1 Introduction

In the literature, the link between types of financing and firm strategic behavior is crucial. The relationship between a firm’s capital structure and strategy is acknowledged as key (Brown et al. 2009), and finance and strategic behavior questions are interrelated (Balakrishnan and Fox 1993). Debt is an important type of funds source (Corbett and Jenkinson 1997), capital structure is measured as the ratio of debt to total capital (Zingales 1998),Footnote 1 and the defining feature of two components in a firm’s capital structure, debt and equity, are the differential rights of providers (La Porta et al. 1998).

Financial structure matters strategically (Myers 1977; Parsons and Titman 2008), since due to agency, signaling and information asymmetry issues (Ross 1977; Myers and Majluf 1984) there is heterogeneity in firms’ financing decisions (Denis and Mihov 2003). The issues of whether to have debt, how much, and what happens are important (Grinblatt and Titman 2001). Relative debt levels impact firms’ behavior. The findings in the leverage and strategy literature indicate that increased debt in firms softens aggressive strategic behavior. Titman (1984) evaluates liquidation decisions; Lambrecht (2001) evaluates entry and exit decisions; Phillips (1995) evaluates output and pricing decisions; Chevalier (1995) evaluates strategic responses of firms, while Zingales (1998) highlights the links between capital structure, firm performance and survival; David et al. (2008) evaluate the impact of debt on long-term oriented R&D activities; Anderson and Makhija (1999) evaluate the relationship between debt and corporate growth opportunities in Japan, and Chirinko and Elston (2006) carry out similar analysis for Germany.

Simultaneously, a large literature discusses why firms merge. Reasons posited for mergers are to gain efficiencies (Farrell and Shapiro 1990), obtain reputation advantages (Dranove and Shanley 1995), reduce risk (Amihud and Lev 1981), diversify (Levy and Sarnat 1970), enjoy spillovers (Tremblay and Tremblay 1988), reconfigure resources (Steiner 1975), and redeploy capabilities (Capron and Mitchell 1998). Conversely, negative motives, such as hubris (Roll 1986), also drive mergers. Mergers determine the extent of competition in a market, and create substantial variations in industry structure (Steiner 1975) via the consolidations of firms. The historical evidence shows that concentration-enhancing structural variations permit firms to gain market power (Stigler 1964), to permit a quiet life (Hicks 1935), to grow the firm (Mueller 1969), and to permit rent-extraction (Marris 1964).

In the finance literature, given a takeover or merger decision, a corpus of work (Bessler et al. 2011) evaluates the financial structure of such a deal. Merger transactions account for a large proportion of global gross domestic product. In the United States between 15,000 and 20,000 mergers occurred in the 2000s, and the role of debt in influencing economic activities is profound. There is a knowledge gap, as no evidence exists on how firm’s leverage influence takeover or merger decisions. A speculation is that, because of a limited liability effect, taken-over or merging firms with high leverage will compete aggressively (Socorro 2007). There is exploration of the relationship between leverage and merger timing (Morellec and Zhdanov 2005), but the evidence on how leverage affects firms’ merger decisions is non-existent.

1.1 Scope of the Study

The study evaluates the relationship between firms’ leverage and mergers. The mergers assessed are all of those occurring among the population of local exchange firms in the United States telecommunications industry in the last two decades. Based on mergers data, the sector provides a natural experiment setting for retrospectively evaluating how leverage had impacted firms’ merger propensities.

In infrastructure-based sectors, changes in firms’ strategic behavior have raised concerns that high debt levels could be leading to severe firms’ financial distress (Bortolotti et al. 2011). For telecommunications firms, the funding of assets such as spectrum and new networks would cost very large sums of money. Conversely, limits on abilities to raise prices, because of intense competition or effective regulation, would compromise the interest-paying capacities of such firms if they were to be highly indebted. Thus, strategies likely to compromise debt servicing would be considered risky, and risky strategic behavior of relatively greater-leveraged telecommunications firms would be constrained.

Based on recent capital structure research (Majumdar 2016), we evaluate the relationship between firms’ leverage and mergers of local exchange firms in the United States telecommunications industry. Past mergers that have been approved are natural experiments, so that pre- and post-merger assessments can be carried out, permitting an evaluation of competition policy decisions. If, after evaluation of an allowed merger, subsequent outcomes, measured via analysis of firms’ behavior, are below expectations, then results indicate that competition agencies may not have got economic assumptions right (Carlton 2009).

The use of historical panel data, from 1988 to 2001, of almost all the local exchange carriers in the United States, containing pre and post-merger data for firms, for two uniquely-different institutional periods, enables tracking of dynamic merger history for each carrier (Majumdar et al. 2012). Table 1 (extracted from Majumdar et al. 2014), lists ownership status and merger activity of local exchange telecommunications companies before 1996. Table 2 lists post-1996 merger events.

Table 1 Status of firms and merger activity in the local exchange sector of the US telecommunications industry before the passage of the telecommunications act of 1996
Full size table
Table 2 Status of firms and merger activity in the local exchange sector of the US telecommunications industry after the passage of the telecommunications act of 1996
Full size table

Two issues can be investigated by retrospective merger assessments. First, whether specific merger decisions were right or wrong can be assessed. Second, the detailed evaluation of a stream of mergers permits an understanding of the outcomes of competition and merger policies. Retrospective assessments of consummated mergers have been reviewed by Hunter et al. (2008), Kwoka Jr (2013) and Pautler (2003). These historical assessments generate performance metrics useful for competition policy development (Kovacic 2006). Retrospective merger appraisals permit the accumulation of knowledge about various behavioral, structural and competitive conditions in different industries. By engaging in retrospective analyses, it is possible to highlight conditions under which merger outcomes are positive or negative. If these conditions can be established, then conclusions can be generalized to develop heuristics that provide policy insights in evaluating mergers based on expected outcomes in terms of price, and impact on other economic parameters.

2 Theory and Institutional Facts

2.1 Debt and Mergers

There are no clear propositions about the impact of leverage on mergers. One literature, on leverage and aggressiveness in strategic behavior, is based on the long-purse argument (Telser 1966), suggesting that having ready access to capital allows a firm to sustain losses until it succeeds in eliminating competition. A firm with low leverage can raise more debt. This will enable it to behave aggressively. A firm with higher leverage can be vulnerable to competitors’ aggressive behavior. The presence of high debt levels provides opportunities for rivals to weaken firms financially by aggressive strategies. Highly-leveraged firms cannot respond similarly, since financial outcomes can be negative. Hence, mergers can be risky actions.

Firms with high debt levels are monitored. They may require periodic re-financing, and negative financial parameters can lead to re-financing denial. Hence, firms with high leverage will be passive to not engender negative financial outcomes (Poitevin 1989), and high leverage levels is associated with a less aggressive strategic stance (Showalter 1995). Concomitantly, firms with lower leverage can signal potential to compete aggressively because of their ability to later tap into additional resources, and be less prone to creditors’ pressures (Poitevin 1989).

Converse arguments are proposed. Lower leverage can be associated with less aggressive behavior (Brander and Lewis 1986). Managers in firms with higher leverage will promote aggressive strategies, because equity holders with limited liability will lose to the extent of their capital. They can afford to sustain uncertain outcomes from risky strategies. Mergers can be risky strategies, in spite of the benefits espoused, given the overwhelming evidence, in the literature, that mergers do not work (Andrade et al. 2001; Datta et al. 1992).

Managers will be aware of merger vicissitudes. Thus, firms may engage in risky mergers, while espousing positive efficiency-enhancing, resource-pooling and technology-spreading gains, anticipating low probability of outcomes. In such cases, higher leverage permits firms to engage in a risky strategy knowing that likely negative outcomes will be borne to a limited extent by firms’ owners as negative financial consequences are passed on to debt holders.

2.2 Institutional Change and Mergers

Because firms’ decisions are driven by context changes (Nelson and Winter 1982), environmental factors play a role in influencing firms. Industry shocks, such as changes in technology or in the regulatory environment, can lead to sequential mergers (Fauli-Oller 2000).Footnote 2 Triggers for resource re-configuration are contingent on environment changes (Ambrosini et al. 2009). Thus, engaging in mergers is capability enhancement by firms (Teece et al. 1997), as mergers lead to reconfiguration of resources (Capron and Mitchell 1998). Mergers allow firms to restructure business activities (Steiner 1975). As part of a process in which firms with excess resources buy less well-endowed firms, mergers permit usage of combined resources to improve merged-firm capabilities (Pleatsikas and Teece 2001). Reconfigurations economies include those from rationalization, stemming from disposal of duplicate assets, and restructuring arising from the recombination of capabilities (Brodley 1987). These activities release resources for other activities.

2.3 Telecommunications Mergers and Hypothesis

The flow of merger events is described next, and summarized in Table 1. In 1984, with the break-up of the old AT&T, the sector consisted of 7 regional holding companies (RHCs), which owned 22 stand-alone Bell operating companies (BOCs). There were 5 other groupings: Central Telephone, Continental Telephone, GTE, Southern New England Telephone and United Telephone, and two large independent companies: Cincinnati Bell and Rochester Telephone. These groupings owned several ILECs. Merger deals were takeover of ILECs by RHCs or other groupings. See Tables 1 and 2 for details. After a merger deal, ILECs’ ownership was transferred from one holding company to another.

A first set of mergers occurred between the 1984 divestiture and the introduction of legislation in 1996. We have classified these as pre-1996 mergers. In the early 1990s, several RHCs amalgamated separate stand-alone ILECs. In 1991, the operations of Mountain States Telephone and Telegraph Company, Northwestern Bell Telephone Company and Pacific Northwest Bell Telephone Company were combined to form US West Communications. In 1992, the amalgamation of South Central Bell Telephone Company and Southern Bell Telephone and Telegraph Company operations, as Bell South, took place. These were successful units that were consolidated for the purposes of exploiting economies of scope.

Simultaneously, several non-RHC groupings, with less-than adequate financial resources, or that were territorially disparate, were acquired by other groupings. Thus, Continental merged its companies, such as Contel of California, Contel of New York, Contel of Virginia and Contel of Texas with GTE. These events occurred in 1990. Thereafter, the operating companies belonging to United were acquired by Sprint in 1991 which became a holding company. Later, the operating companies of Central Telephone Company were acquired by Sprint in 1992.

A second series of mergers occurred after the 1996 legislation. These extremely-large mergers were significant attention-receiving events in the industry (Hazlett 2000), intended to exploit opened-market possibilities (Goldman et al. 2003). We have classified these as post-1996 mergers. The key mergers, by deal size, were the merger of SBC and Pacific Telesis, in April 1996, between two RHCs; the merger of Bell Atlantic and NYNEX, in April 1996, between two RHCs; the merger of WorldCom and MCI, in October 1997, between two IXCs; the merger of AT&T and TCI, in May 1998, between an IXC and a cable operator; and the merger of SBC and Ameritech, in June 1998, between two RHCs.

The 1996 legislation, having opened the local exchange market to entry by CLECs, motivated performance- and growth-enhancing mergers among unattached RHCs. For example, Pacific Telesis, under financial strain in California, was merged with Southwestern Bell Corporation (SBC). To expand it national scope of operations, SBC acquired Southern New England Telephone (SNET) in 1998. In addition, inter-modal competitive threats from the cable sector emerged.Footnote 3

Several RHCs acquired other groupings. Ameritech was acquired by SBC, to obtain financial scale for a major presence in all United States local markets. In 2000, GTE was acquired by Bell Atlantic. Bell Atlantic had also earlier acquired NYNEX, an RHC in its own right. The conglomerate Bell Atlantic re-named itself Verizon. The motive was to bring together complementary assets and strengths, create scale and scope economies, permit innovations and accelerate delivery of advanced services. Additionally, a motive was to tackle competition from cable giants such as AT&T.

Several ILECs went through two mergers. The Continental ILECs went through two mergers. First, they were acquired by GTE. Then, GTE became part of Bell Atlantic. Similarly, the GTE and NYNEX local exchange companies went through two mergers. The first was when they were acquired by Bell Atlantic. The second was when Bell Atlantic acquired Puerto Rico Telephone Company and then re-named the whole group, consisting of erstwhile Bell Atlantic, Contel, GTE, NYNEX and Puerto Rico Telephone Company, as Verizon. Pacific Telesis was first absorbed into SBC. Then, Ameritech was absorbed into SBC and the entire SBC structure recast.

Given institutional conditions, we state our expectations. The institutional environments had been different before and after 1996 in the telecommunications industry. Before 1996, the local exchange sector consisted of regulated monopolies, where each firm had a pre-demarcated market with no competition possible due to entry barriers. We have noted that the impact of the 1996 legislation had been vital in inducing entry by IXCs, CLECs and CSOs in incumbents’ markets. The pre- and post-legislation historical environments had been different, affecting firms’ strategies (Cave et al. 2002).

Institutional entry barriers were eliminated after 1996. Entry in firms’ territories became extensive. In such circumstances, after due scanning and searching, firms might engage in mergers to enhance resources for gaining competitive strength and seize new opportunities. Yet, in spite of capability-augmenting possibilities, mergers could be a risky strategy because of de-novo introduction of competition in historically-regulated environments. Nevertheless, firms might engage in mergers to ward off later competitiveness problems or enjoy new market opportunities. Though efficiency-enhancing, resource-pooling and technology-spreading gains could be theoretically feasible, firms could anticipate a lower probability of these outcomes in a competitive environment, relative to a regulated milieu experienced earlier. In such cases, in anticipations of negative outcomes, higher leverage could constrain firms from undergoing an ownership change.

A greater likelihood of negative financial outcomes could limit funds for meeting interest costs and capital repayments. Thus, value of debt-holders’ security might be compromised. A flow of funds stoppage possibility, in an infrastructure sector, would have many implications. Given debt as an important source of funds, if merger strategies could create negative financial outcomes, with interest payments delayed and debts rolled over, these contingencies could preclude further lending.

3 Analysis

3.1 Data

We assess the impact of debt on mergers using a panel data set. Data are available for each firm for the full period during which it was under prior and, after takeover, new ownership. The data have been used many times for sector analyses (Majumdar et al. 2012). A balanced panel from the Statistics of Communications Common Carriers (SCCC) are used for the period 1988 to 2001. Data are compiled for the principal local operating companies. These companies account for 99 % of the telephone lines in the United States.

3.2 Merger Variables

A merger dummy variable denotes the occurrence of a takeover or ownership-change event. These merger events correlate with a list of mergers maintained at www.cybertelecom.org. The design of dummy variables to control for merger impact is based on prior research (Gugler and Yurtoglu 2004; Majumdar et al. 2012). First, the merger variable (First Merger) is constructed for all first takeover or merger events. Second, for a few firms a Second Merger variables is also constructed, as some of the firms engaged in a sequence of mergers. The second mergers occurred 1996, when the Telecommunications Act was passed.

The way the firms keep financial records, based on accounting and regulatory requirements, each firm retains its accounting identity. Even if a firm has been taken over, by merger, this accounting identity remains and the data reported for the period 1988 to 2001 are based on these accounts-reporting identities. For every firm after merger, its behavior relative to itself in the past, when it had not merged, or relative to other firms in the same period can be evaluated (Majumdar et al. 2012). This facility also applies to identification of separate reasons behind each takeover.

Table 3 lists variables used and their definitions. The data property permits a pre- and post-merger evaluation of outcomes for the same set of firms over a long time-period, a type of analysis in demand (Carlton 2009). A merger dummy variable for merger and acquisition events denotes an event occurrence. The impact of leverage on merger events are evaluated.

Table 3 List of explanatory variables
Full size table

4 Estimation

4.1 Treatment Effects Modeling

A merger occurrence is endogenous, but taking-on debt is also endogenous, as a firm’s debt level is not exogenously given (Parsons and Titman 2008). The literature (Mackay and Phillips 2005; Titman and Wessels 1988) has extensively dealt with endogeneity of debt, and why firms borrow. Micro-econometric causality analysis helps ascertain factors influencing higher debt, given that relative leverage can influence mergers.

In causality analysis of debt and mergers, the debt endogeneity concern (Parsons and Titman 2008) is tackled using the treatment effects approach, in which a dummy explanatory variable denotes the existence of an endogenous phenomenon the impact of which is evaluated on an outcome variable (Heckman 2005).

Firms with greater than average borrowing, measured as above median borrowings, subject themselves to a treatment. The treatment occurrence is given by the transition from below-median to above-median borrowing. Its outcome is evaluated in terms of resulting merger activity. A transition from below-median to above-median borrowing will have been a treatment experienced. Such a transition involves changes in behavior influencing merger activity.

The choice of whether to engage in a treatment is endogenous involving a selection bias, since not all firms will have engaged in such a transition but a selected set of firms. The treatment effects approach is applied in assessing how having greater leverage influences firms’ subsequent mergers. A treatment effects model considers the median borrowing variable as a covariate influencing mergers, after median borrowing has been modeled as a dummy endogenous variable influenced by other exogenous variables. Selection bias arises because treated firms differ from non-treated firms for reasons other than the treatment status, per-se. The process of incurring above-median borrowings is conditioned by several factors, as self-selection into treatment is at play when financial programs, like raising debt, are decided on by firms.

Treatment effect models (Hirano et al. 2003; Rubin 1974) permit natural experiments to be assessed, where a response function, identifying strategic behavior, in this case of firms merging, embodies the effect of interest after the onset of an internal decision or institutional policy (White 2006). Natural experiments are identifiable discrete shifts in within-firm or outside-firm environments such that there is a significant change in behavior (Angrist and Krueger 2001).Footnote 4

4.2 Variables in the Outcome Equation

We use the First Merger and Second Merger as two separate dependent variables in outcome equations. In a treatment effects model, the treatment variable is a 1, 0 variable. We create the explanatory variable in the primary outcome equations as a measure of each local exchange company’s leverage (Leverage), which is coded as a binary indicator with the value of 1 which denotes if a firm has above median debt, relative to other firms, and it is coded as 0 denoting if a firm has below median debt, relative to other firms, in that time period.

Other control variables are included in the outcome equations. Mergers are driven by performance enhancement motives, and contemporaneous and performance indicators are used as merger-influencing variables; contemporaneous (Efficiency) and lagged (Efficiencyt-1) ratios of total operating expenses to total plant in service measures efficiency, and contemporaneous (Financial Performance) as well as lagged (Financial Performancet-1) ratios of total operating revenues to total assets measure financial performance (Cornett and Tehranian 1992).

Since mergers are capability-acquisition activities, relative technology deployment levels will affect mergers. The quantum of fiber cabling is used as an indicator of technological capabilities, since fiber is the key broadband resource in a communications environment; thus, contemporaneous (Broadband) and lagged (Broadbandt-1) ratios of fiber optic cabling to total lines are used as explanatory variables.

Environmental factors likely to affect mergers are business lines and urban population ratios (Sharkey 2002). The business lines construct (Business) is the ratio of business lines to total access lines for each company. The urban population ratio (Urban) is the weighted average ratio of urban population to total population in each firm’s territory. This ratio is weighted by the fraction of lines each firm has operating rights to in a specific territory. The use of market share constructs proxies for market presence, though in regulated industries a high market share does not necessarily imply monopoly power (Spulber 2002). The market share variable (Market Share) is constructed by taking the ratio of a firm’s lines in its operating territory relative to the total lines in the territory.

To control for industry-related factors, we use an: intensity of competition (Competition) variable. The competition variable is the number of possible competitors given a license to operate in the various states. This variable represents the intensity of market competition in each territory. The competition data are collected from the FCC Competition in Telecommunications Industry reports. For each incumbent local exchange carrier, the competition variable is computed as the sum of the number of competitive local exchange carriers operating in the incumbent’s territory.

4.3 Covariates Determining Leverage

This analysis is extensively based on Majumdar (2016). The extent of firms’ leverage indicates expectations about earnings capacities and abilities to repay debts. Yet, on what factors drive debt levels, there is no universal set of covariates determining leverage (Myers 2003). Factors relevant in explaining leverage variations are contingent on time and place specificities (Simerly and Li 2000), industry factors (Vincente-Lorente 2001), and firm-specific attributes (Kayhan and Titman 2007).

The relevant surveys (Frank and Goyal 2009; Harris and Raviv 1991; Rajan and Zingales 1995) highlight the covariates influencing leverage. Leverage increases with fixed assets, non-debt tax shields, growth opportunities, and firm size; it decreases with volatility, spending on intangible assets and superior economic performance (Harris and Raviv 1991). Others (Rajan and Zingales 1995) highlight firm size, profitability, possession of tangible assets and firm growth opportunities as variables significantly impacting leverage.

Frank and Goyal (2009) find six factors accounting for more than 27 % of leverage variations, with other factors contributing 2 %. Firms in industries where the median firm has high leverage have high leverage; firms with greater tangible assets have higher leverage; firms with higher profits have lower leverage; larger firms have higher leverage; and firms with higher market-to-book ratios have lower leverage. While many of these measures are not applicable to our study since many of these telecommunications companies were not publicly traded at the time, we attempt to include as many of these measures as our data sample permits.

4.4 Variables in the Selection Equation

Given the literature, the following are included as explanatory variables in the selection equation for Leverage: contemporaneous as well as lagged sales growth (Growth and Growtht-1), firm size (Size) measured as the log of total assets, financial performance (Performance) which is an asset utilization ratio, an important driver of profitability, measured as the ratio of operating revenues to total tangible assets, and the ratio of total long term assets to total assets (Assets).

4.5 Regulatory Factors and Leverage

Institutional contingencies are relevant for regulated industries, where output prices are regulated. Regulated entities, such as electric utilities and telecommunications firms, in the United States have high leverage (Barclay et al. 2003). Firms subject to rate of return regulation choose high leverage because interest costs are included in the rate base for calculating allowable returns (Dasgupta and Nanda 1993; Spiegel 1996; Spiegel and Spulber 1994).

Regulators have incentives to set high regulated prices to lower the probability of regulated firms becoming financially distressed, allowing interest costs to pass-through to customers and hence permitting higher leverage to occur (Spiegel and Spulber 1994). Price regulation in output markets provides regulated firms with incentives to use higher debt to finance operations, and the introduction of innovative regulations lowers leverage (Ovtchinnikov 2010).

Sector regulation changes have been important. The occurrence of regulatory changes is modeled in assessing the extent of leverage. A shift in risk, from a firm’s customers to its shareholders, can follow from changes in pricing regulations, from rate of return to incentive regulation schemes such as a price cap regulation (Laffont and Tirole 1993). These institutional changes alter the nature of agency costs regulators face, and trigger changes in firms’ incentives for reducing costs (Sappington 2002), as a result of which firms may reduce debt levels to reduce interest burdens. Hence, a shift away from a rate of return regime, to alternative schemes, can significantly influence debt levels in firms.

4.5.1 Regulation Variables in the Selection Equation

Based on the discussion, explanatory variables reflecting the nature of price regulation are included in the selection equation. There have been two types of regulation over prices in the telecommunications sector of the United States; the rate of return and price cap regulation schemes (Sappington 2002). Introduced in 1990, by 2001 over 40 states had implemented price cap regulations. In 1985 there were 50 states with rate of return schemes. By 2001, that number was six. Other incentive schemes were also implemented (Sappington 2002).

Data on telecommunications sector regulatory changes have been used in many other works surveyed in Sappington (2002). We construct variables for five different types of regulatory schemes: Rate of Return Schemes, Other Incentive Schemes, Earnings Sharing Schemes, Hybrid Price Caps, and Pure Price Caps. The base case regulatory scheme, left out in estimation and used for comparison purposes, is the Rate of Return Schemes variable.

4.5.2 Institutional Variables in the Selection Equation

Institutional factors influence firm financing (La Porta et al. 1998; Majumdar 2016). Thus, covariates are included in the selection equation to account for sector-specific institutional contingencies. In the sector, the Telecommunications Act of 1996 has required firms meet certain requirements, under Section 271 of the act, before they enter long distance markets (Economides 1999). Based on Section 271 approvals data (Brown and Zimmerman 2004), a dummy variable (Section 271) is added for the observations obtaining such approvals.

A competition intensity variable is the ratio of competitors in a firm’s territory relative to the average number of industry competitors (Competitive Intensity), based on the available data for competitive entrants given a license to operate in firms’ territories. An important related factor is inter-modal competition (Loomis and Swann 2005). A dummy variable (AT&T Cable), coded as 1 for the years of AT&T ownership of cable assets, and 0 otherwise, is included to account for inter-modal competition.

4.6 Financial Market Considerations and Leverage

The structure of interest rates predicts real economic activity (Fama 1986; McKinnon 1973). The issue is, does the level of interest rates influence borrowings, and do such borrowings influence capital investments? Since Friedman and Schwartz (1963) highlighted the importance of monetary policies in influencing real sector activities, short-term interest rates have been used to influence the cost of capital and spending (Bernanke and Gertler 1995). Monetary policy changes lead to balance sheet restructuring, including of leverage (Adrian and Shin 2010). Financial policies propagate shocks. These constraints affect firms’ leverage (Korajczyk and Levy 2003).

The evidence on policy-induced constraints, however, is mixed. Monetary policy actions may be followed by real output movements lasting two or more years (Bernanke and Blinder 1992); a higher real funds rate has been associated with lower growth in future real output, with the negative correlation interpreted as high interest rates implying low investment opportunities (Estrella and Hardouvelis 1991). Further analysis finds insignificant (Laurent 1988), or mixed evidence that interest rates impact real economic activity (Harvey 1988). Bernanke and Gertler (1995) state that the quantitative identification of interest rate and financial effects on spending remains difficult, and that non-financial factors may influence spending patterns. Analysis also shows that a small proportion of the real output variance in the United States in the last five decades has been attributable to monetary policy shifts (Leeper et al. 1996). Interest rates may not affect leverage, since expenditures are planned ahead and subject to non-monetary influences (Woodford 2003).

4.6.1 Financial Variables in the Selection Equation

In assessing macro-economic factors impacting firm level financial decisions, financial market controls are required. Two variables used are the interest rate on 30-year long term U. S. Treasury bonds (Interest Rate) and a variable measuring period-to-period changes in interest rates (Interest Rate Change).

5 Results

5.1 Findings and Interpretation

The results are in Tables 4 and 5, and relate to the treatment effects specification of the leverage and merger relationships. There are five columns of results for the selection and outcome equations in each table. In Table 4 and model (1), the First Merger variable is introduced on its own, with no controls added. In model (2), along with the First Merger variable all of the control variables are added.

Table 4 Treatments effects estimation results for the first merger outcome variable
Full size table
Table 5 Treatments effects estimation results for the second merger outcome variable
Full size table

Efficiency gains are a major reason for mergers (Farrell and Shapiro 1990), and in model (3) just the prior efficiency variables are included to assess whether their stand-alone effects lead the debt variable to stay the same in magnitude. The levels of past financial performance, reconfiguration of resources and redeployment of capabilities (Steiner 1975) are other reasons noted as merger motivators. In model (4) the efficiency and capability-related variables are included, while in model (5) the financial performance and capability-related variables are included. In Table 5, results for the Second Merger variable are detailed. The models are as for the First Merger variable. In Table 4 models (1) to (5), the debt variable, Leverage, has been negative and significant (p < 0.01). In Table 5 models (1) to (5), the Leverage variable is also negative and significant (p < 0.01). The results support the hypothesis advanced.

Because the dependent variables, First Merger and Second Merger, are 1, 0 dummy variables, as is the Leverage variable, the impact of Leverage on mergers can be interpreted as the probability of merger disapproval, in the case of a negative coefficient, and merger support in case of a positive coefficient. Taking the results of models (1) to (5), with respect to the First Merger variable, the presence of above median debt in firms, as measured by the Leverage variable, leads to an average 75% lower likelihood of a first merger occurrence. Based on the results of models (1) to (5), with respect to the Second Merger variable, the presence of greater debt leads to a 10% lower likelihood of a second merger occurrence. The results stay consistent across the specifications. The telecommunications sector evidence shows that debt negatively affects the merger predilections of the firms studied.

There are two opposite predictions about the impact of firm leverage on mergers. Our results are interpreted in the light of the institutional features of the sector. Given the postulates of the long purse argument (Telser 1966), requiring access to capital, to allow firms to bear negative outcomes until competitive success ensued, firms with low leverage might raise more debt to enable them to behave aggressively. Firms with higher leverage might be vulnerable to competitors’ aggression. High debt levels could engender weakening by other nimble rivals, and older-established firms could become unresponsive to events. Hence, mergers could be risky. Such firms would be monitored, given re-financing requirements, and firms with high leverage would be passive to avoid negative fiscal outcomes. Thus, high leverage levels would be associated with less aggressive behavior by firms, and correspondingly with lower merger propensity.

The first set of mergers by firms in the sector were carried out to consolidate resources. These early events were driven by performance-enhancing considerations. Many smaller firms such as Continental Telephone, Central Telephone and United Telephone consolidated their operations at that time. In the late 1980s and early 1990s, mergers were also carried out to gain scope economies, such that the combined entity could tackle larger players with deeper pockets.

Hence, a-priori weak telecommunications firms were merging to form stronger firms. In such circumstances, the presence of higher-than-average debt would enhance the riskiness of such firms. Thus, for first mergers the presence of high leverage led to higher propensity, as given by the size of the Leverage coefficient, to not merge, as compared to the impact of high leverage on second mergers which led to a lowering of the intensity of the negative propensity to merge.

Most of the second mergers were carried out to exploit opportunities after the passage of the market-opening legislation. The first notable telecommunications merger of this genre was between Pacific Telesis and Southwestern Bell. Many of latter-period telecommunications mergers were undertaken by bigger firms that then became Verizon and AT&T (the former SBC), to gain size and market presence. The latter-period mergers involved numerous companies which progressed through more than one merger event. Several sequential telecommunications mergers were undertaken to gain enhanced competitive leverage in evolving markets which would become contestable, given new entrants’ presence (Ferguson 2004).

Such scaling-up of telecommunications firms’ operations and activities would reduce the risk associated with higher leverage; nevertheless, risk would still exist in a highly-competitive business environment. Removal of restrictions, that would have been keeping other types of communications firms from entering local exchange markets, would bring in an assortment of IXCs, CAPs, CLECs and CSOs to offer services to incumbent companies’ customers (Woroch 2002). While size of incumbent companies might mitigate environmental uncertainties, risks might have remained large enough to ensure that the relationships between higher leverage and merger propensity would be remaining negative.

5.2 An Alternative Explanation

An alternative explanation, based on ideas expressed elsewhere (Majumdar 2016) from which this discussion is abridged, is suggested. Debt can be relational and transactional (Berger and Udell 1995), a classification similar to insider and outsider debt (Carey et al. 1998). Bank and financial institution loans are relational debts. Bonds and securities are transactional debts (Bhattacharya et al. 2004). Transaction lending is arms-length (Boot and Thakor 2000). A firm could have many transactional lenders. Because of free-riding, such lenders would not be able to engender collective action against firms’ managers to constrain managers’ risky strategies, such as engaging in mergers, and the lenders could face hold-up problems.

Conversely, relational debt-holder could influence firms’ strategies. Firms with a single lender would face pressures to engage in growth-enhancing opportunities (Houston and James 1996), as well as constraints hindering them from taking risky decisions. This would be borrower holdup by monopolistic bank creditors (Sharpe 1990), given that bank lenders would have been insiders (Ivashina et al. 2009). Such relational lenders could monitor borrowers to obtain information for interventions, and lending would involve receiving proprietary information (Allen 1990) used in multiple interactions (Greenbaum and Thakor 1995). If firms proposed to engage in possibly-risky strategies, such as large mergers, relationship-based lenders could exert pressures to constrain such behaviors.

In old-established industries facing oversight, such as telecommunications (Woroch 2002), lenders and firms would have engaged in business relationships and interacted for decades (Majumdar 2016). Over time there would have been repeated interactions between borrowers and lenders. Hence, if telecommunications firms were to take decisions involving large amounts on mergers, with the possibility of uncertain outcomes, the lenders would constrain firms’ activities so as to preserve financial performance outcomes and the security of sums lent by them. Thus, a negative relationship between high leverage and the occurrence of mergers would be likely to be observed.

6 Conclusion

Mergers are important phenomena with impacts on competition policy, strategy and economic performance. The analysis has considered the impact of corporate debt in influencing mergers among the population of United States local exchange companies between 1988 and 2001 How mergers are influenced is important. Yet, there has been no evidence as to how firms’ financial structures influence mergers. Our analysis is the first, we are aware of, to examine the relationship between firm leverage and merger propensity, an important issue given high debt concerns. The extent of firm leverage can significantly influence merger decisions, either in retarding or promoting them. There are complexities involved in analyses of debt and mergers, because of the endogeneity of financial decisions by firms. The literatures have dealt with the extent of debt as a strategic decision to be modeled, using a range of covariates to explain the debt-choice decision. The need to account for debt as an endogenous variable is important, and the extent of debt as a strategic choice has been appropriately modeled.

We model debt-choice as an endogenous process, and augment the policy and corporate finance literatures by demonstrating the disentangling of complexities inherent in simultaneously evaluating corporate financing and merger decisions. For this evaluation, we use treatment effects modeling. This enables selection effects in strategic choice decisions, such as the taking-on of debt, to be accounted for. The technique enables the analysis of causality implicit in simultaneous strategic leverage and merger decisions. The use of treatment effects modeling advances analyses of firms’ strategic behavior in competition policy and economic analyses. Our results have shown that the impact of relatively higher corporate debt levels on merger events in the sector has been negative. The results are consistent with the idea of monitoring of firms with high debt levels; and such firms will be strategically passive, by not undertaking in risky mergers so as not to engender potentially-negative financial performance outcomes.