The effect of product differentiation on strategic financing decisions. Richard Fairchild University of Bath School of Management Working Paper Series 2004.07

This working paper is produced for discussion purposes only. The papers are expected to be published in due course, in revised form and should not be quoted without the author’s permission.

1

University of Bath School of Management Working Paper Series University of Bath School of Management Claverton Down Bath BA2 7AY United Kingdom Tel: +44 1225 826742 Fax: +44 1225 826473 http://www.bath.ac.uk/management/research/papers.htm 2004 2004.01

Stephan C. M. Henneberg

The Conundrum of Leading or Following in Politics? An Analysis of Political Marketing Postures

2004.02

Y. L Chen and Stephan C. M. Henneberg

Political Pulling Power. Celebrity Political Endorsement and Campaign Management for the Taipei City Councillor Election 2002

2004.03

Stephan C. M. Henneberg, Stefanos Mouzas and Pete Naudé

Network Pictures – A Concept of Managers’ Cognitive Maps in Networks

2004.04

Peter Reason

Education for Ecology: Science, Aesthetics, Spirit and Ceremony

2004.05

Yvonne Ward & Andrew Graves

A New Cost Management & Accounting Approach For Lean Enterprises

2004.06

Jing Lin Duanmu & Felicia Fai

Assessing the context, nature, and extent of MNEs’ Backward knowledge transfer to Chinese suppliers

2004.07

Richard Fairchild

The effect of product differentiation on strategic financing decisions.

2

The effect of product differentiation on strategic financing decisions.

Richard Fairchild E-mail: [email protected]

Abstract. In a model of duopoly Bertrand competition in differentiated products, rival firms may use the debt level strategically to soften price competition and increase profits. The equilibrium choice of debt level depends on the degree of differentiation, and the firms’ short-term and long-term incentives. At maximum product differentiation, the rivals always use zero debt. As differentiation (and market power) reduces, firms may be induced to increase the debt level to keep prices high. At low levels of differentiation (high levels of competition), firms may return to low debt levels, as the predation effect dominates.

1. Introduction. Financial and industrial economists have increasingly recognised the interaction between product market competition and financing decisions of firms. A firm may use financial leverage strategically to affect a rival’s behaviour. There are two main modelling approaches; limited liability models and deep purse, or predation, models. In limited liability models, firms pursue policies that transfer wealth from debt holders to equity holders. Hence, firms have an incentive to set high debt levels. Under Cournot quantity competition (Brander and Lewis 1986), firms with higher debt levels compete more aggressively by increasing output and profit at their rivals’ expense. Under Bertrand price competition (Showalter 1995) rivals can use debt financing to soften price competition. Therefore, the limited liability models predict a positive relationship between market power and debt.

©Richard Fairchild 2004

3

In the deep purse or predation models (Brander and Lewis 1986; Bolton and Scharfstein 1990), an unlevered firm can enter a market to steal market share from a highly levered firm, or even drive it into bankruptcy. Therefore, in contrast to the limited liability models, the predation models predict a negative relationship between market power and debt. Empirical tests of these models have provided mixed results.

Some researchers find a

negative relationship ( Titman and Wessels 1988; Chevalier 1995; Barclay et al 1995; Rajan and Zingales 1995; Barclay and Smith 1996), while others find a positive relationship (Krishnawamy et al 1992, Phillips 1995, Michaelas et al 1999; and Rathinsamy et al 2000). Pandey (2000) finds a non-linear (cubic) relationship. Recognition of these conflicting theoretical and empirical results has motivated development of the model in this paper. We present a new theoretical approach which analyses the effects of the degree of market power on equilibrium debt levels. We base our model on the analysis of Dasgupta and Titman (1998), in which long-term debt softens price competition by inducing the rivals to focus on short-term pricing decisions. We develop a model of duopoly Bertrand competition in differentiated products, in which rival firms may use the debt level strategically to soften price competition and increase profits. The equilibrium choice of debt level depends on the degree of differentiation, and the firms’ short-term and long-term incentives. At maximum product differentiation, the rivals always use zero debt. As differentiation (and market power) reduces, firms may be induced to increase the debt level to keep prices high. At low levels of differentiation (high levels of competition), firms may return to low debt levels. Hence, our results support Pandey’s finding of a non-monotonic relationship.

©Richard Fairchild 2004

4

2. The Model. Consider a financial structure/product pricing game played under duopoly competition. At date 0, the firms simultaneously choose their financial structures. At date 1, the rivals simultaneously set product prices. At this stage, they face Bertrand price competition in differentiated products. The firms then operate in the product market over two periods (date 1 and date 2). The players are risk-neutral, and the risk-free rate is zero. We build on the model of Dasgupta and Titman (1998). In deciding on their date 1 prices, the firms face the following trade-off. The short-term date 1 price affects the firms’ market shares. Assuming that there is a certain customer ‘stickiness’ (customers buying from one firm in date 1 tend to buy from that same firm in date 2), the date 1 price affects date 1 and 2 profits. The higher the date 1 price, the higher the date 1 profits, but also the lower the market share, and the lower the date 2 profits. In deciding on date 1 prices, the firms face this tradeoff between date 1 and date 2 profits. In addition, firms face price competition from each other. In summary, there are two forces driving prices down; Bertrand price competition, and the firms’ desire for long-term market share. Increasing long-term financial leverage softens price competition by inducing firms to focus on short-term pricing and profits. Since they are less interested in future market share, they compete less aggressively in short-term prices.

The period 1 industry demand function is Q = 2[α − p ], where Q is the quantity demanded

and p is the price. The period 1 demand function for firm i is

qi = α − pi + γ [ p j − pi ].

©Richard Fairchild 2004

(1)

5

where γ represents the degree of differentiation between firm i and firm j. Increasing γ reflects reducing product differentiation (and, therefore, reducing market power), which implies that firm i can capture greater market share by undercutting firm j. When γ = 0, each firm has a local monopoly, and the pricing decision of one firm has no effect on the quantity demanded from the other firm.

We assume ‘no-switching’ of customers between date 1 and date 2; the quantity of customers buying from firm i in period 1, qi , buys from the same firm in period 2. Hence, each firm has a local monopoly in period 2, and the quantity demanded in period 2 from firm i is q2 = qi . Note that this implies that customers who did not buy from either firm in date 1 drop out of the market completely at date 2.

The firms face date 2 demand uncertainty. With equal probability, all customers have one of two possible reservation prices; P2 > 0, or zero. Since each firm has a local monopoly in date 2, each firm can charge the customers their date 2 reservation price. We assume zero costs of production. Therefore, each firm’s date 2 profit is either q2 P2 = qi P2 or zero, with equal probability. Hence, at date 1, firm i' s expected period 2 profits are ∏ 2 = qi

P2 . Firm i' s 2

period 1 profit is ∏1 = pi qi . Since players are risk-neutral and the risk-free rate is zero, the value of firm i can be written as Vi = ∏1 + ∏ 2 . That is,

Vi = qi ( pi +

P2 ). 2

©Richard Fairchild 2004

(2)

6

Define the rivals’ security issue as S i and S j respectively. For simplicity, we allow each firm to choose from one of two possible levels of long-term debt D. Each firm can either choose an all-equity (zero debt) structure; S i = E , or each firm can choose a “high-debt” structure

with date 2 repayment value Si = D = qi P2 . In summary, S i , S j ∈ {E , D}.

We solve the financial contracting/pricing game by backward induction.

3. Pricing Decisions for given debt levels.

Firstly, take the date 0 financial contracts Si , S j as given, and solve for the firms’ equilibrium prices pi ( Si , S j ), and p j ( S j , Si ). . In order to do so, we substitute qi from equation (1) into equation (2) to obtain;

Vi = αpi − pi + γpi p j − γpi + (α − pi + γ [ p j − pi ]) 2

2

P2 . 2

(3)

If both firms have chosen the all-equity contract at date 0 ( D = 0), they set date 1 prices to maximise (3), taking the other firm’s price as given. We obtain the equilibrium prices by solving ∂Vi / ∂Pi = 0, and recognising that, in equilibrium, pi * = p j * .

If both firms issue high debt level D = qi P2 , they set date 1 prices to maximise the expected value of date 1 equity;

∏ i = αpi − pi + γpi p j − γpi . 2

©Richard Fairchild 2004

2

(4)

7

We obtain the equilibrium prices by solving ∂ ∏ i / ∂Pi = 0, and recognising that, in equilibrium, pi * = p j * .

If firm i has chosen the all-equity contract, and firm j has chosen the high debt contract at date 0, we solve ∂Vi / ∂Pi = 0, and ∂ ∏ j / ∂Pj = 0, to obtain equilibrium prices.

Therefore, the equilibrium prices for each combination of Si , S j ∈ {E , D} are given in lemma 1.

Lemma 1: The equilibrium prices for given Si , S j ∈ {E , D} are;

a) pi * ( E , E ) = p j * ( E , E ) =

α − 0.5(1 + γ ) P2 . 2+γ

b) pi * ( D, D ) = p j * ( D, D ) =

α 2+γ

.

2α + 3γα − 0.5γ (1 + γ ) P2 2α + 3γα − (1 + γ ) 2 P2 c) pi * ( E , D ) = , p j * ( D, E ) = . 2 3γ 2 + 8γ + 4 3γ + 8γ + 4

i) If γ = 0, firm i' s equilibrium price is independent of firm j ' s security choice. Further, debt provides a higher price than equity; pi * ( D, D ) = pi * ( D, E ) > pi * ( E , D ) = pi * ( E , E ). ii) If γ > 0, debt softens price competition. Further, each firm’s equilibrium price depends on the other firm’s equilibrium price; pi * ( D, D ) > pi * ( D, E ) > pi * ( E , D ) > pi * ( E , E ).

©Richard Fairchild 2004

8

Note that, the highest equilibrium prices occur when both firms issue debt. The lowest equilibrium prices occur when both firms issue equity. If firm j issues equity, firm i can commit to a higher price by setting debt, but this price is lower than when both firms issue debt.

Now move back to date 0 to solve for the rivals’ equilibrium financial contracts.

4. The firms’ simultaneous choice of debt levels.

The rivals choose date 0 financial contracts Si , S j ∈ {E , D}, to maximise firm value, given the other firm’s choice of debt level. In doing so, each firm recognises that its own choice, and that of its rival, will affect date 1 prices, as given by lemma 1. In order to solve for the equilibrium date 0 financial contracts, we substitute the equilibrium prices given in lemma 1 for each pair of debt levels into (1) to obtain equilibrium quantities. We then substitute equilibrium prices and equilibrium quantities into (2) to obtain equilibrium firm values Vi ( S i , S j ) and V j ( S j , S i ) for each pair of debt levels. Finally, we solve for the date 0 debt levels by finding the Nash equilibria of the normal form game.

Lemma 2: The firm values for given S i , S j ∈ {E , D} are;

a) Vi ( E , E ) = [

α + γα + 0.5(1 + γ ) P2 α − 0.5(1 + γ ) P2 P2 ][ + ] 2+γ 2+γ 2

b) Vi ( D, D ) = [

α + γα α P ][ + 2] 2+γ 2+γ 2

©Richard Fairchild 2004

9

c) Vi ( E , D ) = [

3γ 2α + 5γα + 2α + (1 + γ ) 3 P2 − 0.5γ 2 (1 + γ ) P2 2α + 3γα − (1 + γ ) 2 P2 P2 + ], ][ 3γ 2 + 8γ + 4 2 3γ 2 + 8γ + 4

d) Vi ( D, E ) = [

3γ 2α + 5γα + 2α − 0.5γ (1 + γ ) 2 P2 2α + 3γα − 0.5γ (1 + γ ) P2 P2 + ]. ][ 3γ 2 + 8γ + 4 2 3γ 2 + 8γ + 4

Therefore, firm i' s best response functions, given S j = E or S j = D, are

Vi ( D, E ) − Vi ( E , E ) =

αP2 [1.5γ 4 + 2.5γ 3 + γ 2 ] − P2 2 (1 + 5γ + 9γ 2 + 7.25γ 3 + 2.75γ 4 + 0.5γ 5 ) , (3γ 2 + 8γ + 4) 2

αP2 [1.5γ 4 + 2.5γ 3 + γ 2 ] − P2 2 (1 + 5γ + 9γ 2 + 7γ 3 + 2.25γ 4 + 0.25γ 5 ) , Vi ( D, D ) − Vi ( E , D ) = (3γ 2 + 8γ + 4) 2

respectively.

Examination of the best response functions Vi ( D, E ) − Vi ( E , E ) and Vi ( D, D ) − Vi ( E , D ) reveals the following;

Lemma 3: The equilibrium security issuance is as follows;

a) If γ = 0, Vi ( D, E ) − Vi ( E , E ) = Vi ( D, D ) − Vi ( E , D ) = −

2

P2 . Firm i' s dominant strategy is 16

to issue equity. By symmetry, this is also firm j' s dominant strategy. Hence, the equilibrium is {Si = S j = E}.

©Richard Fairchild 2004

10

b) Vi ( D, D ) − Vi ( E , D ) > Vi ( D, E ) − Vi ( E , E ), ∀γ > 0. If Vi ( D, D ) − Vi ( E , D ) > Vi ( D, E ) − Vi ( E , E ) ≥ 0, the equilibrium is {Si = S j = D}. If Vi ( D, D ) − Vi ( E , D ) ≥ 0 > Vi ( D, E ) − Vi ( E , E ), the multiple equilibria are {Si = S j = D}, and {Si = S j = E}. If 0 > Vi ( D, D ) − Vi ( E , D ) > Vi ( D, E ) − Vi ( E , E ), the equilibrium is {Si = S j = E}.

Examination of the best response functions Vi ( D, E ) − Vi ( E , E ) and Vi ( D, D ) − Vi ( E , D ) reveals that the equilibria depend on product differentiation γ , the firms’ long term incentives (affected by P2 ), and the firms’ short-term incentives (affected by α ). For simplicity, we take the firms’ short-term and long-term incentives as given, and focus on the effects of γ on the equilibria.

We solve using the parameter values α = 3,500 and P2 = 1,000.

We consider product

differentiation in the interval γ ∈ [0,6]. This ensures that equilibrium prices are non-negative for any combination of securities; pi ( Si , S j ) ≥ 0, p j ( Si , S j ) ≥ 0 ∀Si , S j ∈ {E , D}. Diagram 1 demonstrates firm i' s best response functions; Vi ( D, D ) − Vi ( E , D ) (the upper line), and Vi ( D, E ) − Vi ( E , E ) (the lower line).

We obtain our main result (this relationship is demonstrated in diagram 2);

©Richard Fairchild 2004

11

Proposition 1: For the parameter values α = 3,500 and P2 = 1,000, the equilibrium is

{Si = S j = E} ∀γ ∈ (0,1.5]; {Si = S j = E} and {Si = S j = D} ∀γ ∈ (1.5,1.8]; {Si = S j = D} ∀γ ∈ (1.8,5]; and {Si = S j = E} and {Si = S j = D} ∀γ ∈ (5,6].

Therefore, for these parameters, the relationship between market power and debt is nonmonotonic (as in Pandey 2000). This is affected by a combination of the limited liability and predation effects. In the low range of γ (that is, high differentiation/high market power) the limited liability effect dominates. That is, the rivals use debt to soften price competition. However, note that debt and market power are substitutes. At maximum differentiation (local monopolies; γ = 0), the rivals do not need to issue debt to soften price competition. As differentiation reduces, the rivals increase debt (debt substitutes for market power in softening price competition).

When differentiation reduces sufficiently, there are multiple equilibria. Now the predation effect dominates (a firm with low debt can steal market share by setting a lower price than a firm with high debt), thus driving debt down.

©Richard Fairchild 2004

12

Increase in value due to issuing debt

Firm i's security reaction functions: alpha = 3500, P2 = 1000. 100000 50000 0 0

-50000

1

2

3

4

5

6

-100000 Gamma

Diagram 1.

Effect of Product Differentiation on Equilibrium Debt Levels 3500000

Debt Level.

3000000 2500000 2000000 1500000 1000000 500000 0 0

1

2

3

4

5

6

Differentiation Parameter

Diagram 2.

©Richard Fairchild 2004

13

Conclusion.

We have developed a financial contracting model in which firms use debt strategically to soften Bertrand price competition. Our main result is that, when firms have local monopolies, they both issue equity in equilibrium. As differentiation reduces, firms substitute debt for equity in order to soften price competition (limited liability effect). When differentiation reduces sufficiently, the predation effect dominates, and the debt level is driven down. Our simple approach provides a good basis for future theoretical and empirical research. We have analysed the effects of the differentiation parameter γ on equilibrium debt levels for particular long-term and short-term parameters P2 and α . Development of the model will analyse the effect of these parameters on the equilibria. Furthermore, we have only considered two possible debt levels (high or zero). We may develop this model to include more debt levels, at the expense of tractability (see, for example, Fairchild (2004), who includes zero, medium and high debt levels). Furthermore, future research will examine the impact of agency costs, and sequential predation (as in Bolton and Scharfstein 1990) in this model.

©Richard Fairchild 2004

14

References:

Bolton, P., and Scharfstein D., 1990. A Theory of predation based on agency problems in financial contracting. American Economic Review. 93-106.

Barclay, M.J. and Smith, C.W., 1996. On financial architecture: leverage, maturity and priority. Journal of Applied Corporate Finance 8 (4), 4-17.

Barclay, M.J., Smith, C.W., and Watts, R.L. 1995. The determinants of corporate leverage and dividend policies. Journal of Applied Corporate Finance 7 (4), 4-19.

Brander, J.A., and Lewis, T.R., 1986. Oligopoly and financial structure: the limited liability effect. American Economic Review 956-970.

Chevalier, J., 1995. Capital structure and product-market competition: empirical evidence from the supermarket industry. American Economic Review; 415-435.

Dasgupta, S, and Titman, S.,

1998. Pricing Strategy and Financial Policy. Review of

Financial Studies; 705-737.

Fairchild, R. 2004. Potential product market competition, financial structure, and actual competitive intensity. Mimeo, SSRN database.

Grullon, G; Kanatas, G., and Kumar. P., 2002. Financing decisions and advertising: an empirical study of capital structure and product market competition. Mimeo, SSRN database.

©Richard Fairchild 2004

15

Krishnawamy, C.R., Mangla, I. And Rathinasamy, R.S. 1992. An empirical analysis of the relationship between financial structure and market structure. Journal of Financial and Strategic Decisions, 5 (3), 75-88.

Michaelas, N., Chittenden, F. and Poutziouris, P. 1999. Financial policy and capital structure choice in U.K. SMEs: empirical evidence from company panel data.

Small Business

Economics, 12, 113-130.

Pandey, I.M. 2002. Capital structure and market power interaction: evidence from Malaysia. Mimeo, SSRN database.

Philllips, G. 1995. Increased debt and industry product markets: An empirical analysis.” Journal of Financial Economics 189-238.

Rajan, R. G. and Zingales, L. 1995. What do we know about capital structure? Some evidence from international data, Journal of Finance 50 (5), 1421-1460.

Rathinasamy, R.S., Krishnaswamy, C.R. and Mantipragada, K.G. 2000. Capital structure and product market interaction: an international perspective. Global Business and Finance Review 5 (2), 51-63.

Showalter, D.

1995 Oligopoly and financial structure: comment.

American Economic

Review. 647-653.

©Richard Fairchild 2004

16

Titman, S. and Wessels, R. 1988. The Determinants of Capital Structure Choice. Journal of Finance, 43 (1) 1-19.

©Richard Fairchild 2004

17

University of Bath School of Management Working Paper Series Past Papers

University of Bath School of Management Claverton Down Bath BA2 7AY United Kingdom Tel: +44 1225 826742 Fax: +44 1225 826473 http://www.bath.ac.uk/management/research/papers.htm 2003 2003.01

Stephan C. M. Henneberg

The Conundrum of Leading or Following in Politics? An Analysis of Political Marketing Postures (Forthcoming in Journal of Political Marketing, Vol. 5, 2005)

2003.02

Richard Fairchild

Management’s optimal financial contracts, the degree of alignment with investors, and the ‘carrot and stick’ role of debt.

2003.03

Richard Fairchild

An Investigation of the Determinants of BT’s Debt Levels from 1998-2002: What does it tell us about the Optimal Capital Structure?

2003.04

Steve Brown & Felicia Fai

Strategic Resonance Between Technological and Organisational Capabilities in the Innovation Process within Firms

2003.05

Paul Goodwin

Providing Support for Decisions based on Time Series Information Under Conditions of Asymmetric Loss (Forthcoming in the European Journal of Operational Research, 2004)

2003.06

Will Liddell & John H Powell

Are you still here? Reconciling patient access and GP effectiveness in the management of a large medical practice: a case study using QPID

2003.07

Felicia Fai

A Structural Decomposition Analysis of Technological Opportunity in Firm Survival and Leadership

2003.08

John Purcell & Nick Kinnie

Employment Regimes for the Factories of the Future: Human Resource Management in Telephone Call Centres

2003.09

Juani Swart & Nick Kinnie

The impact of client-relationships on organisational form and HR practices

2003.10

Sue Hutchinson, Nick Kinnie & John Purcell

HR Practices and Business Performance: what makes a difference?

2003.11

Bruce Rayton, Kim Hoque & John Purcell

Does one size fit all? : Exploring the impact of employee perceptions of HR practices on motivation levels

2003.12

Sue Hutchinson, Nick Kinnie & John Purcell

Bringing Policies to Life: Discretionary Behaviour and the Impact on Business Performance

18

2003.13

Stephan C. M. Henneberg

CRM Implementation: Hard Choices and Soft Options (Forthcoming in Journal of Relationship Marketing, 2005, Vol. 3, No. 1)

2003.14

Stephan C. M. Henneberg

Move aside, advocatus diaboli: It is time to hear the position of the advocatus dei on political marketing (Forthcoming in Journal of Public Affairs, Vol. 4, No. 3, 2004)

2003.15

Nick Kinnie, Juani Swart & John Purcell

Influences on the choice of HR systems: the network organisation perspective

2003.16

Juani Swart, Nick Kinnie & John Purcell

Managing the careers of IT professionals: A competing identity perspective

2003.17

Catherine Phillips

How do consumers express their identity through the choice of products that they buy?

2003.18

Peter Reason

Choice and quality in action research practice

2003.19

Stephan C. M. Henneberg

Generic Functions of Political Marketing Management

2003.20

Niki Panteli

Situating Trust within Virtual Teams

2003.21

Louise Knight & Annie Pye

Network learning: An empirically-derived model of learning by groups of organizations

19