Dastidar (2009) showed that US firms self-select into international diversification and that, once this selection is corrected with a Heckman-type two-stage procedure and chop-shop imputed-value benchmarks (imputing the value of each segment from industry-median valuation multiples of comparable stand-alone firms), multinational firms trade at a premium rather than the discount reported by the earlier global-diversification literature. The paper's own limitations, however, define what remains open: the correction rests on a weakly specified first-stage choice model, ownership and governance variables are endogenous and not identified, multinational conglomerates cannot be decomposed jointly by industry and geography because segment-level data were unavailable, and the imputed benchmarks may be sensitive to multiplier heterogeneity, cross-country accounting differences, and the segment reporting regime. The open problem is to establish the causal effect of international diversification on firm value under these strengthened conditions. Later work has advanced pieces of this agenda — Heckman-based replications finding a global premium (Chang, Kogut and Yang 2016), dynamic-endogeneity methodology for international business panels (Li, Ding, Hu and Wan 2021), and exploitation of post-1998 geographic segment disclosures (Olibe, Rezaee, Flagg and Ott 2019; Chuwonganant, Olibe, Strawser and Strawser 2023) — but no located study delivers the selection-corrected joint industrial-by-geographic valuation of multinational conglomerates or credibly identifies ownership endogeneity, and reviews through 2017–2022 still report no consensus on the multinationality–value relation. The problem remains open.

Background

A classic question in corporate finance is whether corporate diversification — operating in several industries or countries — creates or destroys shareholder value. The standard measurement tool is the excess-value or 'chop-shop' method of Berger and Ofek (1995): each business segment of a diversified firm is assigned an imputed value by applying industry-median valuation multiples (ratios of market value to sales, assets, or earnings) computed from single-segment, stand-alone firms in the same industry, and the sum of the imputed segment values is compared with the firm's actual market value. The logarithm of actual value divided by imputed value is the firm's excess value; a systematically negative excess value for diversified firms is the 'diversification discount'. Berger and Ofek documented such a discount for industrially diversified US firms, and Denis, Denis and Yost (2002) showed that US firms diversifying both industrially and globally trade at the largest discounts.

The discount need not mean that diversification destroys value, because firms are not randomly assigned to diversification: they choose it. This is self-selection or endogeneity of the diversification decision. Campa and Kedia (2002) and Villalonga (2004), using two-stage selection corrections in the tradition of Heckman (1979) and alternative establishment-level data, showed that much of the industrial diversification discount reflects pre-existing differences between firms that diversify and firms that do not. In a Heckman two-stage design, a first-stage choice model — here a probit, a maximum-likelihood model for a binary outcome, estimating the probability that a firm is internationally diversified as a function of firm, industry, and home-country characteristics — is used to construct a correction term that is then included in the second-stage valuation regression, so that the estimated effect of diversification on value is net of the selection process.

Dastidar (2009) applied this design to international diversification. A multinational enterprise (MNE) is a firm with operations in more than one country; foreign direct investment (FDI) is the acquisition or creation of productive assets abroad. Using US firm-level data with industrial and geographic segment detail, and imputed-value benchmarks constructed so that an MNE is compared with the sum of stand-alone parts of comparable international exposure, the paper finds that the decision to diversify internationally is endogenous, that firms more likely to diversify internationally have lower firm values, and that — contrary to the global diversification discount literature — multinational firms are valued at a premium once the endogeneity of the diversification decision is controlled for. This parallels the selection critique in the industrial diversification literature.

The paper's own limitations section, which is the source of the present problem, identifies weaknesses that bear on how definitive this conclusion can be. First, the first-stage probit has a low pseudo-R-squared (a goodness-of-fit measure for discrete-choice models) and omits determinants of the diversification decision such as merger-and-acquisition activity and stock-listing characteristics, because the data did not permit them. Second, ownership and governance variables — managerial ownership, blockholders (large outside shareholders), and bank ownership — are endogenous and not identified, although they may interact with the diversification–value relationship. Third, the analysis could not handle multinational conglomerates (firms diversified both industrially and geographically), because consolidated segment detail did not permit decomposing segment-level assets and earnings across regions; this blocks a chop-shop analysis that is simultaneously industrial and geographic. Fourth, the imputed values may be affected by regional or reporting-induced differences in valuation multiples and by the accuracy of the commercial database's adjustments for cross-country accounting differences. Fifth, the sample period predates the replacement, after 1998, of the US segment-reporting rule FASB 14 by SFAS 131 (now ASC 280), under which firms report operating rather than strictly industrial or geographic segments; this limits temporal generalizability. Each limitation bears on the credibility of the selection-corrected premium, and together they define directions in which the identification of that premium can be strengthened.

Problem Statement

Establish whether international corporate diversification causally affects firm value once firms' self-selection into multinationality is accounted for, under strengthened identification — in particular, whether the premium reported for US multinationals survives that strengthening or whether some part of it is an artifact of the original identification. The overarching objective — a defensible causal valuation estimate for the international diversification effect — is advanced by the following strengthening directions drawn from the source paper's limitations. (i) Re-estimate the first-stage model of the diversification decision with richer determinants, including merger-and-acquisition activity and listing characteristics, so that the selection correction is not driven by a weakly specified choice model (the original probit had a low pseudo-R-squared). (ii) Credibly address the endogeneity of ownership and governance structures — managerial ownership, blockholders, and bank ownership — in the diversification–value relation. (iii) Extend the chop-shop imputed-value analysis to multinational conglomerates by decomposing segment-level assets and earnings jointly by industry and geography, so that the industrial and geographic components of the valuation effect are separately identified. (iv) Establish robustness of the resulting selection-corrected valuation effect to alternative multiplier constructions, to cross-country accounting adjustments, and to the post-1998 segment-reporting regime and longer sample periods. The re-estimation (with the sample, the data sources, the variables, the identification device, and the imputed-value benchmark construction fixed) rests on firm-level industrial and geographic segment data and comparable-firm valuation data covering both the pre-1998 FASB 14 and the post-1998 ASC 280 segment-reporting regimes, drawn from publicly accessible sources — the firms' own segment and financial-statement disclosures and public market data; the imputed-value benchmarks follow the Berger and Ofek (1995) imputed-value conventions defined in the Background.

References

  1. Dastidar, Protiti (2009). International corporate diversification and performance: Does firm self-selection matter? Journal of International Business Studies 40(1):71–85. DOI: 10.1057/palgrave.jibs.2008.57. https://doi.org/10.1057/palgrave.jibs.2008.57
  2. Berger, Philip G.; Ofek, Eli (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
  3. Denis, David J.; Denis, Diane K.; Yost, Keven (2002). Global diversification, industrial diversification, and firm value. The Journal of Finance 57(5):1951–1979. DOI: 10.1111/0022-1082.00485. https://doi.org/10.1111/0022-1082.00485
  4. Campa, Jose Manuel; Kedia, Simi (2002). Explaining the diversification discount. The Journal of Finance 57(4):1731–1762. DOI: 10.1111/1540-6261.00476. https://doi.org/10.1111/1540-6261.00476
  5. Villalonga, Belén (2004). Diversification discount or premium? New evidence from the Business Information Tracking Series. The 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
  6. Heckman, James J. (1979). Sample selection bias as a specification error. Econometrica 47(1):153–161. DOI: 10.2307/1912352. https://doi.org/10.2307/1912352
  7. Chang, Sungyong; Kogut, Bruce; Yang, Jae-Suk (2016). Global diversification discount and its discontents: A bit of self-selection makes a world of difference. Strategic Management Journal 37(11):2254–2274. DOI: 10.1002/smj.2574. https://doi.org/10.1002/smj.2574
  8. Eckert, Stefan; Dittfeld, Marcus; Muche, Thomas; Rässler, Susanne (2010). Does multinationality lead to value enhancement? An empirical examination of publicly listed corporations from Germany. International Business Review 19(6):562–574. DOI: 10.1016/j.ibusrev.2010.04.001. https://doi.org/10.1016/j.ibusrev.2010.04.001
  9. Eckert, Stefan; Koppe, Max; Burkatzki, Eckhard; Eichentopf, Simon; Scharf, Constantin (2022). Economies of scale: The rationale behind the multinationality–performance enigma. Management International Review 62(5):681–710. DOI: 10.1007/s11575-022-00473-2. https://doi.org/10.1007/s11575-022-00473-2
  10. Hennart, Jean-François (2011). A theoretical assessment of the empirical literature on the impact of multinationality on performance. Global Strategy Journal 1(1–2):135–151. DOI: 10.1002/gsj.8. https://doi.org/10.1002/gsj.8
  11. Nguyen, Quyen T. K. (2017). Multinationality and performance literature: A critical review and future research agenda. Management International Review 57(3):311–347. DOI: 10.1007/s11575-016-0290-y. https://doi.org/10.1007/s11575-016-0290-y
  12. Li, Jiatao; Ding, Haoyuan; Hu, Yichuan; Wan, Guoguang (2021). Dealing with dynamic endogeneity in international business research. Journal of International Business Studies 52(3):339–362. DOI: 10.1057/s41267-020-00398-8. https://doi.org/10.1057/s41267-020-00398-8
  13. Olibe, Kingsley O.; Rezaee, Zabihollah; Flagg, James; Ott, Richard (2019). Corporate diversification, debt maturity structures and firm value: The role of geographic segment data. The Quarterly Review of Economics and Finance 74:206–219. DOI: 10.1016/j.qref.2019.01.011. https://doi.org/10.1016/j.qref.2019.01.011
  14. Chuwonganant, Chairat; Olibe, Kingsley O.; Strawser, Jeffrey W.; Strawser, William R. (2023). Revisiting the impact of ASC 280: Liquidity, operating performance, and market value. Journal of Corporate Accounting & Finance 34(3):318–336. DOI: 10.1002/jcaf.22627. https://doi.org/10.1002/jcaf.22627