Banca dItalia Fourth BICEPR Conference on Money Banking

Banca d’Italia Fourth BI-CEPR Conference on Money, Banking and Finance “Corporate Governance, Capital Structure and Firm Performance” Banca d’Italia October 2 -3, 2009, Rome The Deep Pocket Effect of Internal Capital Markets by Chiara Fumagalli, Xavier Boutin, Giacinta Cestone, Giovanni Pica and Nicolas Serrano-Velarde discussion by Francesco Columba The usual disclaimer applies. The opinions are those of the discussant only and in no way involve the responsibility of the Bank of Italy.

Banca d’Italia Summary I • Role of financial strength in industry entry • Literature examined the role of individual firm deep pockets and the effect of groupaffiliation per se • Access to groups’ deep pockets, opposed to own financial resources, is a source of market power for firms? Yes, according to the authors • First finding: liquid wealth owned by affiliated subsidiaries in other markets is negatively related to entry in the incumbents’ market • But endogeneity is a major concern. Unobserved factors (eg group efficiency) could be driving simultaneously entry in a market and group cash holdings in other markets. • To address endogeneity the authors adopt a theory-driven empirical approach and test two predictions from Cestone and Fumagalli (2005) Rome, October 2, 2009 discussion by Francesco Columba 2

Banca d’Italia Summary II • First prediction tested: entry deterrence of group deep pockets stronger when group-affiliated incumbents have a more difficult access to external finance • Finding: industries where firms hold less collateralizable assets entry is more sensitive to group liquidity and less sensitive to incumbent liquidity • Second prediction tested: entry-deterring effect of group deep pockets boosted by the intensity of internal resource reallocation within the group • Finding: group deep pockets have a larger effect on entry in markets where within group capital market is more active (activity measured by intra-group lending, group diversification, financial intermediaries) • Contribution to deep pocket literature (financial muscle as competitive advantage) disentangling group financial strength from firm one and to internal capital markets literature analyzing business groups instead of multidivisional firms. • Overall results are against perfect capital market hypothesis Rome, October 21, 2009 discussion by Francesco Columba 3

Banca d’Italia Relevance of internal capital markets • Own funding in normal times matters, and its interaction with prodcut market competition may inform the competitive behavior of multimarket firms and groups • During times of financial distress internal resources relevance may be magnified. • As the debate on the existence of financial frictions (and on pro-cyclicality of the financial sector) underlines, during crises access to external finance may have a prohibitive price or be impossible at all as markets dry up or disappear. Access to internal resources becomes even more a strategic factor • According to the authors after a shock hits one of the sector within groups winner-picking (cash-poor groups) may coexist with cross-subsidization (cash-rich groups). What happens if the shock is not idiosyncratic to a sector? Rome, October 2, 2009 discussion by Francesco Columba 4

Banca d’Italia empirical design • Basic entry equation – Information set of entrant refers to one-year lagged variables – Exogeneity of cash holdings of the rest of the group in other markets to shocks to market i. What if common shocks? Behavior along the cycle? • Group deep pockets and external capital markets – Is the proxy of ease of access to external capital market (tangibles/assets) robust ? What if a deleveraging process develops due to funding and liquidity problems? – Is it the relevant one for access to securities and stock markets, to bank finance? – Rating? Relationship lending? • Group deep pockets and internal capital markets – 3 measures of ICM intensity: intra-group loans, number of financial intermediaries, group diversification • Data – Foreign owned groups Rome, October 2, 2009 discussion by Francesco Columba 5

Banca d’Italia results • Basic equation: – effect of business group total cash on entry deterrence is smaller than that of incumbent total cash • Deep pockets and access to external finance: – results using sub-sampling coherent with prediction: but conditional on accuracy of proxy of ease of access to credit – proxy measured at market level not at firm level • Deep pockets and ICM activity proxied from: – – intra-group loans financial intermediaries diversification more direct proxies, financial flows? Rome, October 2, 2009 discussion by Francesco Columba 6

Banca d’Italia Issues • Determinants of cash accumulation within a group; drivers of groups cash holdings decisions. Is endogeneity tackled? • Dynamics? Is the cash flow held to pick up opportunities or to do corporate hedging? Or to support needy firms (socialist equilibrium)? • Welfare implications? • Liquidity measures? Is access to external finance controlled for? Relationship lending? Rome, October 2, 2009 discussion by Francesco Columba 7

Banca d’Italia Policy implications • potential threat to entry posed by the deep pockets of the group incumbent is affiliated with. • detrimental effect to be traded off with efficiency gains from group deep pockets and the financial slack provided • Cash-rich groups may have a pro-competitive effect favoring entry in opposition to incumbents’ predatory strategies Rome, October 2, 2009 discussion by Francesco Columba 8