PT

Engages in Mergers and Asset Sales and Are There Efficienc y Gains?" The Journal of

CE

Finance 56: 2019-2065.

AC

Maksimovic, Vojislav, Gordon Phillips, and N.R. Prabhala. 2011. "Post-merger restructuring and the boundaries of the firm." Journal of Financial Economics 102: 317-343. Malmendier, Ulrike, and Geoffrey Tate. 2005. "CEO overconfidence and corporate investment." The Journal of Finance 60: 2661-2700. Masulis, Ronald W, Cong Wang, and Fei Xie. 2007. "Corporate governance and acquirer returns." The Journal of Finance 62: 1851-1889.

ACCEPTED MANUSCRIPT Moeller, Sara B., and Frederik P. Schlingemann. 2005. "Global diversification and bidder gains: A comparison between cross-border and domestic acquisitions." Journal of Banking and Finance 29: 533-564. Mulherin, J. Harold, and Audra L. Boone. 2000. "Comparing acquisitions and divestitures."

IP

T

Journal of Corporate Finance 6: 117-139.

CR

Netter, Jeffry, Mike Stegemoller, and M. Babajide Wintoki. 2011. "Implications of Data Screens on Merger and Acquisition Analysis: A Large Sample Study of Mergers and Acquisitions

US

from 1992 to 2009." The Review of Financial Studies 24: 2316-2357.

AN

Nocke, Volker, and Stephen Yeaple. 2007. "Cross-border mergers and acquisitions vs. greenfield

Economics 72: 336-365.

M

foreig direct investment : The role of firm heterogeneity." Journal of International

ED

Opler, Tim, and Sheridan Titman. 1993. "The Determinants of Leveraged Buyout Activity: Free

PT

Cash Flow vs. Financial Distress Costs." The Journal of Finance 48: 1985-1999.

CE

Palepu, Krishna G. 1986. "Predicting takeover targets: A methodologica l and empirical

AC

analysis." Journal of Accounting and Economics 8: 3-37. Pavcnik, Nina. 2002. "Trade Liberalization, Exit, and Productivity Improvements: Evidence from Chilean Plants." The Review of Economic Studies 69: 245-276. Phillips, Gordon M. 1995. "Increased debt and industry product markets. An Empirical Analysis." Journal of Financial Economics 37: 189-238. Phillips, Gordon, and Alexei Zhdanov. 2013. "R&D and the Incentives from Merger and Acquisition Activity." The Review of Financial Studies 26: 34-78.

ACCEPTED MANUSCRIPT Piotroski, Joseph D. 2000. "Value Investing: The Use of Historical Financial Statement Information to Separate Winners from Losers." Journal of Accounting Research 38: 1-41. Piotroski, Joseph D., and Eric C. So. 2012. "Identifying Expectation Error in Value/Glamor Strategies: A Fundamental Analysis Approach." The Review of Financial Studies 25:

IP

T

2841-2875.

CR

Rhodes-Kropf, Matthew, and S Viswanathan. 2004. "Market valuation and merger waves." The

US

Journal of Finance 59: 2685-2718.

Romalis, John. 2005. "NAFTA's and CUSFTA's impact on international trade." NBER Working

AN

Paper # 11059.

M

Shleifer, Andrei, and Robert W Vishny. 2003. "Stock market driven acquisitions." Journal of

ED

Financial Economics 70: 295-311.

Valta, Philip. 2012. "Competition and the cost of debt." Journal of Financial Economics 105:

PT

661-682.

CE

Xu, Jin. 2012. "Profitability and capital structure: Evidence from import penetration."

AC

Journal of Financial Economics 106: 427-446.

ACCEPTED MANUSCRIPT Figure 1: Tariff cuts by year

AC

CE

PT

ED

M

AN

US

CR

IP

T

This figure shows the average number of tariff cuts faced by firms each year between 1998 and 2014. Decdummy1, Decdummy2, and Decdummy3 are defined as percentage decreases in ad-valorem import duty rates by 2%, 20%, or 30% or more, respectively. Ad-valorem duty rates are defined at the industry-year level, and are aggregated across all countries of import with Most Favored Nation (MFN) status. Data on import tariffs are from USITC’s website.

ACCEPTED MANUSCRIPT Figure 2: Acquisitions around tariff cuts

AC

CE

PT

ED

M

AN

US

CR

IP

T

This figure shows the acquisition activity around changes in ad -valorem import tariff rates for the sample period of 1998-2014. The bars represent the tariff changes and are aggregated across all industries and countries with Most Favored Nation (MFN) status. The line shows the number of acquisitions made each year, aggregated across all industries. Data on import tariffs are from the USITC website, whereas data on mergers and acquisitions are from SDC.

ACCEPTED MANUSCRIPT Figure 3: Average ad-val orem import tariff rates by year

AC

CE

PT

ED

M

AN

US

CR

IP

T

This figure compares the average ad-valorem import duty rate against the average percentage change in the tariff rates each year. The bars represent the tariff changes and are aggregated across all industries and countries with Most Favored Nation (MFN) status . The dotted line shows the average import duty rate each year, aggregated across all industries. Data on import tariffs are from the USITC website.

ACCEPTED MANUSCRIPT Figure 4: Impact of tariff cuts on acquisitions in the years around treatment This figure plots coefficient estimates and confidence intervals from regressing acquisition probability on dummy variables for two years before tariff cuts through two years after tariff cuts. The figure shows a lack of any significant trend in acquisition likelihoods in the two years before and two years after the treatment event. On the other hand, we observe a sharp increase in the average firm’s acquisition propensity in the treatment year. The sample period is 1998-2014. Treatment event is defined as a percentage decrease in ad-valorem import duty rate by 30% or more relative to the previous year (defined as Decdummy3 in the paper).

T

Acquisitions in the years around treatment

IP

0.1

CR

0.08

US

0.06

AN

0.04

M

0.02

0 -1

0

ED

-2

-0.02

AC

CE

PT

Years relative to treatment

1

2

ACCEPTED MANUSCRIPT

Table 1 Summary Statistics This table reports descriptive statistics for variables of interest. Acq takes a value of one for firms that make an

T

acquisition in year t, zero otherwise. Decdummy1, Decdummy2, and Decdummy3 take a value of one for industry-

IP

years that face decreases in ad-valorem import duty rates by 2%, 20%, or 30% or more, respectively. Total assets is the total asset value from Compustat and is measured in millions of U.S. Dollars. Cash is cash and marketable

CR

securities (che) divided by book value of assets. Book leverage is the book value of debt (dltt+dlc) scaled by book value of assets. Market leverage is the book value of debt scaled by market value of assets (at-ceq+prcc_f*csho).

US

Return volatility is the standard deviation of daily returns over 365 days before end of year t. R&D is research and

AN

development expenses (xrd) scaled by book value of assets. Market-to-Book assets is the ratio of market value of assets to book value of assets. Debt/Equity is the book value of debt divided by market value of equity

M

(prcc_f*csho).

Mean

Median

Std Dev

Min

Max

15803

0.122

0

0.328

0

1

15803

0.218

0

0.412

0

1

15803

0.116

0

0.320

0

1

15803

0.081

0

0.274

0

1

15803

1,795.098

180.222

7,031.173

3.532

172,384

15800

0.239

0.159

0.236

0.000

0.904

ED

Obs

Acq

PT

Decdummy1 Decdummy2 Total assets Cash

15803

0.182

0.127

0.200

0.000

1.003

Market leverage

15789

0.130

0.070

0.159

0.000

0.789

Return volatility

15803

0.044

0.038

0.026

0.006

0.820

R&D

15803

0.309

0.051

1.050

0.000

10.218

Market-to-Book assets

15789

2.201

1.565

1.971

0.439

21.887

D/E ratio

15789

0.463

0.092

1.455

0

32.315

AC

Book leverage

CE

Decdummy3

ACCEPTED MANUSCRIPT Table 2 Average changes in import tariffs by industry This table reports the average import duties as well as average and median percentage changes in import tariffs by 3digit NAICS industry. Since crop production and animal production industries represent only 0.03% of the overall sample, these two industries are excluded in further analyses. Percentage Change in Import duties Percentage of sample

Mean Duty Rate

Mean

Median

3

0.02%

-0.009

0.015

0.01% 5.13%

0.016 0.003

-0.066 -0.019

-0.066 -0.144

Industry Description

111

Crop Production

0.035

112 211

Animal production Oil and Gas Extraction

1 811

212 311

Mining (except oil and gas) Food manufacturing

26 290

0.16% 1.84%

0.002 0.076

-0.021 0.133

0.000 0.001

312 313

Beverage and tobacco product manufacturing Textile mills

149 98

0.94% 0.62%

0.012 0.108

-0.053 0.059

-0.060 -0.002

314 315

Textile product mills Apparel manufacturing

42 21

0.27% 0.13%

0.043 0.191

0.015 0.006

-0.048 0.006

316 321 322

Leather and allied product manufacturing Wood product manufacturing Paper manufacturing

104 89 211

0.66% 0.56% 1.34%

0.187 0.034 0.015

0.026 0.665 -0.176

0.018 0.000 -0.163

323 324

Printing and related support activities Petroleum and coal products manufacturing

58 126

0.37% 0.80%

0.017 0.006

-0.294 -0.033

-0.193 -0.142

325 326

Chemical manufacturing Plastics and rubber products manufacturing

2336 284

14.78% 1.80%

0.034 0.042

0.177 0.069

0.000 -0.001

327 331

Nonmetallic mineral product manufacturing Primary metal manufacturing

108 275

0.68% 1.74%

0.027 0.027

0.081 0.084

0.000 -0.009

332 333

Fabricated metal product manufacturing Machinery manufacturing

352 1389

2.23% 8.79%

0.046 0.031

0.083 0.466

0.000 0.000

334 335

Computer and electronic product manufacturing Electrical equipment, appliance, and component manufacturing

5041 515

31.90% 3.26%

0.041 0.028

0.035 -0.023

0.000 -0.002

336 337

Transportation equipment manufacturing Furniture and related product manufacturing

706 108

4.47% 0.68%

0.023 0.086

0.006 1.462

0.000 0.000

339 511

Miscellaneous manufacturing Publishing industries (Except internet)

1055 1605

6.68% 10.16%

0.041 0.000

0.027 -0.258

0.000 0.000

15803

100%

0.033

0.070

0.000

CR US

AN

M

ED

PT

CE

AC

All

IP

3-digit NAICS

T

# firmyears

ACCEPTED MANUSCRIPT

Logit model Dependent variable

Decdummy3

# of acquisitions

Dutycut

Pctchange

(2)

(3) 0.015 (0.99)

5.101*

-0.018

1.005**

(1.92)

(-0.84)

(1.99)

-0.965

-0.031*

-0.228

(-0.62)

(-1.87)

(-0.58)

0.015

-0.001

-0.046**

(0.16)

(-0.98)

(-2.38)

0.406

0.000

0.151

(0.37)

(0.04)

(0.83)

Industry median MTB assets

-0.106

0.007**

0.079**

(-0.35)

(2.57)

(2.03)

ED

Industry median leverage

0.001

(1.23)

PT

M

Industry median cash holdings

-0.029

(-1.11)

AN

Industry median ROA

Industry median size

OLS model

US

(1)

CR

IP

T

Table 3 Are tariff cuts independent of industry factors? This table reports results from regressions of tariff cuts on industry level variables. The sample period is from 1998 to 2014. The three variables for tariff cuts are as follows: Decdummy3 is a binary variable that takes a value of one if an industry faces a tariff cut of 30% or more in year t, zero otherwise. Dutycut is the year-on-year change in the advalorem import duty rate for the Most Favored Nations (MFN). Pctchange is the percentage change in the advalorem import duties for the MFN countries. The control variables are industry level medians values measured in year t-1. # of acquisitions is the number of acquisitions within the 6 digit NAICS industry -year. ROA is earnings before interest and taxes (ebit) divided by book value of assets (at). Cash holdings is defined as cash and marketable securities (che) scaled by the book value of assets. Size is the natural log of total assets. Leverage is the book value of debt (dltt+dlc) scaled by book value of assets. MTB assets is the ratio of market value of assets (atceq+prcc_f*csho) to the book value of assets (at). # firms is the number of firms operating in that 6 digit NAICS industry-year. Model 1 uses a logit model whereas Models 2 and 3 use OLS models. Industry-years with less than three firms are excluded from the regressions. Standard errors are corrected for heteroscedasticity (t -stats are in parentheses). *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively.

CE

# firms

Constant

0.026***

-0.000

-0.001

(3.98)

(-0.47)

(-0.46)

-3.664***

-0.002

0.150

(-5.64)

(-0.28)

(1.09)

0.004

0.003

1651

1651

AC

Adj. R

Pseudo R

0.033

N

1651

ACCEPTED MANUSCRIPT Table 4 Foreign Competition and Likelihood of Acquisition This table reports results from linear probability models that estimate the association between increased foreign competition and firms’ acquisition propensities. Dependent variable is the likelihood of acquisition in year t. Decdummy1, Decdummy2, and Decdummy3 take a value of one for industry-years that face decreases in ad-valorem import duty rates by 2%, 20%, and 30% or more, respectively. Control variables are measured in year t-1. R&D is research and development expenses (xrd) scaled by book value of assets (at). Cash is cash and marketable securities (che) divided by book value of assets. Size is the natural log of total assets. ROA is earnings before interest and taxes (ebit) divided by total assets. Market-to-Book is the ratio of market value of assets (atceq+prcc_f*csho) to book value of assets. Debt/Equity is the book value of debt (dltt+dlc) divided by market value of equity (prcc_f*csho). Net loss is a binary variable that takes a value of one for firms that report a net loss for year t. Return volatility is the standard deviation of daily returns over 365 days before end of year t. # of firms is the number of firms in the 6-digit industry year. Net loss takes a value of one if a firm suffers a net loss in year t, zero otherwise. Industry leverage, Industry cash, and Industry ROA are industry average book leverage, cash, and ROA in year t-1. Standard errors are corrected for heteroskedasticity and are clustered at the 3-digit NAICS industry level (t-statistics are in parentheses). Models 3, 6, and 9 include firm and year fixed effects. Number of observations varies across models due to perfect predictability of acquisition likelihood for some firms. *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively. (1) (2) (3) (4) (5) (6) (7) (8) (9) Decdummy1 0.032*** 0.034*** 0.023** (5.13) (2.81) (2.23) Decdummy2 0.057*** 0.040** 0.024** (6.97) (2.18) (2.20) Decdummy3 0.078*** 0.048** 0.033* (8.16) (2.24) (1.96) R&D -0.032 -0.181*** -0.031 -0.180*** -0.034 -0.181*** (-0.98) (-4.12) (-0.96) (-4.03) (-0.99) (-4.06) Cash 0.016 0.151*** 0.016 0.151*** 0.016 0.151*** (1.44) (4.16) (1.44) (4.18) (1.45) (4.20) Size 0.023*** -0.008 0.023*** -0.008 0.023*** -0.008 (5.87) (-0.87) (5.66) (-0.86) (5.70) (-0.83) ROA 0.035 0.036** 0.035 0.036** 0.035 0.036** (1.47) (2.20) (1.49) (2.19) (1.46) (2.20) Market-to-Book 0.008* 0.007*** 0.009* 0.007*** 0.008* 0.007*** (1.95) (3.30) (2.00) (3.25) (1.99) (3.28) Debt/Equity -0.013*** -0.021*** -0.013*** -0.020*** -0.013*** -0.020*** (-5.32) (-4.81) (-5.01) (-4.78) (-5.13) (-4.78) Return volatility 0.025 -0.357** 0.024 -0.362** 0.019 -0.365** (0.13) (-2.53) (0.13) (-2.57) (0.10) (-2.56) Net loss -0.025*** -0.037*** -0.025*** -0.037*** -0.025*** -0.037***

T P

I R

C S

A

U N

D E

T P

C A

E C

M

ACCEPTED MANUSCRIPT

Industry leverage Industry cash # firms Industry ROA Constant Firm fixed effects Year fixed effects Adj. R2 N

0.115*** (39.13) No No 0.002 15799

(-8.31) -0.176** (-2.28) -0.023 (-0.32) 0.001*** (19.35) 0.087* (1.97) -0.008 (-0.22) No No 0.048 15799

(-5.21) -0.051 (-0.71) -0.099 (-1.06) 0.000 (0.99) 0.057 (1.33)

Yes Yes 0.213 15282

D E

T P

C A

E C

0.116*** (41.79) No No 0.003 15799

(-8.10) -0.171** (-2.24) -0.027 (-0.40) 0.001*** (19.78) 0.090* (2.02) -0.003 (-0.08) No No 0.048 15799

Yes Yes 0.213 15282

(-7.83) -0.166** (-2.18) -0.037 (-0.57) 0.001*** (14.95) 0.081* (2.04) 0.000 (0.01) No No 0.048 15799

T P

I R

C S

U N

A

M

(-5.41) -0.049 (-0.68) -0.102 (-1.11) 0.000 (0.78) 0.058 (1.31)

0.116*** (42.71) No No 0.004 15799

(-5.48) -0.047 (-0.66) -0.108 (-1.16) 0.000 (0.70) 0.049 (1.13)

Yes Yes 0.213 15282

ACCEPTED MANUSCRIPT Table 5 Foreign competition and likelihood of acquisitions within industry acquisitions This table reports results from tests that examine the association between increased foreign competition and firm propensities by type of acquisitions. The dependent variable in Panel A is the likelihood of acquisition within the same 2 digit SIC industry in year t. The dependent variable in Panel B is the likelihood of acquiring a target outside the bidder firm’s 2 digit SIC industry in year t. Decdummy1, Decdummy2, and Decdummy3 take a value of one for industry-years that face decreases in ad-valorem import duty rates by 2%, 20%, and 30% or more, respectively. Control variables are defined in Table 4 and are measured in year t-1. Standard errors are corrected for heteroskedasticity and are clustered at the 3-digit NAICS industry level (t-statistics are in parentheses). Models 3, 6, and 9 include firm and year fixed effects. Number of observations varies across models due to perfect predictability of acquisition likelihood for some firms. *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively. Panel A: Dependent variable is same industry acquisitions (1) (2) (3) (4) (5) (6) (7) (8) (9) Decdummy1 0.032*** 0.040** 0.026*** (6.27) (2.08) (2.84) Decdummy2 0.062*** 0.069* 0.025** (9.39) (2.00) (2.09) Decdummy3 0.077*** 0.040* 0.030* (10.01) (2.02) (1.79) R&D 0.056 -0.115*** 0.047 -0.114*** -0.042 -0.114*** (1.00) (-2.94) (0.91) (-2.97) (-1.09) (-2.99) Cash 0.051 0.082*** 0.047 0.082*** 0.007 0.082*** (1.24) (6.02) (1.27) (6.08) (0.34) (6.13) Size 0.017*** -0.002 0.017*** -0.001 0.015*** -0.001 (4.03) (-0.15) (3.91) (-0.12) (4.33) (-0.10) ROA 0.043* 0.023* 0.042* 0.023* 0.020 0.023* (1.87) (1.93) (1.89) (1.92) (1.35) (1.96) Market-to-Book 0.008** 0.004* 0.007** 0.004 0.005* 0.004 (2.50) (1.72) (2.51) (1.69) (1.83) (1.70) Debt/Equity -0.010** -0.011*** -0.010*** -0.011*** -0.006*** -0.011*** (-2.76) (-3.40) (-2.81) (-3.41) (-3.71) (-3.46) Return volatility 0.300 -0.204* 0.266 -0.211* -0.107 -0.214* (1.61) (-1.86) (1.58) (-1.91) (-0.82) (-1.90) Net loss -0.016*** -0.026*** -0.015*** -0.027*** -0.019*** -0.027*** (-3.30) (-3.49) (-3.18) (-3.49) (-4.10) (-3.54) Industry leverage -0.036 -0.033 -0.037 -0.031 (-0.69) (-0.65) (-0.61) (-0.61) Industry cash -0.013 -0.017 0.048 -0.022

T P

I R

C S

A

U N

D E

T P

C A

E C

M

ACCEPTED MANUSCRIPT (-0.36) 0.000 (1.27) 0.078** (2.10)

# firms Industry ROA Constant Firm fixed effects Year fixed effects Adj. R2 N

0.070*** (29.16) No No 0.002 15799

-0.056* (-1.75) No No 0.026 15799

(-0.45) 0.000 (1.03) 0.078** (2.13) 0.070*** (30.96) No No 0.005 15799

Yes Yes 0.181 15282

Panel B: Dependent variable is probability of conglomerate acquisitions (1) Decdummy1

(2)

(3)

(4)

0.000

0.003

-0.003

(0.06)

(0.76)

(-0.63)

Decdummy2

M

(-1.01)

D E

Decdummy3 R&D

T P

0.009

ROA Market-to-Book Debt/Equity

-0.001

(0.34)

(-0.14)

(7)

(8)

(9)

0.000

0.008*

0.003

(0.06)

(1.72)

(0.33)

-0.066

0.009

-0.066

(-1.14)

(0.39)

(-1.14)

(0.38)

(-1.14)

0.069**

0.009

0.069**

0.009

0.069**

(2.47)

(0.70)

(2.48)

(0.70)

(2.48)

0.008***

-0.007

0.008***

-0.007

0.008***

-0.007

(3.03)

(-1.47)

(3.00)

(-1.47)

(3.02)

(-1.46)

E C 0.009

C A

0.001

Yes Yes 0.181 15282

0.009

(0.70)

Size

I R

(-0.61) 0.000 (1.00) 0.071** (2.21)

-0.066

(0.39) Cash

A

(6)

0.070*** (31.98) No No 0.006 15799

T P

Yes Yes 0.181 15282

C S

U N

(5)

-0.005

-0.053 (-1.69) No No 0.029 15799

(0.79) 0.001*** (11.54) 0.098** (2.22) -0.037 (-1.36) No No 0.049 15799

0.015

0.013

0.015

0.013

0.015

0.013

(1.43)

(1.27)

(1.44)

(1.27)

(1.43)

(1.27)

0.003*

0.002

0.003*

0.002

0.003*

0.002

(1.84)

(1.29)

(1.83)

(1.29)

(1.86)

(1.29)

-0.007***

-0.009***

-0.007***

-0.009***

-0.007***

-0.009***

(-3.71)

(-3.65)

(-3.69)

(-3.64)

(-3.73)

(-3.61)

ACCEPTED MANUSCRIPT Return volatility Net loss Industry leverage Industry cash # firms Industry ROA Constant

0.125

-0.153**

0.124

-0.152**

0.126

-0.151**

(0.99)

(-2.49)

(0.97)

(-2.51)

(0.99)

(-2.50)

-0.007*

-0.010

-0.007*

-0.010

-0.007*

-0.010

(-2.02)

(-1.30)

(-2.06)

(-1.30)

(-2.01)

(-1.29)

-0.131***

-0.015

-0.131***

-0.015

-0.129***

-0.016

(-3.07)

(-0.39)

(-3.05)

(-0.40)

(-3.08)

(-0.41)

-0.084**

-0.086

-0.085**

-0.086

-0.085**

-0.086

(-2.20)

(-1.29)

(-2.29)

(-1.29)

(-2.34)

(-1.29)

-0.000*

-0.000

-0.000*

-0.000

-0.000*

-0.000

(-1.74)

(-0.46)

(-1.79)

(-0.45)

(-1.89)

(-0.53)

-0.016

-0.020

-0.016

-0.021

-0.017

-0.022

(-0.70)

(-0.67)

(-0.70)

(-0.70)

(-0.77)

(-0.77)

0.045***

0.037**

0.046***

(24.25)

(2.54)

Firm fixed effects

No

No

Yes

No

Year fixed effects

No

No

Yes

No

Adj. R

-0.000

0.011

0.125

N

15799

15799

15282

2

U N

A

I R

C S

0.038**

(26.15)

T P

0.037**

(26.29)

(2.58)

No

Yes

No

No

Yes

No

Yes

No

No

Yes

0.000

0.011

0.125

-0.000

0.011

0.125

15799

15799

15282

15799

15799

15282

D E

M

(2.66)

0.045***

T P

E C

C A

Table 6 Causal impact of increased foreign competition on firms’ acquisition propensities This table reports results from LPM regressions of the indicators for the timing of acquisitions on indicators for the timing of the tariff cuts. The sample period is from 1998 to 2014. Panel A presents results from regressions of acquisition likelihood (acq) on the timing of tariff cuts. Decdummy3 t-1 , Decdummy3 t , and Decdummy3 t+1 are equal to one if the 6- digit NAICS industry faces a tariff cut of 30% or more in year t-1, year t, or year t+1, respectively. Panel B presents results from falsification tests where acquisitions are assigned to random firm-years. Other control variables are defined in Table 4. All models include firm and

ACCEPTED MANUSCRIPT year fixed effects. Standard errors are corrected for heteroscedasticity and are clustered at the 3-digit NAICS industry level (t-statistics are in parentheses). Number of observations varies across models due to perfect predictability of acquisition likelihood for some firms. *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively.

T P

I R

C S

A

U N

D E

T P

C A

E C

M

ACCEPTED MANUSCRIPT Panel A: Timing of tariff cuts and firm acquisition propensities -0.009

(-0.80)

(-0.73) 0.033*

0.028*

(1.91)

(1.93)

(1.87)

0.007

0.005

(0.61)

(0.42)

-0.183***

-0.172***

-0.173***

(-4.28)

(-3.75)

(-4.01)

0.153***

0.151***

0.154***

(4.16)

(4.07)

(3.94)

-0.007

-0.005

-0.005

(-0.83)

(-0.57)

(-0.58)

0.036**

0.036**

0.037*

(2.14)

(2.07)

0.007***

0.007***

(3.45)

(4.45)

-0.021***

-0.021***

-0.021***

(-4.81)

(-4.98)

(-5.01)

-0.374**

-0.354**

-0.364**

(-2.59)

(-2.73)

(-2.71)

-0.036***

-0.041***

-0.040***

R&D Cash Size ROA Market-to-Book Debt/Equity Return volatility

(2.02)

0.007*** (4.72)

(-0.62)

ED

Net loss

US

Decdummy3t+1

(-0.60)

(-0.54)

-0.088

-0.083

-0.066

(-0.90)

(-0.96)

(-0.71)

0.000

0.000

0.000

(1.00)

(0.56)

(0.79)

0.061

0.052

0.064

(1.31)

(1.17)

(1.28)

Firm fixed effects

Yes

Yes

Yes

Year fixed effects

Yes

Yes

Yes

0.214

0.220

0.220

15167

14607

14496

(-5.41) -0.044

PT

Industry leverage

# firms

2

Adj. R N

AC

Industry ROA

CE

Industry cash

CR

0.028*

T

-0.009

AN

Decdummy3t

(3)

M

Decdummy3t-1

(2)

IP

(1)

(-5.97)

(-5.81)

-0.046

-0.041

Panel B: Falsification tests Decdummy1 Decdummy2

(1) -0.008 (-0.84)

(2)

-0.014

(3)

ACCEPTED MANUSCRIPT (-1.05)

Size ROA Market-to-Book Debt/Equity Return volatility Net loss

AN

Industry leverage Industry cash

M

# firms

ED

Industry ROA

T

Cash

0.065** (2.45) -0.032 (-1.01) 0.013 (1.51) -0.004 (-0.31) 0.003*** (2.93) 0.002 (0.61) -0.121 (-0.89) -0.015** (-2.58) 0.134** (2.20) 0.023 (0.41) -0.000 (-0.13) 0.046 (1.04)

CR

0.065** (2.47) -0.032 (-1.01) 0.013 (1.50) -0.004 (-0.31) 0.003*** (2.91) 0.002 (0.58) -0.122 (-0.90) -0.015** (-2.55) 0.134** (2.23) 0.022 (0.38) -0.000 (-0.24) 0.045 (1.07)

US

R&D

-0.018 (-1.03) 0.065** (2.50) -0.032 (-1.01) 0.013 (1.51) -0.003 (-0.29) 0.003** (2.80) 0.002 (0.60) -0.120 (-0.89) -0.015** (-2.54) 0.133** (2.19) 0.026 (0.47) -0.000 (-0.08) 0.050 (1.07)

IP

Decdummy3

Yes

Yes

Yes

Year fixed effects

Yes

Yes

Yes

0.120 15282

0.121 15282

0.121 15282

AC

CE

Adj. R2 N

PT

Firm-year fixed effects

ACCEPTED MANUSCRIPT Table 7 Robustness of main results to alternate model specifications This table reports results from analyses of the results in Table 4 with alternate model specifications . The dependent variable is the likelihood of acquisition in year t. Decdummy1, Decdummy2, and Decdummy3 take a value of one for industry-years that face decreases in ad-valorem import duty rates by 2%, 20%, and 30% or more from the previous year, respectively. Control variables are defined in Table 4 and are measured in year t -1. Standard errors are corrected for heteroskedasticity and are clustered at the 3-digit NAICS industry level (t-statistics are in parentheses). Models 1 through 3 report results from Linear Probability Models and include firm and industry-year fixed effects. Models 4 through 6 report results from Conditional Logit regressions and include firm and year fixed effects. Standard errors are corrected for heteroscedasticity and are clustered at the 3-digit NAICS industry level (t-statistics are in parentheses). Number of observations varies across models due to perfect predictability of acquisition likelihood for some firms. *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively. Linear Probability Model (1) Decdummy1

(2)

Conditional Logit Model

(3)

(4)

(3.12)

U N

0.023** (2.43)

Decdummy3 R&D

-0.196***

(2.41)

(-2.75)

(-2.60)

(-2.62)

0.151***

0.151***

1.829***

1.841***

1.845***

T P

(4.10)

(4.11)

(11.92)

(12.02)

(11.82)

-0.008

-0.008

0.065

0.069

0.074

(-0.89)

(-0.90)

(-0.89)

(0.49)

(0.50)

(0.53)

0.033**

0.033**

0.033**

1.109***

1.101***

1.097***

(2.08)

(2.08)

(2.09)

(2.87)

(2.85)

(2.83)

0.007***

0.007***

0.007***

0.035***

0.035***

0.035***

(3.57)

(3.53)

(3.54)

(3.87)

(3.64)

(3.58)

-0.021***

-0.021***

-0.021***

-0.924***

-0.924***

-0.925***

(-5.54)

(-5.56)

(-5.57)

(-4.35)

(-4.26)

(-4.21)

-0.392**

-0.393**

-0.392**

-4.508

-4.567

-4.638

(-2.51)

(-2.52)

(-2.53)

(-1.35)

(-1.35)

(-1.35)

-0.035***

-0.035***

-0.036***

-0.378***

-0.379***

-0.380***

(-5.40)

(-5.42)

(-5.40)

(-5.71)

(-5.62)

(-5.66)

-0.008

Net loss

0.316**

(2.94)

(-4.06)

Size

Return volatility

(3.86)

-1.639***

(4.09)

E C

C A

0.270***

-1.658***

D E

0.151***

Debt/Equity

(3.28)

-1.656***

(-4.11)

Market-to-Book

0.277***

0.034***

M

-0.196***

(6)

-0.196***

Cash

ROA

A

(5)

C S

0.022***

Decdummy2

I R

T P

(-4.07)

ACCEPTED MANUSCRIPT Industry leverage

-0.058

-0.057

-0.056

0.300

0.345

0.359

(-0.78)

(-0.77)

(-0.76)

(0.30)

(0.33)

(0.34)

-0.143

-0.149

-0.150

-0.713

-0.768

-0.848

(-1.36)

(-1.43)

(-1.44)

(-0.92)

(-0.97)

(-1.07)

-0.000

-0.000

-0.000

-0.000

-0.001

-0.001

(-0.48)

(-0.53)

(-0.57)

(-0.28)

(-0.40)

-0.025

-0.027

-0.025

0.092

0.103

(-0.36)

(-0.36)

(-0.33)

(0.15)

(0.16)

0.221***

0.225***

0.225***

(3.26)

(3.31)

(3.32)

Firm fixed effects

Yes

Yes

Yes

Year fixed effects

No

No

No

Industry-year fixed effects

Yes

Yes

Yes

0.213

0.213

0.213

Industry cash # firms Industry ROA Constant

2

Adj. R

Pseudo R2 N

15270

D E

T P

C A

E C

15270

M

15270

0.005

(0.01)

Yes

C S

Yes

Yes

No

No

No

0.078

0.078

0.078

7177

7177

7177

Yes

U N

A

I R

T P (-0.46)

Yes

Yes

ACCEPTED MANUSCRIPT

CR

IP

T

Table 8 Robustness of main results to excluding industries with largest representations This table presents results from additional robustness tests of the main results on subsamples after excluding industries with the largest representations . Panel A provides results from the analyses in Table 4 after excluding firms in the Chemical Manufacturing Industry (NAICS 3-digit industry 325). Panel B presents results from the analyses in Table 4 after excluding firms in the Computer and Electronic Product Manufacturing Industry (NAICS 3-digit industry 334), and Panel C excludes firms in the Publishing Industry (except Internet) (NAICS 3-digit industry 511). Panel D presents results from running the analyses in Table 4 on only the NAICS industries 325, 334, and 511. The dependent variable, Acq, takes a value of one for firms that make an acquisition in year t. Decdummy1, Decdummy2, and Decdummy3 take a value of one for industry-years that face decreases in ad-valorem import duty rates by 2%, 20%, and 30% or more from the previous year, respectively. Control variables are defined in Table 4 and are measured in year t-1. Standard errors are corrected for heteroskedasticity and are clustered at the 3-digit NAICS industry level (t-statistics are in parentheses). Models 3,6, and 9 include firm and year fixed effects. Number of observations varies across models due to perfect predictability of acquisition likelihood for some firms. *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively. (1) Decdummy1

(2)

(3)

0.028***

0.030**

0.022**

(4.16)

(2.45)

(2.20)

Decdummy2

(4)

(5)

0.048***

US

Panel A: Main results after excluding Chemical Manufacturing Industry (NAICS 3-digit industry 325)

0.006

-0.199***

M

Decdummy3 R&D

0.031*

0.021**

(1.72)

(2.17)

AN

(5.66)

(6)

(7)

(8)

0.069***

0.038*

(7.01)

(9)

0.029*

(1.78)

(1.79)

0.006

-0.198***

0.004

-0.199***

(0.14)

(-3.30)

(0.09)

(-3.35)

0.005

0.176***

0.005

0.176***

(-3.38)

0.005

0.176***

(0.45)

(7.49)

(0.46)

(7.58)

(0.46)

(7.58)

Size

0.024***

-0.015

0.024***

-0.015

0.024***

-0.014

(5.44)

(-1.54)

(5.24)

(-1.54)

(5.27)

(-1.51)

0.070***

0.053*

0.070***

0.053*

0.070***

0.052*

(5.15)

(1.77)

(5.20)

(1.76)

(5.06)

(1.76)

0.012***

0.008***

0.012***

0.008***

0.012***

0.008***

(3.44)

(8.17)

(3.51)

(7.92)

(3.52)

(8.00)

-0.013***

-0.020***

-0.013***

-0.020***

-0.013***

-0.020***

(-4.68)

(-4.34)

(-4.45)

(-4.29)

(-4.53)

(-4.28)

-0.086

-0.498***

-0.089

-0.504***

-0.092

-0.507***

(-0.45)

(-7.40)

(-0.47)

(-7.51)

(-0.48)

(-7.19)

-0.023***

-0.033***

-0.023***

-0.033***

-0.023***

-0.033***

(-9.03)

(-3.97)

(-8.60)

(-4.14)

(-8.04)

(-4.19)

-0.046

-0.086

-0.042

-0.083

-0.037

-0.081

(-0.78)

(-1.27)

(-0.69)

(-1.22)

(-0.62)

(-1.21)

Debt/Equity Return volatility Net Loss Industry leverage Industry Cash # firms

PT

CE

Market-toBook Assets

AC

ROA

ED

(0.15) Cash

0.080

-0.178**

0.075

-0.180**

0.066

-0.187***

(0.90)

(-2.72)

(0.85)

(-2.81)

(0.79)

(-2.91)

0.001***

0.000*

0.001***

0.000*

0.001***

0.000

ACCEPTED MANUSCRIPT

Industry ROA Constant

(11.08)

(1.89)

(9.66)

(1.80)

(8.58)

(1.67)

0.058

0.063

0.062

0.064

0.055

0.055

(1.24)

(1.16)

(1.27)

(1.16)

(1.31)

(1.00)

-0.050

0.124***

-0.046

0.124***

-0.043

(37.19)

(-1.51)

(39.84)

(-1.35)

(40.76)

(-1.31)

No

No

Yes

No

No

Yes

No

No

Yes

No

No

Yes

No

No

Yes

No

No

Yes

Adj. R

0.001

0.052

0.224

0.002

0.051

0.224

0.004

0.051

0.224

N

13463

13463

13019

13463

13463

13019

13463

13463

13019

IP

2

CR

Firm fixed effects Year fixed effects

T

0.124***

Panel B: Main results after excluding Computer and Electronic Product Manufacturing Industry (NAICS 3-digit industry 334) (3)

(4)

0.050***

0.042***

0.034***

(6.68)

(3.33)

(4.31)

Decdummy2

(6)

0.092***

0.051**

0.028*

(9.18)

(2.30)

(1.90)

M

Decdummy3 R&D

(5)

(7)

(8)

(9)

US

(2)

AN

Decdummy1

(1)

0.117*** (10.04)

0.060**

0.029

(2.33)

(1.08)

-0.053

-0.130**

-0.061

-0.131**

(-2.38)

(-1.32)

(-2.35)

(-1.55)

(-2.39)

0.110***

0.016

0.111***

0.017

0.111***

-0.132**

(-1.37) Cash

0.017 (0.92)

(3.77)

(0.90)

(3.76)

(0.95)

(3.80)

Size

0.020***

-0.017

0.019***

-0.016

0.019***

-0.016

(4.05)

(-1.34)

(3.92)

(-1.29)

(3.91)

(-1.29)

0.046*

0.019

0.046*

0.017

0.045*

(1.92)

(0.99)

(1.91)

(0.91)

(1.90)

PT

0.018

CE

ROA

(0.95)

Debt/Equity Return volatility Net Loss Industry leverage Industry Cash

0.003

0.004

0.003

0.004

0.003

0.004

(1.02)

(1.37)

(1.07)

(1.33)

(1.07)

(1.32)

-0.011***

-0.018***

-0.011***

-0.018***

-0.011***

-0.018***

(-8.44)

(-4.80)

(-7.75)

(-4.90)

(-8.01)

(-4.97)

-0.231

-0.362*

-0.230

-0.369*

-0.237

-0.376*

(-1.45)

(-2.00)

(-1.45)

(-2.02)

(-1.45)

(-2.01)

-0.028***

-0.031***

-0.027***

-0.032***

-0.028***

-0.032***

(-7.40)

(-3.23)

(-6.59)

(-3.39)

(-6.61)

(-3.45)

-0.148*

0.005

-0.150*

0.004

-0.141*

0.007

(-1.87)

(0.06)

(-1.79)

(0.05)

(-1.73)

(0.09)

0.008

-0.032

-0.003

-0.035

-0.019

-0.037

AC

Market-toBook Assets

ED

-0.053

ACCEPTED MANUSCRIPT

# firms Industry ROA

(-0.03)

(-0.34)

(-0.21)

(-0.36)

0.000

0.001***

0.000

0.001***

0.000

(13.26)

(0.64)

(11.23)

(0.43)

(9.84)

(0.38)

0.095

0.052

0.091

0.049

0.070

0.038

(1.20)

(0.92)

(1.15)

(0.84)

(0.95)

(0.63)

0.106***

0.015

0.107***

0.024

0.108***

0.027

(30.36)

(0.32)

(32.81)

(0.54)

(33.69)

(0.61)

No

No

No

No

No

No

Yes

Yes

T

Firm fixed effects Year fixed effects

(-0.30)

IP

Constant

(0.09) 0.001***

Yes

No

Yes

No

No

Yes

No

No

Yes

0.004

0.049

0.213

0.008

0.048

0.212

0.009

0.048

0.212

N

10758

10758

10380

10758

10758

10758

10758

10380

(7)

(8)

(9)

0.021*

0.023*

0.016

CR

No

Adj. R2

US

10380

Panel C: Main results after excluding Publishing (except Internet) Industry (NAICS 3-digit industry 511) (2)

(3)

0.017***

0.024***

0.016**

(2.59)

(3.34)

(2.22)

Decdummy3

(6)

0.019**

0.021**

0.012**

(2.14)

(2.23)

(2.17) (2.03)

(0.93)

-0.011

-0.170***

-0.011

-0.170***

-0.013

-0.170***

ED

R&D

(5)

M

Decdummy2

(4)

AN

Decdummy1

(1)

(1.88)

(-3.46)

(-0.33)

(-3.45)

(-0.38)

(-3.45)

0.009

0.155***

0.009

0.155***

0.009

0.155***

(3.46)

(0.81)

(3.49)

(0.83)

(3.51)

Size

0.021***

-0.004

0.020***

-0.004

0.020***

-0.004

(5.07)

(-0.51)

(4.95)

(-0.49)

(4.98)

(-0.48)

0.037

0.036*

0.038

0.036*

0.037

0.036*

(1.42)

(1.82)

(1.43)

(1.83)

(1.41)

(1.84)

0.009

0.006**

0.009

0.006*

0.009

0.006*

(1.60)

(2.10)

(1.61)

(2.07)

(1.60)

(2.07)

-0.013***

-0.021***

-0.013***

-0.021***

-0.013***

-0.021***

(-5.92)

(-4.95)

(-5.70)

(-4.91)

(-5.74)

(-4.90)

0.109

-0.331**

0.104

-0.336**

0.100

-0.339**

(0.61)

(-2.21)

(0.58)

(-2.25)

(0.56)

(-2.25)

-0.026***

-0.035***

-0.026***

-0.035***

-0.026***

-0.035***

(-6.63)

(-4.45)

(-6.49)

(-4.63)

(-6.33)

(-4.64)

-0.161**

-0.058

-0.157**

-0.055

-0.157**

-0.054

(-2.31)

(-0.78)

(-2.25)

(-0.75)

(-2.24)

(-0.74)

-0.069

-0.117

-0.074*

-0.120

-0.081*

-0.122

Market-toBook Assets Debt/Equity Return volatility Net Loss Industry leverage Industry

AC

ROA

CE

(0.81)

PT

(-0.34) Cash

ACCEPTED MANUSCRIPT Cash

# firms Industry ROA

(-1.17)

(-1.81)

(-1.21)

(-2.01)

(-1.22)

0.001***

0.000

0.001***

0.000

0.001***

0.000

(4.20)

(0.21)

(3.83)

(0.12)

(4.05)

(0.11)

0.039

0.034

0.040

0.034

0.036

0.031

(1.04)

(0.74)

(0.97)

(0.70)

(0.88)

(0.64)

0.011

0.108***

0.016

0.108***

0.018

(35.96)

(0.32)

(39.04)

(0.46)

(40.09)

(0.52)

No

No

Yes

No

No

Yes

No

No

Yes

No

No

Yes

No

No

Yes

No

No

Yes

Adj. R

0.000

0.032

0.189

0.000

0.032

0.189

0.000

0.032

0.189

N

14194

14194

13743

14194

14194

13743

14194

14194

13743

(6)

(7)

(8)

(9)

0.096***

0.066

0.053***

(8.24)

(1.86)

(3.73)

CR

2

US

Firm fixed effects Year fixed effects

IP

0.106***

T

Constant

(-1.65)

Panel D: Main results for only firms in NAICS 3-digit industries 325, 334, and 511 (3)

0.038***

0.037

0.022**

(4.24)

(1.34)

(2.06)

(4)

(5)

Decdummy2

0.073***

0.057

0.043***

(1.55)

(3.16)

Decdummy3

M

Decdummy1

(2)

AN

(1)

Cash

-0.038

-0.191***

-0.040

-0.192***

-0.037

-0.192***

(-0.92)

(-3.05)

(-0.93)

(-3.06)

(-0.83)

(-3.06)

0.025

0.157***

0.026

0.156***

0.025

0.156***

(5.72)

(1.25)

(5.71)

(1.23)

(5.70)

0.001

0.030**

0.001

0.030*

0.001

(0.11)

(4.30)

(0.11)

(4.28)

(0.15)

Size

0.030**

ROA

CE

(1.24)

PT

R&D

ED

(6.92)

(4.43)

Debt/Equity Return volatility Net Loss Industry leverage

0.019

0.025

0.019

0.026

0.019

(1.20)

(0.93)

(1.18)

(0.95)

(1.18)

0.010

0.007***

0.010

0.007***

0.010

0.007***

(2.03)

(3.74)

(2.07)

(3.70)

(2.09)

(3.72)

AC

Market-toBook Assets

0.025

(0.94)

-0.020

-0.022***

-0.021

-0.022***

-0.021

-0.022***

(-1.80)

(-3.01)

(-1.77)

(-3.01)

(-1.79)

(-3.02)

0.375

-0.183

0.386

-0.192

0.384

-0.190

(1.88)

(-0.80)

(2.12)

(-0.84)

(-0.83)

-0.024***

-0.048***

-0.023***

-0.047***

(2.17) 0.024***

-0.047***

(-19.92)

(-5.18)

(-14.55)

(-5.13)

(-12.80)

(-5.13)

-0.369

-0.114

-0.351

-0.115

-0.354

-0.109

ACCEPTED MANUSCRIPT

Industry ROA Constant

(-1.19)

(-2.42)

(-1.20)

(-2.34)

(-1.14)

-0.038

-0.055

-0.040

-0.061

-0.048

-0.070

(-0.41)

(-0.67)

(-0.51)

(-0.75)

(-0.67)

(-0.85)

0.001**

0.000

0.001**

-0.000

0.001**

-0.000

(9.80)

(0.04)

(6.24)

(-0.07)

(5.56)

(-0.10)

0.110

0.034

0.114

0.026

0.112

0.025

(1.74)

(0.56)

(2.25)

(0.42)

(2.57)

(0.40)

0.126***

-0.024

0.124***

-0.026

(32.09)

(-0.61)

(33.02)

(-0.64)

No

No

Yes

No

No

No

No

Yes

No

No

Adj. R

0.002

0.062

0.238

0.005

0.063

N

9302

9302

8991

9302

9302

-0.019

(33.34)

(-0.52)

No

No

Yes

Yes

No

No

Yes

0.239

0.007

0.064

0.239

8991

9302

9302

8991

CR

Yes

US

AN M ED PT CE

2

AC

Firm fixed effects Year fixed effects

0.124***

T

# firms

(-2.54)

IP

Industry Cash

ACCEPTED MANUSCRIPT Table 9 Likelihood of acquisition for all firms, including industries with less than three firms This table reports results from tests that examine the association between increased foreign competition and firm propensities for all firms, including industries with less than three firms. The dependent variable takes a value of one for firms that acquire in year t, zero otherwise. Decdummy1, Decdummy2, and Decdummy3 take a value of one for industry-years that face decreases in ad-valorem import duty rates by 2%, 20%, and 30% or more, respectively. Control variables are defined in Table 4 and are measured in year t-1. Standard errors are corrected for heteroskedasticity and are clustered at the 3-digit NAICS industry level (t-statistics are in parentheses). Models 3, 6, and 9 include firm and year fixed effects. Number of observations varies across models due to perfect predictability of acquisition likelihood for some firms. *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively. Decdummy1

(1)

(2)

(3)

0.029***

0.030***

0.022**

(5.12)

(2.81)

(2.29)

Decdummy2

(4)

(5)

(6)

0.059***

0.039**

0.025**

(7.70)

(2.21)

(2.34)

R&D

-0.023

-0.189***

0.079*** (8.84)

(8)

(9)

0.047**

0.033**

(2.27)

(2.27)

-0.024

-0.188***

-0.027

-0.189***

(-0.65)

(-4.45)

(-0.69)

(-4.48)

0.013

0.144***

0.013

0.144***

A

M

C S

U N

Decdummy3

T P

I R (7)

(-0.63)

(-4.53)

Cash

0.013

0.144***

(1.23)

(3.91)

(1.23)

(3.91)

(1.25)

(3.93)

Size

0.021***

-0.012

0.020***

-0.012

0.020***

-0.011

(-1.37)

(4.96)

(-1.36)

(5.00)

(-1.34)

0.038**

0.040

0.038**

0.040

0.038**

(2.37)

(1.70)

(2.35)

(1.66)

(2.38)

0.008*

0.007***

0.008*

0.007***

0.008*

0.007***

(1.94)

(3.30)

(1.98)

(3.27)

(1.98)

(3.29)

-0.012***

-0.017***

-0.012***

-0.017***

-0.012***

-0.017***

(-5.66)

(-5.20)

(-5.35)

(-5.12)

(-5.53)

(-5.15)

0.075

-0.281**

0.073

-0.287**

0.068

-0.290**

(0.39)

(-2.62)

(0.38)

(-2.68)

(0.35)

(-2.67)

-0.025***

-0.037***

-0.025***

-0.037***

-0.026***

-0.037***

(-7.80)

(-5.71)

(-7.56)

(-5.76)

(-7.50)

(-5.84)

T P

(5.12) ROA

0.040

E C (1.67)

Market-toBook Assets Debt/Equity Return volatility Net Loss

C A

D E

ACCEPTED MANUSCRIPT Industry leverage

-0.136**

-0.090**

-0.135**

-0.090**

-0.131**

-0.088**

(-2.63)

(-2.24)

(-2.61)

(-2.24)

(-2.56)

(-2.18)

-0.025

-0.106*

-0.027

-0.108*

-0.035

-0.112*

(-0.45)

(-1.71)

(-0.51)

(-1.76)

(-0.67)

(-1.80)

0.001***

0.000

0.001***

0.000

(17.71)

(1.63)

(18.72)

(1.38)

0.051

0.048

0.054

0.048

(1.36)

(1.59)

(1.46)

(1.56)

Industry Cash # firms Industry ROA Constant Firm fixed effects Year fixed effects 2

0.109***

-0.006

0.109***

-0.002

(41.15)

(-0.20)

(43.97)

(-0.05)

No

No

Yes

No

No

No

No

Yes

No

Adj. R

0.001

0.044

0.201

N

18428

18428

17930

D E

T P

C A

E C

0.003

M

18428

No

0.000

I R

(14.75)

(1.31)

0.047

0.041

(1.42)

(1.41)

0.110***

0.001

(44.91)

(0.03)

Yes

No

No

Yes

Yes

No

No

Yes

C S

U N

A

T P

0.001***

0.044

0.201

0.004

0.044

0.201

18428

17930

18428

18428

17930

ACCEPTED MANUSCRIPT Table 10 Tariff cuts and acquisitions by efficiency gains This table reports results from tests that examine the association between increased foreign competition and firm propensitie s by firm characteristics. The dependent variable, Acq, takes a value of one for firms that make an acquisition in year t. Single Segment takes a value of one for firms operating in a single segment, zero for multi-segment firms. Capital/Labor ratio is the ratio of capital expenditure (icapt) to number of employees (emp). Operating profit margin is the earnings before interest and taxes (ebit) plus depreciations (dp) divided by net sales (sale). Net profit margin is the ratio of net income (ni) to net sales (sale). Growth opportunities is the ratio of market value of assets (at-ceq+prcc_f*csho) to the book value of assets (at). Samples split by continuous variables are partitioned at the industry-year median values in year t-1. Decdummy3 takes a value of one for industry-years that face decreases in ad-valorem import duty rates by 30% or more, zero otherwise. Control variables are defined in Table 4 and are measured in year t-1. Standard errors are corrected for heteroskedasticity and are clustered at the 3-digit NAICS industry level (t-statistics are in parentheses). All models include firm and year fixed effects. Number of observations varies across models due to perfect predictability of acquisition likelihood for some firms. *, **, and *** denote significance at t he 10%, 5%, and 1% levels, respectively. Growth opportunities Single segment Capital/Labor ratio Operating profit margin Net profit margin

T P

I R

No Decdummy3 Cash Size ROA Market-to-Book Debt/Equity

Net loss Industry leverage Industry cash

Low

High

Low

High

-0.006

0.059*

-0.009

0.055**

0.024

(-0.28)

(1.77)

(-0.44)

(2.61)

(1.09)

0.289***

0.100*

0.156***

0.161***

0.139***

(5.89)

(1.81)

(3.33)

(3.11)

-0.020**

-0.004

-0.012

-0.007

(-2.77)

(-0.50)

(-0.93)

0.065

0.028***

0.020

(0.89)

(3.46)

(1.63)

D E

0.002

0.005*

(0.29) -0.012*

(-0.51)

C S

A

U N

Low

High

Low

High

0.036**

0.022

0.057**

0.048

0.026*

(2.12)

(1.15)

(2.46)

(1.59)

(1.83)

0.169***

0.109***

0.203***

0.162***

0.135***

(3.79)

(3.13)

(2.89)

(5.39)

(3.11)

(6.25)

-0.011

-0.024**

-0.009

-0.008

-0.011

0.006

(-0.95)

(-2.61)

(-0.71)

(-0.86)

(-0.96)

(0.53)

M

0.119*

0.035***

0.114

0.024*

0.099

0.124***

0.009

(2.04)

(3.58)

(1.67)

(1.86)

(1.22)

(3.31)

(0.78)

0.007***

0.005**

0.008***

0.005

0.006**

0.002

(1.95)

(4.53)

(2.22)

(4.80)

(1.09)

(2.16)

(0.65)

-0.019*

T P

E C

-0.021***

-0.020***

-0.016***

-0.026***

-0.018***

-0.025**

-0.015***

-0.028**

(-1.91)

(-3.86)

(-3.39)

(-2.96)

(-2.90)

(-4.86)

(-2.30)

(-4.19)

(-2.22)

-0.819***

-0.235

-0.329

-0.274

0.022

-1.410**

-0.090

-0.743

-0.084

-0.695**

(-3.40)

(-0.81)

(-1.29)

(-0.87)

(0.10)

(-2.65)

(-0.32)

(-1.50)

(-0.30)

(-2.14)

-0.029*

-0.038***

-0.039***

-0.028***

-0.028***

-0.039**

-0.029***

-0.031*

-0.024*

-0.044***

(-1.94)

(-4.88)

(-3.26)

(-3.20)

(-3.28)

(-2.31)

(-4.51)

(-1.77)

(-1.95)

(-4.15)

-0.208

0.086

0.102

-0.272**

-0.021

-0.046

-0.086

-0.014

0.012

-0.085

(-1.35)

(1.12)

(0.97)

(-2.16)

(-0.20)

(-0.57)

(-0.61)

(-0.18)

(0.14)

(-0.67)

-0.124

0.052

0.010

-0.140

-0.248**

0.046

-0.129

-0.049

-0.007

-0.113

(-1.88) Return volatility

Yes

C A

ACCEPTED MANUSCRIPT

# firms

(-1.17)

(0.59)

(0.10)

(-1.38)

(-2.75)

(0.35)

(-1.31)

(-0.31)

(-0.04)

(-0.92)

0.001***

-0.000

-0.000

0.000

0.000**

0.000

0.000

0.000

0.000

0.000

(4.55)

(-0.45)

(-0.65)

(1.01)

(2.12)

(1.18)

(0.52)

(0.30)

(0.26)

(1.39)

Industry RoA

0.118

0.113

0.020

0.054

0.095**

-0.003

0.034

0.026

0.050

0.018

(0.83)

(1.55)

(0.33)

(0.67)

(2.51)

(-0.03)

(0.64)

R&D

-0.153

-0.125**

-0.069

-0.221***

-0.077

-0.554***

-0.071

Firm fixed effects

(-0.45) Yes

(-2.29) Yes

(-0.88) Yes

(-6.16) Yes

(-0.92) Yes

(-3.50) Yes

(-0.78) Yes

Year fixed effects

Yes

Yes

Yes

Yes

Yes

Yes

Adj. R

0.253

0.241

N

5036

6708

0.191 7316

0.258 7490

0.191 7311

0.246 7439

2

D E

T P

C A

E C

M

(1.09)

(0.23)

-0.167*

-0.189*

(-4.97) Yes

(-1.97) Yes

(-1.76) Yes

Yes

Yes

Yes

Yes

0.224 7254

0.246

0.195

0.267

7399

7300

7287

T P

I R

C S

U N

A

(0.24)

-0.433***

ACCEPTED MANUSCRIPT Table 11 Tariff cuts and acquisition propensities by financial constraints and access to capital

This table reports results from tests that examine the association between increased foreign competition and firm propensitie s by indicators of firms’ financial constraints. The dependent variable, Acq, takes a value of one for firms that make an acquisition in year t. Cash holdings is the ratio of cash and marketable securities (che) to total assets (at). Book leverage is the book value of debt (dltt + dlc) divided by book value of assets (at). Bond rating and Investment grade rating take a value of one if a firm has an S&P rating, and has S&P rating of BBB- or better, respectively. Z score is defined as the sum of 3.3 times earnings before interest and taxes (ebit), 1.4 times retained earnings (re), net sales (sale), and 1.2 times working capital (wcap), all scaled by total assets (at). C&I spread is the Commercial and Industrial Loan rates spreads over the intended Fed Funds rate. Samples split by continuous variables are partitioned at the industry -year median values in year t-1. Decdummy3 takes a value of one for industry-years that face decreases in ad-valorem import duty rates by 30% or more, zero otherwise. Control variables are defined in Table 4 and are measured in year t -1. Standard errors are corrected for heteroskedasticity and are clustered at the 3digit NAICS industry level (t-statistics are in parentheses). All models include firm and year fixed effects. Number of observations varies across models due to perfect predictability of acquisition likelihood for some firms. *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively. Cash holdings Book leverage Bond rating Investment grade rating Z score C&I spread

T P

I R

Decdummy3

Low

High

Low

High

No

Yes

0.016 (0.89)

0.047** (2.15)

0.057** (2.27) 0.149***

0.008 (0.43) 0.216*

0.017 (0.71) 0.208***

0.048** (2.31) 0.169***

(1.81) -0.013 (-0.67)

(3.49) 0.010 (0.72)

Cash Size

-0.034*** (-2.90)

0.009 (0.78)

(6.50) -0.003 (-0.27)

ROA

0.054* (2.01) 0.009**

0.027 (1.48) 0.004*

0.023 (0.83) 0.003*

0.033 (1.17) 0.012***

D E

(2.51) -0.016*** (-3.97)

(1.95) -0.022* (-2.00)

(1.96) -0.034 (-1.28)

-0.680*** (-3.67) -0.051***

-0.002 (-0.01) -0.016*

(-3.48) -0.083 (-1.01) -0.044 (-0.42)

M arket-to-Book Debt/Equity Return volatility Net loss Industry leverage Industry cash

C S

U N

A

M

No

0.034* (1.99) 0.206***

Yes

Low

High

Low

High

0.052** (2.56) 0.153***

0.017 (0.77) 0.123**

0.049** (2.50) 0.175***

0.051*** (3.39) 0.215***

0.020 (0.93) 0.105**

(3.81) -0.030** (-2.26)

(3.38) -0.001 (-0.11)

(4.76) -0.025* (-1.83)

(2.66) -0.006 (-0.59)

(6.04) -0.030** (-2.72)

(4.82) -0.004 (-0.32)

(2.43) -0.002 (-0.21)

0.037*** (4.36) 0.008***

0.066 (1.37) 0.001

0.043** (2.60) 0.010***

0.022 (0.88) -0.001

0.035** (2.47) 0.007***

0.104* (1.83) 0.002

0.017 (0.38) -0.007***

0.050*** (2.97) 0.012*

(3.19) -0.014*** (-3.29)

(3.34) -0.024*** (-6.62)

(0.31) -0.019** (-2.52)

(4.38) -0.017*** (-4.09)

(-0.25) -0.105*** (-3.20)

(4.35) -0.019*** (-3.66)

(0.29) -0.021** (-2.67)

(-4.55) -0.018*** (-3.75)

(1.79) -0.027*** (-3.60)

-0.183 (-0.58) -0.004

-0.500*** (-2.96) -0.069***

-0.056 (-0.29) -0.040***

-0.771*** (-2.93) -0.041***

-0.201 (-1.39) -0.046***

-0.934*** (-3.10) -0.018

-0.082 (-0.42) -0.047***

-1.007*** (-4.12) -0.017

-0.445** (-2.18) -0.029

-0.331 (-0.57) -0.046***

(-1.90) 0.019 (0.18)

(-1.06) 0.082 (0.83)

(-4.34) -0.050 (-0.52)

(-6.53) 0.046 (0.55)

(-2.94) -0.170 (-1.55)

(-6.74) -0.041 (-0.46)

(-1.38) -0.069 (-0.87)

(-2.93) -0.044 (-0.34)

(-1.53) -0.015 (-0.20)

(-1.68) -0.013 (-0.11)

(-4.01) 0.013 (0.13)

-0.133 (-1.53)

-0.127 (-1.38)

0.022 (0.18)

-0.056 (-0.44)

-0.154 (-1.24)

-0.068 (-0.61)

-0.161 (-1.41)

-0.210** (-2.81)

-0.011 (-0.08)

-0.059 (-0.45)

-0.171 (-1.50)

T P

E C

C A

ACCEPTED MANUSCRIPT # firms Industry ROA

R&D Firm fixed effects Year fixed effects Adj. R2 N

-0.000 (-1.59)

0.000* (1.92)

0.000 (1.66)

-0.000*** (-2.86)

0.000* (1.99)

-0.000 (-0.29)

0.000 (1.55)

0.000 (0.05)

0.000 (0.62)

0.000 (0.93)

-0.000 (-0.71)

0.000 (1.41)

0.015 (0.20) -0.278***

0.099 (1.22) -0.191**

0.134 (1.57) -0.144***

0.029 (0.37) -0.261**

-0.080 (-1.53) -0.120**

0.134** (2.54) -0.152*

-0.003 (-0.05) -0.158***

0.084 (0.81) -0.176*

0.094* (1.77) -0.108**

-0.012 (-0.14) -0.624***

0.017 (0.47) -0.250***

0.016 (0.31) -0.170**

(-3.20) Yes Yes

(-2.51) Yes Yes

(-3.20) Yes Yes

(-2.26) Yes Yes

(-2.67) Yes Yes

(-1.87) Yes Yes

(-4.07) Yes Yes

(-1.83) Yes Yes

(-2.07) Yes Yes

(-4.35) Yes Yes

(-3.79) Yes Yes

(-2.27) Yes Yes

0.214 7375

0.232 7316

0.245 7451

0.230 7313

0.228 8505

0.249 6410

0.224 10156

0.259 4789

0.230 7268

0.244 7526

0.212 5829

0.238 7118

A

U N

D E

T P

C A

E C

M

C S

I R

T P

ACCEPTED MANUSCRIPT

Table 12 CARs to announcement of acquisitions in response to tariff cuts This table reports the cumulative abnormal returns (CARs) surrounding the announcement of acquisitions in the years of tariff cuts. The sample spans the 19982014 period. The CARs are calculated over the event windows [-1, +1], [-3, +3], and [-5, +5] using the three-factor model and CRSP value weighted market returns. Decdummy3 takes a value of one for industry-years that face decreases in ad-valorem import duty rates by 30% or more, zero otherwise. All cash deal takes a value of one for deals that are funded completely by cash, zero otherwise. All stock deal takes a value of one for deals funded entirely by stocks, zero otherwise. Relative size is the deal value scale by bidder’s market value of equity in year t-1. Public target takes a value of one if the target firms is public, zero otherwise. Foreign target takes a value of one if the target firm is a foreign firm. Conglomerate deal takes a value of one if the acquirer and target are in different 2-digit SIC industries. Market-to-Book assets, Debt/Equity, Industry leverage, and Industry cash are as defined in Table 4. Standard errors are corrected for heteroscedasticity (t-statistics are in parentheses). *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively.

T P

I R

C S

A

U N

D E

T P

C A

E C

M

ACCEPTED MANUSCRIPT [-1,+1] (1)

(2)

(3)

(4)

-0.001

-0.007

-0.005

-0.015

-0.010

(-0.17)

(-0.91)

(-0.58)

(-1.62)

(-0.99)

Public target Foreign target

0.006

0.016**

(2.12)

(0.98)

(2.20)

0.003

0.001

(1.78)

(0.37)

(0.13)

0.014***

0.013**

0.011*

(3.38)

(2.50)

-0.038***

-0.035***

(-7.15)

(-5.34)

-0.009

-0.012*

-0.011

(-1.61)

(-1.76)

(-1.35)

0.000 (0.28)

Market-to-Book assets

-0.001

AN

(-0.18) Debt/Equity

0.002 -0.028 (-1.38) -0.012

(-3.87)

0.001

-0.001

(0.58)

(-0.43)

-0.002

-0.002

(-0.36)

(-0.24)

0.001

0.002

(0.42)

(0.52)

-0.062**

-0.055*

(-2.43)

(-1.82)

-0.013

0.007

(-0.34)

(0.15)

(4.46)

(2.08)

(3.93)

(2.66)

(3.09)

(1.76)

-0.001

0.027

-0.000

0.018

0.001

0.012

1929

1926

1929

1926

1929

1926

0.021**

0.034***

0.011***

0.027*

PT

0.010***

CE AC

-0.030***

0.011***

(-0.40)

N

(1.74)

ED

Industry leverage

M

(0.89) Industry cash

T

0.011*

US

Conglomerate deal

0.011**

IP

Relative size

Adj. R

(6)

-0.000

All stock deal

2

(5)

(-0.03) All cash deal

Constant

[-5,+5]

CR

Decdummy3

[-3,+3]

ACCEPTED MANUSCRIPT

IP

T

Table 13 Long run returns to announcement of acquisitions in response to tariff cuts This table reports the long-run cumulative abnormal returns (CARs) surrounding the announcement of acquisitions in the years of tariff cuts. The sample spans the 1998-2014 period. The CARs are calculated over the event windows [-1, +60] and [-1, +120]. CARs in Panel A are calculated in excess of the three-factor model using CRSP value weighted market returns. CARs in Panel B are calculated using the market model. Decdummy3 takes a value of one for industry-years that face decreases in ad-valorem import duty rates by 30% or more, zero otherwise. All cash deal takes a value of one for deals that are funded completely by cash, zero otherwise. All stock deal takes a value of one for deals funded entirely by stocks, zero otherwise. Relative size is the deal value scale by bidder’s market value of equity in year t-1. Public target takes a value of one if the target firms is public, zero otherwise. Foreign target takes a value of one if the target firm is a foreign firm. Conglomerate deal takes a value of one if the acquirer and target are in different 2-digit SIC industries. Market-to-Book assets, Debt/Equity, Industry leverage, and Industry cash are as defined in Table 4. Standard errors are corrected for heteroscedasticity (t-statistics are in parentheses). *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively. [-1,+60] Decdummy3

-0.012 (-0.54)

All stock deal

M

Relative size

PT

ED

Public target Foreign target

(2) 0.013

(0.86)

(3)

(0.58)

-0.007

Conglomerate deal

CE

Market-to-Book assets

AC

Debt/Equity

Industry cash Industry leverage

0.035

(1.73)

(1.48)

-0.036*

-0.100***

(-1.69)

(-3.36)

-0.014

-0.045**

(-1.04)

(-2.39)

-0.014

0.022

(-0.76)

(0.91)

-0.032*

-0.058**

(-1.69)

(-2.23)

-0.007*

-0.008*

(-1.96)

(-1.70)

-0.005

0.002

(-0.29)

(0.09)

-0.012

-0.034***

(-1.58)

(-3.27)

-0.164**

-0.171*

(-2.39)

(-1.80)

0.023

0.106

(0.22) Constant 2

Adj. R N

(4)

0.029*

AN

All cash deal

0.027

(-0.22)

[-1,+120]

US

(1)

CR

Panel A: Long run returns using the three-factor model

(0.73)

-0.035***

0.033

-0.070***

0.005

(-4.52)

(0.95)

(-6.41)

(0.10)

-0.000

0.015

-0.000

0.025

1929

1926

1929

1926

Panel B: Long run returns using the market model

ACCEPTED MANUSCRIPT [-1,+60] (1)

(2)

(3)

(4)

0.031

0.056**

0.065**

0.097***

(1.36)

(2.36)

(2.05)

(2.96)

All cash deal All stock deal Relative size

0.039**

0.044*

(2.19)

(1.78)

-0.063***

-0.134***

(-2.78)

(-4.26)

-0.018

-0.058***

T

Decdummy3

[-1,+120]

-0.014 (-0.76)

Foreign target

-0.035* (-1.80) -0.001

US

Conglomerate deal

Debt/Equity

M

Industry cash

ED

Industry leverage

Adj. R

AC

CE

N

PT

2

(0.81)

-0.060** (-2.18) -0.000

(-0.39)

(-0.04)

-0.007

-0.006

(-0.46)

(-0.26)

AN

Market-to-Book assets

Constant

0.021

CR

Public target

(-2.91)

IP

(-1.23)

-0.013

-0.033***

(-1.64)

(-3.04)

-0.152**

-0.169*

(-2.13)

(-1.69)

0.087

0.146

(0.80)

(0.96)

-0.041***

0.006

-0.076***

-0.016

(-5.00)

(0.18)

(-6.63)

(-0.32)

0.000

0.018

0.002

0.029

1929

1926

1929

1926

ACCEPTED MANUSCRIPT Table 14 Post-acquisition performance This table reports acquirer firms’ performance in the year after the tariff cuts. ROA is the ratio of earnings before interest and taxes (ebit) to book value of assets (at). Net profit margin is the ratio of net income (ni) to net sales (sale). The dependent variables in Panel A are ROA and net profit margin in year t+1. The dependent variables in Panel B are ROA and net profit margins adjusted for industry averages in year t+1. Models 1 and 4 present firm

T

performance following all acquisitions in the years that bidder firms face tariff cuts. Models 2 and 5 show firm

IP

performance following same industry acquisitions in the years that bidders face tariff cuts. Models 3 and 6 report

CR

firm performance in the year following acquisitions where targets and bidders are in different 3-digit NAICS industries. Other control variables are defined in Table 4. All models include firm and year fixed effects. Standard

US

errors are corrected for heteroscedasticity and are clustered at the 3-digit NAICS industry level (t-statistics are in

AN

parentheses). *, **, and *** denote significance at the 10%, 5%, and 1% levels, respectively.

Panel A: Accounting performance by acquirer firms in the year after tariff cuts ROA in year t+1

Decdummy3

(3)

(4)

(5)

(6)

-0.010

-0.010

-0.006

0.004

0.004

0.011

(-1.40)

(-1.52)

(-1.10)

(0.27)

(0.26)

(0.67)

-0.024

(-0.47)

(-1.46)

0.017

0.042***

(1.01)

(2.99)

CE

PT

Decdummy3 x Acquisition Same industry acquisition

-0.005

ED

Acquisition

Decdummy3 x same industry acquisition

-0.007

-0.026

(-0.56)

(-1.44)

0.027**

0.057***

(2.56)

(3.45)

AC

Conglomerate acquisition

Decdummy3 x conglomerate acquisition R&D Cash Size Market-to-Book Assets

Net profit margin in year t+1

(2)

M

(1)

-0.001

-0.015

(-0.13)

(-1.27)

-0.008

-0.002

(-0.23)

(-0.06)

-0.122**

-0.122**

-0.122**

0.056

0.058

0.058

(-2.50)

(-2.54)

(-2.50)

(0.50)

(0.52)

(0.52)

-0.070***

-0.070***

-0.070***

0.034

0.033

0.033

(-7.20)

(-7.22)

(-7.32)

(0.58)

(0.54)

(0.53)

-0.023***

-0.023***

-0.023***

-0.046***

-0.046***

-0.046***

(-3.08)

(-3.04)

(-3.08)

(-8.06)

(-7.78)

(-8.12)

-0.003

-0.003

-0.003

-0.013

-0.013

-0.013

(-0.87)

(-0.88)

(-0.87)

(-1.14)

(-1.16)

(-1.15)

ACCEPTED MANUSCRIPT

Industry leverage Industry Cash # firms Industry ROA

0.007*

-0.003

-0.003

-0.003

(2.03)

(2.02)

(-0.36)

(-0.33)

(-0.30)

-0.267***

-0.267***

-0.267***

-0.185

-0.182

-0.181

(-5.87)

(-5.81)

(-5.69)

(-1.16)

(-1.15)

(-1.13)

-0.068***

-0.068***

-0.068***

-0.061**

-0.060**

-0.060**

(-9.19)

(-9.15)

(-9.07)

(-2.47)

(-2.46)

(-2.46)

0.024

0.023

0.024

0.030

0.030

0.031

(0.44)

(0.44)

(0.45)

(0.37)

(0.36)

(0.38)

-0.057*

-0.057*

-0.058*

-0.081

-0.080

-0.081

(-1.80)

(-1.83)

(-1.82)

(-1.23)

(-1.25)

(-1.27)

-0.000***

-0.000***

-0.000***

-0.001***

-0.001***

-0.001***

(-6.33)

(-6.22)

(-6.03)

(-4.87)

(-4.95)

(-4.99)

-0.051

-0.051

-0.051

0.043

0.043

0.043

(-1.65)

(0.84)

(0.84)

(0.83)

Yes

Yes

Yes

Yes

T

Net Loss

0.007*

(2.06)

IP

Return volatility

0.007*

CR

Debt/Equity

(-1.61)

Yes

Yes

Year fixed effects

Yes

Yes

Yes

Yes

Yes

Yes

Adj. R

0.702

0.702

0.701

0.212

0.212

0.212

N

11890

11890

11890

11890

11890

11890

AN

M

2

US

(-1.60) Firm fixed effects

ED

Panel B: Industry adjusted accounting performance by acquiring firms in the year after tariff cuts Industry Adjusted Net profit margin Industry adjusted ROA in year t+1 in year t+1

Acquisition

CE

Decdummy3 x Acquisition

PT

Decdummy3

(1)

(2)

(3)

(4)

(5)

(6)

0.005

0.006*

0.009***

-0.001

-0.002

0.003

(1.07)

(1.75)

(3.46)

(-0.20)

(-0.27)

(0.64)

-0.003

-0.014

(-0.28)

(-0.67)

0.024

0.028

(1.01)

(1.20)

AC

Same industry acquisition

Decdummy3 x same industry acquisition

-0.006

-0.017

(-0.47)

(-0.74)

0.031**

0.042*

(2.10)

(1.93)

Conglomerate acquisition Decdummy3 x conglomerate acquisition R&D Cash

0.002

-0.005

(0.37)

(-0.36)

0.004

-0.009

(0.10)

(-0.22)

-0.093**

-0.094**

-0.094**

0.073

0.074

0.075

(-2.52)

(-2.54)

(-2.50)

(0.70)

(0.71)

(0.70)

-0.071***

-0.071***

-0.071***

0.036

0.035

0.035

(-5.84)

(-5.95)

(-6.09)

(0.57)

(0.55)

(0.55)

Net Loss Industry leverage Industry Cash # firms Industry ROA

-0.030***

-0.030***

-0.030***

(-2.67)

(-4.68)

(-4.61)

(-4.62)

0.000

0.000

0.000

-0.008

-0.008

-0.008

(0.09)

(0.10)

(0.09)

(-0.86)

(-0.87)

(-0.87)

0.005**

0.005**

0.005**

-0.006

-0.006

-0.006

(2.22)

(2.18)

(2.14)

(-0.90)

(-0.87)

(-0.83)

0.030

0.029

0.029

0.305***

0.306***

0.307***

(0.39)

(0.38)

(0.39)

(2.81)

(2.84)

(2.84)

-0.052***

-0.052***

-0.052***

-0.035

-0.035

-0.035

(-7.42)

(-7.40)

(-7.40)

(-1.29)

(-1.30)

(-1.31)

-0.026

-0.027

-0.026

-0.034

-0.035

-0.034

(-0.46)

(-0.46)

(-0.45)

(-0.46)

(-0.47)

(-0.46)

-0.023

-0.023

-0.023

-0.095**

-0.095**

-0.096**

(-0.77)

(-0.78)

(-0.77)

(-2.46)

(-2.47)

(-2.47)

0.000*

0.000*

0.000*

0.001*

0.001*

0.001*

(1.77)

(1.75)

(1.74)

(1.98)

(1.96)

(1.96)

-0.022

-0.022

-0.021

0.067

0.067

0.067

(-0.41)

(-0.41)

(-0.38)

(0.72)

(0.72)

(0.71)

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

Yes

0.632

0.631

0.164

0.164

0.164

11890

11890

11890

11890

11890

Firm fixed effects

Yes

Year fixed effects

Yes

2

0.632

N

11890

AC

CE

PT

ED

M

Adj. R

IP

Return volatility

-0.013**

(-2.64)

CR

Debt/Equity

-0.013**

(-2.72)

US

Market-to-Book Assets

-0.013**

AN

Size

T

ACCEPTED MANUSCRIPT

ACCEPTED MANUSCRIPT Appendix Figure A1: Acquisitions in the year before treatment

AC

CE

PT

ED

M

AN

US

CR

IP

T

This figure shows the average number of acquisitions by treatment and control firms in the year before the treatment firms face tariff cuts. The sample period is 1998-2014. Treatment event is defined as a percentage decrease in ad valorem import duty rate by 30% or more relative to the previous year (defined as Decdummy3 in the paper). Firms that do not face a tariff cut of 30% or more are classified as control firms that year. Since firms can face tariff cuts of 30% or more in multiple years over the sample period, the composition of treatment and control groups can be different each year.

ACCEPTED MANUSCRIPT Figure A2: Acquisitions in the year of the treatment event

AC

CE

PT

ED

M

AN

US

CR

IP

T

This figure shows the average number of acquisitions by treatment and control firms in the year the treatment firms face tariff cuts. The sample period is 1998-2014. Treatment event is defined as a percentage decrease in ad valorem import duty rate by 30% or more relative to the previous year (defined as Decdummy3 in the paper). Firms that do not face a tariff cut of 30% or more are classified as control firms that year. Since firms can face tariff cuts of 30% or more in multiple years over the sample period, the composition of treatment and control groups can be different each year.

ACCEPTED MANUSCRIPT Figure A3: Acquisitions in the year after treatment

AC

CE

PT

ED

M

AN

US

CR

IP

T

This figure shows the average number of acquisitions by treatment and control firms in the year after the treatment firms face tariff cuts. The sample period is 1998-2014. Treatment event is defined as a percentage decrease in ad valorem import duty rate by 30% or more relative to the previous year (defined as Decdummy3 in the paper). Firms that do not face a tariff cut of 30% or more are classified as control firms that year. Since firms can face tariff cuts of 30% or more in multiple years over the sample period, the composition of treatment and control groups can be different each year.

ACCEPTED MANUSCRIPT Table A1 Robustness of main results to excluding small deals This table presents results from additional robustness tests of the main results on subsamples after excluding deals with deal values less than $10 million and $50 million, respectively. Panel A provides results from tests in Table IV after excluding acquisitions with deal values less than $10 million. Panel B looks at the main results after excluding acquisitions with deal values less than $50 million. The dependent variable, Acq, takes a value of one for firms that make an acquisition in year t. Decdummy1, Decdummy2, and Decdummy3 take a value of one for industry-years that face decreases in ad-valorem import duty rates by 2%, 20%, and 30% or more from the previous year, respectively. Control variables are defined in Table 4 and are measured in year t-1. Standard errors are corrected for heteroskedasticity and are clustered at the 3-digit NAICS industry level (t-statistics are in parentheses). Models 3, 6, and 9 include firm and year fixed effects. Number of observations varies across models due to perfect predictability of acquisition likelihood for some firms. *, **, an d *** denote significance at the 10%, 5%, and 1% levels, respectively.

T P

I R (7)

(8)

(9)

0.047***

0.027**

0.016

(5.15)

(2.45)

(1.47)

-0.125***

-0.033

-0.125***

Panel A: Main results after excluding deals less than $10 million in size Decdummy1

(1)

(2)

(3)

0.015**

0.020**

0.014*

(2.45)

(2.58)

(1.99)

Decdummy2

(5)

0.029***

0.022*

0.010*

A

(1.94)

(3.81) Decdummy3 R&D

-0.031 (-0.58)

Cash

(-3.59)

ROA Market-to-Book Assets Debt/Equity Return volatility Net Loss

(6)

U N

(2.02)

M

-0.031 (-0.57)

(-3.52)

(-0.58)

(-3.54)

0.123***

0.015

0.123***

0.015

0.123***

(3.76)

(1.11)

(3.78)

(1.11)

(3.79)

-0.001

0.031***

-0.001

0.030***

-0.001

(6.23)

T P (-0.19)

(6.12)

(-0.17)

(6.12)

(-0.15)

0.031

0.034**

0.032

0.034**

0.031

0.034**

(1.56)

(2.29)

(1.58)

(2.29)

(1.55)

(2.29)

0.011**

0.008***

0.011***

0.008***

0.011***

0.008***

0.015 (1.11)

Size

D E

-0.125***

C S

(4)

0.031***

E C

C A

(2.79)

(3.78)

(2.82)

(3.74)

(2.82)

(3.76)

-0.012***

-0.019***

-0.012***

-0.019***

-0.012***

-0.019***

(-4.47)

(-4.78)

(-4.31)

(-4.74)

(-4.36)

(-4.72)

0.078

-0.253**

0.075

-0.257**

0.073

-0.258**

(0.40)

(-2.23)

(0.39)

(-2.25)

(0.37)

(-2.23)

-0.018***

-0.029***

-0.018***

-0.030***

-0.018***

-0.030***

ACCEPTED MANUSCRIPT

Industry leverage Industry Cash # firms Industry ROA Constant

(-6.26)

(-4.36)

(-6.11)

(-4.53)

(-6.02)

(-4.58)

-0.137**

-0.057

-0.134*

-0.054

-0.131*

-0.054

(-2.08)

(-0.91)

(-2.03)

(-0.86)

(-2.00)

(-0.86)

-0.011

-0.060

-0.015

-0.061

-0.020

-0.064

(-0.20)

(-0.73)

(-0.26)

(-0.75)

0.001***

0.000

0.001***

0.000

(12.61)

(1.20)

(12.29)

(1.11)

0.075**

0.051

0.077**

0.052

(2.44)

(1.21)

(2.41)

(1.22)

0.099***

-0.081*

0.099***

-0.077*

(36.01)

(-2.06)

(38.20)

(-1.92)

Firm fixed effects

No

No

Yes

No

No

Year fixed effects

No

No

Yes

No

No

2

Adj. R N

0

0.065

0.223

0.001

15444

15444

14914

15444

D E

T P

C A

E C

(-0.79) 0.000

I R

(10.63)

(1.05)

0.072**

0.048

(2.53)

(1.10)

0.098***

-0.076*

(38.80)

(-1.88)

Yes

No

No

Yes

Yes

No

No

Yes

T P

C S

U N

A

M

(-0.37)

0.001***

0.064

0.223

0.002

0.064

0.223

15444

14914

15444

15444

14914

ACCEPTED MANUSCRIPT Panel B: Main results after excluding deals less than $50 million in size Decdummy1

(1)

(2)

(3)

0.004

0.011**

0.005

(0.87)

(2.40)

(1.07)

Decdummy2

(4)

(5)

(6)

0.012*

0.012

0.000

(1.80)

(1.50)

(0.01)

(7)

Cash Size

Debt/Equity

-0.056*

0.005

-0.056

(0.09)

(-1.72)

(0.09)

(-1.71)

# firms Industry ROA

0.003

(1.66)

(0.36)

0.004

-0.056*

(0.08)

(-1.71)

-0.014

0.051*

-0.014

0.051*

(1.95)

(-0.91)

(1.97)

(-0.91)

(1.97)

0.031***

0.004

0.031***

0.004

0.031***

0.004

(6.95)

(0.59)

(6.93)

(0.59)

(6.91)

(0.60)

U N

A

0.014

0.022*

(0.98)

(1.97)

0.012***

0.007***

(2.83)

(2.82)

M

0.014

0.022*

0.014

0.022*

(0.99)

(1.98)

(0.97)

(1.97)

0.012***

0.007**

0.012***

0.007***

(2.85)

(2.81)

(2.85)

(2.81)

-0.012***

-0.010***

-0.012***

-0.010***

-0.012***

(-4.06)

(-7.06)

(-4.06)

(-7.12)

(-4.03)

-0.442***

0.095

-0.444***

0.093

-0.444***

(-4.01)

(0.56)

(-4.01)

(0.54)

(-3.97)

-0.011***

-0.019***

-0.011***

-0.019***

-0.011***

-0.019***

(-4.07)

(-3.26)

(-4.08)

(-3.35)

(-4.04)

(-3.38)

-0.093*

-0.042

-0.091*

-0.041

-0.090*

-0.041

(-1.96)

(-1.04)

(-1.92)

(-0.99)

(-1.87)

(-0.99)

E C 0.096

C A

D E

T P

-0.010***

(0.56)

Industry Cash

0.013

(2.71)

0.051*

Return volatility

Industry leverage

0.020***

-0.014

(-7.21)

Net Loss

(9)

(-0.92)

ROA Market-to-Book Assets

C S

0.005

T P

I R

Decdummy3 R&D

(8)

-0.015

-0.035

-0.017

-0.035

-0.019

-0.036

(-0.30)

(-0.64)

(-0.34)

(-0.64)

(-0.41)

(-0.65)

0.000***

0.000**

0.000***

0.000**

0.000***

0.000**

(6.59)

(2.16)

(7.26)

(2.16)

(6.48)

(2.08)

0.047*

0.077*

0.048*

0.078**

0.046*

0.077*

ACCEPTED MANUSCRIPT (1.79) Constant Firm fixed effects Year fixed effects

(2.06)

(1.75)

(2.11)

(1.76)

0.063***

-0.123***

0.063***

-0.121***

0.063***

-0.120***

(27.88)

(-3.41)

(29.45)

(-3.34)

(29.88)

(-3.29)

No

No

No

No

No

No

Yes

Yes

No

No

Yes

No

No

Yes

No

Adj. R2

-0.000

0.085

0.222

0.000

0.085

0.222

0.000

N

14818

14818

14285

14818

14818

14285

I R

C S

A

U N

D E

T P

C A

E C

M

14818

(1.99)

Yes

No

Yes

0.085

0.222

14818

14285

T P

ACCEPTED MANUSCRIPT Table A2 Likelihood of acquisition by firms facing tariff increases This table reports results from tests that examine the association between decreased foreign competition and firm acquisition propensities. Incdummy1, Incdummy2, and Incdummy3 take a value of one for industry-years that face increases in ad-valorem import duty rates by 2%, 20%, and 30% or more, respectively. Control variables are defined in Table 4 and are measured in year t -1. Standard errors are corrected for heteroskedasticity and are clustered at the 3digit NAICS industry level (t-statistics are in parentheses). Models 3, 6, and 9 include firm and year fixed effects. Number of observations varies across models due to perfect predictability of acquisition likelihood for some firms. *, **, and *** denote significance at the 10%, 5%, an d 1% levels, respectively. Incdummy1

(1)

(2)

(3)

0.007

0.016*

0.004

(0.94)

(1.87)

(0.51)

Incdummy2

(4)

(5)

(6)

0.025***

0.029**

0.001

(2.72)

(2.58)

(0.18)

R&D

-0.031

-0.180***

(-0.94)

(-4.14)

Cash

0.017

0.153***

(1.54)

(4.21)

Size

0.023***

-0.008

(5.97)

(-0.86)

Industry leverage

-0.005

(1.62)

(-0.57)

-0.180***

-0.032

-0.180***

(-0.97)

(-4.16)

(-0.99)

(-4.16)

0.018

0.152***

0.017

0.152***

(1.55)

(4.21)

(1.53)

(4.21)

0.023***

-0.008

0.023***

-0.008

(5.99)

(-0.85)

(5.97)

(-0.85)

0.036**

0.035

0.036**

0.034

0.036**

D E

(1.49)

(2.22)

(1.49)

(2.21)

0.008*

0.007***

0.008*

0.007***

0.008*

0.007***

(1.93)

(3.26)

(1.91)

(3.22)

(1.92)

(3.23)

-0.013***

-0.020***

-0.013***

-0.020***

-0.013***

-0.020***

(-4.95)

(-4.72)

(-4.92)

(-4.71)

(-4.94)

(-4.71)

-0.009

-0.370**

-0.005

-0.368**

-0.007

-0.365**

(-0.04)

(-2.59)

(-0.02)

(-2.58)

(-0.04)

(-2.57)

-0.026***

-0.037***

-0.026***

-0.037***

-0.026***

-0.037***

(-7.02)

(-5.56)

(-6.83)

(-5.54)

(-6.92)

(-5.44)

-0.177**

-0.045

-0.179**

-0.045

-0.177**

-0.048

(1.49)

Net Loss

0.024

(2.22)

0.035

E C

Market-toBook Assets

Return volatility

(2.09)

(9)

T P

ROA

Debt/Equity

0.021**

(8)

-0.032

A

M

T P

I R

C S

U N

Incdummy3

(7)

C A

ACCEPTED MANUSCRIPT (-2.18)

(-0.62)

(-2.19)

(-0.63)

(-2.17)

(-0.67)

-0.037

-0.105

-0.036

-0.104

-0.040

-0.106

(-0.54)

(-1.09)

(-0.51)

(-1.10)

(-0.57)

(-1.12)

0.001***

0.000

0.001***

0.000

0.001***

0.000

(15.66)

(0.98)

(13.15)

(0.98)

0.087

0.063

0.090*

0.064

(1.69)

(1.32)

(1.73)

(1.33)

Industry Cash # firms Industry ROA Constant

0.121***

0.004

0.120***

0.004

(42.30)

(0.11)

(44.08)

(0.12)

No

No

Yes

No

No

No

No

Yes

No

Adj. R

-0.000

0.046

0.213

0.000

N

15799

15799

15282

15799

Firm fixed effects Year fixed effects 2

D E

T P

C A

E C

(1.00)

I R

0.088

0.065

(1.71)

(1.36)

0.121***

0.006

(44.66)

(0.17)

Yes

No

No

Yes

No

Yes

No

No

Yes

0.047

0.213

0.000

0.047

0.213

15799

15282

15799

15799

15282

C S

U N

A

M

T P

(13.44)

ACCEPTED MANUSCRIPT Highlights :

Manuscript: CORFIN_2018_350

T P



I find that firms are more likely to make acquisitions when faced with increased foreign competition.



I use tariff cuts as exogenous shocks to industry competition structure.



The results are stronger for the subsamples of firms that are affected more by increased foreign competition and firms that stand to gain

I R

C S

U N

more from such acquisitions.

A



Only financially unconstrained firms adopt the acquisition channel to respond to increased foreign competition.



When faced with increased foreign competition, firms make acquisitions to become more efficient. Such firms have better profitability

D E

M

ratios in the year following such acquisitions in response to increased foreign competition.

T P

C A

E C