Laeven and Levine (2007) adapted the excess-value methodology of the industrial-diversification literature to financial institutions and found that financial conglomerates combining lending with non-lending financial services trade below the value of comparable specialized intermediaries - a diversification discount consistent with intensified agency problems outweighing economies of scope - while explicitly declining to identify a single causal factor, because agency costs were not measured and activity-level data were coarse. The subsequent literature has contested rather than settled the question. US evidence confirms a persistent discount that appears to be caused by diversification (Schmid and Walter, 2009), but European and cross-country studies find no universal discount or even valuation gains (van Lelyveld and Knot, 2009; Elsas, Hackethal, and Holzhäuser, 2010; Filson and Olfati, 2014); quasi-experimental causal identification exists only for geographic, not activity, diversification (Goetz, Laeven, and Levine, 2013); and the excess-value matching methodology itself is now challenged as manufacturing the discount (Villalonga, 2004; Hund, Monk, and Tice, 2024). The open problem is to establish, with credible identification and measurement robust to the matching critique, whether activity diversification causally reduces the market value of financial conglomerates, and through which mechanism.

Background

Financial conglomerates are institutions that conduct several financial activities under one corporate roof, most commonly traditional lending alongside non-lending services such as securities underwriting, brokerage, asset management, or insurance. Whether this organizational form creates or destroys shareholder value is an old question in the economics of financial intermediation, and one with direct regulatory stakes: the United States prohibited the combination of commercial and investment banking from 1933 until the Gramm-Leach-Bliley Act of 1999 re-permitted it, several European countries maintained universal banking throughout, and post-crisis separation proposals after 2008 revisited the same trade-off. The theoretical tension is between economies of scope - cost savings or revenue synergies from producing several financial services jointly, for example reusing client information across lending and underwriting - and agency problems: insiders of a complex, opaque, multi-activity institution are harder for outside shareholders and creditors to monitor, so managers can extract larger private benefits at the owners' expense.

The empirical benchmark comes from the corporate finance of industrial (non-financial) firms. Lang and Stulz (1994) and Berger and Ofek (1995) documented that diversified firms trade at a discount relative to portfolios of comparable focused firms. Berger and Ofek's measure, excess value, is a chop-shop construction: estimate the value the conglomerate would command if split into specialized pieces by weighting each segment's assets or sales with the valuation multiple of median single-segment firms in that industry, then compare the sum with the conglomerate's actual market value, typically expressed through Tobin's q - the ratio of the market value of a firm's assets to their replacement cost, approximated for banks by market value over book assets. A negative excess value means the market values the integrated firm below its specialized parts. Whether this correlation reflects causation was contested from the start: Campa and Kedia (2002) show that much of it disappears once the selection of weak firms into diversification is modeled, and Villalonga (2004), using establishment-level rather than business-segment data, finds the discount vanishes or reverses.

Laeven and Levine (2007) transplanted this methodology to financial institutions, which had been largely absent from the debate despite being the sector where the scope-versus-conflicts policy argument is loudest and where opacity plausibly makes agency problems most severe. Using data on 836 banks across 43 countries over 1998-2002, they compare each diversified bank's q with the value-weighted q of hypothetical specialized lenders and non-lenders constructed from focused banks. Diversified banks exhibit a statistically and economically significant discount (on the order of 6 percent of average q in the working-paper version), robust to bank- and country-level controls, alternative diversification measures, and four instrumental-variable strategies that use, among other instruments, country-level regulatory restrictions on bank activities. Their reading: the results are consistent with theories in which diversification intensifies agency problems, and economies of scope are not large enough to produce a diversification premium. But they state explicitly that they do not measure agency problems directly, so they cannot unequivocally attribute the discount to agency costs, and they flag the coarseness of activity-level data (net income or aggregate non-loan assets rather than activity-by-activity gross revenues) as a limit on quantifying scope economies.

Later work has contested rather than settled the question. Schmid and Walter (2009) confirm a substantial, persistent conglomerate discount in a large US sample covering 1985-2004, with patterns suggesting diversification causes the discount rather than the reverse. But van Lelyveld and Knot (2009) find no universal discount among European bank-insurance conglomerates, with the discount varying systematically with size, risk profile, and familiarity with the conglomerate business model; Elsas, Hackethal, and Holzhäuser (2010), in a nine-country panel, find no conglomerate discount at all - revenue diversification is associated with higher profitability and higher market valuation; and Filson and Olfati (2014), studying acquisitions after the Gramm-Leach-Bliley liberalization, find that diversification into investment banking, brokerage, and insurance created value on average. On the mechanism side, Goetz, Laeven, and Levine (2013) provide the cleanest causal evidence to date - using the dynamic process of US interstate bank deregulation, they show that exogenous increases in geographic diversity reduce bank holding company valuations and worsen insider lending and loan quality, consistent with agency theories - but their evidence concerns geographic rather than activity diversification. Klein and Saidenberg (2010) show that at US bank holding companies the discount is partly a matter of organizational structure (the number of banking subsidiaries) rather than industry mix, and Stiroh and Rumble (2006) show that diversification gains are offset by exposure to volatile non-interest income. Finally, the measurement foundation itself has come under attack: Hund, Monk, and Tice (2024) show that the Berger-Ofek procedure systematically matches large, old diversified firms with small, young focused firms and thereby manufactures the discount - with sales-and-age-consistent matching the return-weighted discount for industrial firms is zero - which raises, but does not answer, the question of how much of the financial-conglomerate discount is a construction artifact.

Problem Statement

Laeven and Levine's title question - is there a diversification discount in financial conglomerates? - was answered affirmatively for their own sample, but the question they themselves left open, and that the later literature has contested without settling, is its causal and mechanistic core. The problem: for financial conglomerates, determine whether combining lending with non-lending financial activities reduces market value relative to comparable specialized intermediaries, and why. The question's causal, mechanism, and measurement components are linked parts of one objective - explaining the valuation effect of financial conglomeration: (1) Causal effect: an estimate of the effect of activity diversification on market value that separates the consequences of diversifying from the selection of particular institutions into diversification and from country- and period-level institutional confounders. (2) Mechanism: attribution of the estimated effect (including a zero or positive effect) to specific channels - intensified agency problems, insufficient economies of scope, organizational-structure frictions, or measurement artifact - with evidence that discriminates among them, ideally including direct or proxy measurement of agency costs in financial conglomerates, which the source paper lacked. (3) Measurement robustness: a valuation counterfactual that does not mechanically generate the result through peer mismatch - the standard excess-value matching procedure is known, for industrial firms, to pair large, old diversified firms with small, young focused firms and thereby to manufacture a discount (Hund, Monk, and Tice, 2024) - so that the answer is robust to the critique that the standard procedure manufactures the discount in this setting. An answer that establishes no causal discount - a credible-identification null, or a demonstration that the measured financial-conglomerate discount is an artifact of the matching or data construction in this setting - resolves the problem as fully as one that establishes and explains a discount.

References

  1. Laeven, Luc, and Ross Levine (2007). Is there a diversification discount in financial conglomerates? Journal of Financial Economics 85(2), 331-367. DOI: 10.1016/j.jfineco.2005.06.001. https://doi.org/10.1016/j.jfineco.2005.06.001
  2. Schmid, Markus M., and Ingo Walter (2009). Do financial conglomerates create or destroy economic value? Journal of Financial Intermediation 18(2), 193-216. DOI: 10.1016/j.jfi.2008.07.002. https://doi.org/10.1016/j.jfi.2008.07.002
  3. van Lelyveld, Iman, and Klaas Knot (2009). Do financial conglomerates create or destroy value? Evidence for the EU. Journal of Banking & Finance 33(12), 2312-2321. DOI: 10.1016/j.jbankfin.2009.06.007. https://doi.org/10.1016/j.jbankfin.2009.06.007
  4. Elsas, Ralf, Andreas Hackethal, and Markus Holzhäuser (2010). The anatomy of bank diversification. Journal of Banking & Finance 34(6), 1274-1287. DOI: 10.1016/j.jbankfin.2009.11.024. https://doi.org/10.1016/j.jbankfin.2009.11.024
  5. Goetz, Martin R., Luc Laeven, and Ross Levine (2013). Identifying the valuation effects and agency costs of corporate diversification: Evidence from the geographic diversification of U.S. banks. Review of Financial Studies 26(7), 1787-1823. DOI: 10.1093/rfs/hht021. https://doi.org/10.1093/rfs/hht021
  6. Filson, Darren, and Saman Olfati (2014). The impacts of Gramm-Leach-Bliley bank diversification on value and risk. Journal of Banking & Finance 41, 209-221. DOI: 10.1016/j.jbankfin.2014.01.019. https://doi.org/10.1016/j.jbankfin.2014.01.019
  7. Stiroh, Kevin J., and Adrienne Rumble (2006). The dark side of diversification: The case of US financial holding companies. Journal of Banking & Finance 30(8), 2131-2161. DOI: 10.1016/j.jbankfin.2005.04.030. https://doi.org/10.1016/j.jbankfin.2005.04.030
  8. Klein, Peter G., and Marc R. Saidenberg (2010). Organizational structure and the diversification discount: Evidence from commercial banking. Journal of Industrial Economics 58(1), 127-155. DOI: 10.1111/j.1467-6451.2010.00409.x. https://doi.org/10.1111/j.1467-6451.2010.00409.x
  9. Lang, Larry H. P., and René M. Stulz (1994). Tobin's q, corporate diversification, and firm performance. Journal of Political Economy 102(6), 1248-1280. DOI: 10.1086/261970. https://doi.org/10.1086/261970
  10. Berger, Philip G., and Eli Ofek (1995). Diversification's effect on firm value. Journal of Financial Economics 37(1), 39-65. DOI: 10.1016/0304-405X(94)00798-6. https://doi.org/10.1016/0304-405X(94)00798-6
  11. Campa, Jose Manuel, and Simi Kedia (2002). Explaining the diversification discount. Journal of Finance 57(4), 1731-1762. DOI: 10.1111/1540-6261.00476. https://doi.org/10.1111/1540-6261.00476
  12. Villalonga, Belén (2004). Diversification discount or premium? New evidence from the Business Information Tracking Series. Journal of Finance 59(2), 479-506. DOI: 10.1111/j.1540-6261.2004.00640.x. https://doi.org/10.1111/j.1540-6261.2004.00640.x
  13. Hund, John E., Donald Monk, and Sheri Tice (2024). The Berger-Ofek diversification discount is just poor firm matching. Critical Finance Review 13, 1-44. DOI: 10.1561/104.00000135. https://doi.org/10.1561/104.00000135