Equity Frictions and Firm Ownership

Firm equity in the Eurozone is far more concentrated in insiders’ hands than the standard modelling assumption allows, and how much so varies systematically across countries. Insiders — defined as the three individuals with the largest shares of the company, identified by tracing multi-level ownership structures of public firms — hold on average 40% of the equity of each public firm across countries and firms. Privately held firms in Europe are typically held by one single shareholder. Combining the two, the paper reports that on average 64% of firm equity is held by insiders. The country spread is the striking part: the share of total firm equity held by insiders ranges from less than 25% in the Netherlands to 90% in Austria, so the complementary “outside equity share” runs from around 10% in Austria to over 75% in the Netherlands. Debt varies too but much less: aggregate leverage, measured as outstanding debt over firm assets net of cash, ranges from 43% in Belgium to 63% in Italy.

Four datasets, combined so that private and public firm ownership can be measured on a common basis. Private firm values and household wealth come from the first wave (2009–10) of the ECB’s Household Finance and Consumption Survey, a survey modelled on the U.S. Survey of Consumer Finances, designed to be comparable across Eurozone countries and oversampling rich households. The paper uses nine of the 15 countries, excluding very small economies (Luxembourg, Cyprus, Malta) and former socialist economies (Slovenia, Slovakia) on the grounds that their recent transition makes them unlikely to be at a steady-state wealth distribution; the nine account for over 93% of Eurozone GDP. Public firm values come from Compustat Global, measured as 2009 market value. Leverage and the firm size distribution come from Amadeus Financials, which covers small and private companies. Shareholder identities come from the Amadeus ownership module, where the average (median) share of total firm equity recorded is 78% (86%). The paper stresses that ownership structures are often multilayered — 36% of total recorded equity is directly held by other firms — so an analysis based on direct ownership alone would overstate concentration.

The paper reports it as suggestive rather than causal. It presents suggestive evidence that the extent to which firms rely on debt and equity correlates with the quality of legal debtor and investor protection respectively, building on the law-and-finance literature pioneered by LaPorta et al. (1997, 1998). The framing throughout is that the estimated frictions capture the strength of creditor and shareholder protection, and an appendix discusses which specific laws and institutions the three parameters might correspond to, along with the potential role of taxes in explaining cross-country differences in ownership and financing. The paper’s own quantitative contribution is described as inferring the level of these frictions from observable firm decisions rather than constructing institutional indices directly.

In countries with more debt and less equity financing, entrepreneurs hold a larger share of aggregate wealth and overall wealth inequality is higher. The paper documents this as a data pattern before modelling it, and reads it as consistent with outside equity being not merely a source of external finance that helps entrepreneurs grow their firms, but also a means of sharing risk: when there is more outside equity, firm financing and risk exposure are more equally distributed, entrepreneurs are less exposed to idiosyncratic business risk, and accumulate less wealth to self-insure. This is the observation the model is built to rationalise.

A Bewley–Imrohoroglu–Huggett–Aiyagari economy augmented with entrepreneurial production and three financing instruments: inside equity, debt, and outside equity. Workers supply labour to entrepreneurs; workers’ skill and entrepreneurs’ productivity are both subject to idiosyncratic risk. An entrepreneur can run a private firm financed with debt and inside equity, or go public and issue outside equity — and if she goes public she also chooses the fraction of firm equity to sell. So the size of the entrepreneurial sector is endogenous, and publicly traded firms have entrepreneurial origins and are run by risk-averse insiders. This last feature is what separates the model from the macro-finance capital-structure literature (Cooley–Quadrini, Hennessy–Whited, Jermann–Quadrini, Begenau–Salomao), where firms are typically run by risk-neutral investors, and from the entrepreneurship literature (Buera–Shin, Midrigan–Xu, Cagetti–DeNardi), where entrepreneurs typically own 100% of their businesses and have access only to debt.

A fixed cost of going public, a proportional monitoring cost that scales with the outside-equity share, and a maximum leverage constraint. The first captures the costs of going public such as underwriting fees. The second captures agency frictions arising because selling equity separates ownership from control — it is modelled as a proportional cost scaling with the share of outside equity. The third is a standard collateral constraint limiting debt issuance. The paper describes the resulting comparative statics: the lower an entrepreneur’s wealth and the higher her productivity, the greater her benefit from external finance and risk sharing, and the more likely she is to sell outside equity; and the lower the frictions in debt markets, the less entrepreneurs rely on equity for external finance. It is this mapping from frictions to the joint distribution of leverage and outside equity that underpins the quantitative strategy.

For France, Germany, Austria and the Netherlands — chosen to span the European range of equity ownership — using the share of equity in private firms, the insider share of public firms, aggregate leverage and the firm size distribution. These moments, together with the model structure, determine the levels of the three financial frictions and the distribution of entrepreneurial productivity. The discount factor is set separately for each country to match its net foreign asset position, so that each economy’s savings position vis-à-vis the rest of the world is reproduced; the paper notes it measures NFA as net foreign assets over GDP plus government debt to GDP, because the model has no supply of public-sector bonds.

Debt frictions are similar; equity frictions differ sharply, and in ways that map onto each country’s ownership pattern. The share of firm assets that can be collateralised ranges from 49% in the Netherlands to 63% in Germany. The fixed cost of going public is estimated at 7% of the median firm value in Germany and close to zero in Austria and the Netherlands. The monitoring cost on outsider-held equity reaches up to 12% in Austria, 5% in Germany, and again close to zero in the Netherlands. The paper walks through the logic country by country. In France — the benchmark, sitting in the middle on both dimensions — firms borrow close to half their capital stock, about 37% of firms by value are privately held, insider ownership is about one-third, and the model implies going-public costs on the order of 4% of firm value, which the paper notes is consistent with Abrahamson et al. (2011), who find European IPO fees average just over 4%. Germany’s firms are more levered (58.3% versus 48.5% in France) and more likely to stay private (59% of firm value privately held versus 33% in France), which the model reads as a higher fixed IPO cost but a slightly lower monitoring cost, alongside a higher leverage constraint that could reflect stronger creditor rights. Austria’s distinguishing feature is that nearly 60% of public equity is held by insiders — the highest in the Eurozone — implying a monitoring cost more than twice France’s, and correspondingly a very low fixed IPO cost, since otherwise no firm would clear the hurdle. The Netherlands is the outlier in the other direction: less than 15% of firm value is privately held, the insider share is 16.1%, leverage is low, and the firm size distribution is less concentrated (top 25% of firms account for 67.1% of the wage bill, against 81.1% in France).

Yes — the paper reports several non-targeted dimensions, and is candid about one where the fit is imperfect. The model reproduces the observed positive relationship between outside equity and firm size: a regression of the insider share on size yields a coefficient of −0.05 in the model against −0.024 in the data. It also reproduces the share of wealth held by entrepreneurs and the differences in top wealth inequality across countries, neither of which was explicitly targeted. On the other hand it over-predicts the insider share for the Netherlands (25.2% in model versus 16.1% in data) even with the monitoring cost set to zero — the paper attributes this to insiders discounting the future more heavily than outside investors, and to entrepreneurs assigning zero weight to firm profits in the state where their own death triggers firm failure. It also notes that the Netherlands is given double the maturity rate of peer countries (π_o = 0.2) to help match its high share of public firms, possibly reflecting faster firm “seasoning” and listing throughput.

Equity frictions dominate on both the level and the dispersion margin. Comparing each calibrated economy to a frictionless benchmark — setting the leverage constraint λ to one for debt, or both equity costs to zero and allowing young firms to issue outside equity for equity — removing equity frictions raises output by an average of 5.3%, against 3% for debt frictions. Country by country the frictionless-equity gains are 5.3% (France), 5.8% (Germany), 8.3% (Austria) and 1.7% (Netherlands), against frictionless-debt gains of 2.7%, 2.1%, 4.7% and 2.7%. On the harmonisation margin, adopting France’s creditor-rights protection changes output by a modest average of 0.5% of GDP, while adopting French equity-market frictions changes it by 1.9% — nearly four times as large. Note the signs differ by country: harmonising Germany’s and Austria’s equity frictions to France raises output (+0.8% and +2.7%), while harmonising the Netherlands’ to the French level lowers it (−2.3%), because Dutch equity markets are the least frictional in the sample.

Because equity does two jobs where debt does one: it supplies funds and it offloads risk, and the risk-sharing channel further raises investment. The paper’s stated mechanism is that increasing access to either form of external finance lets productive entrepreneurs invest more and hire more workers, but when the finance comes as equity, investment and hence output are further boosted because entrepreneurs who share more of the associated risk are willing to invest more in their firms. Put in terms of transition paths: in an economy with frictionless debt markets, entrepreneurs holding large shares of their own firms would still ratchet up investment slowly as they accumulate wealth to self-insure, whereas with frictionless equity markets all firm risk is shared across the population and firms can reach their efficient scale instantly. The paper also flags a scope condition on the magnitudes: the aggregate gains from harmonisation and from frictionless markets may be significantly smaller in closed economies where the interest rate adjusts to clear the domestic capital market, though it argues the finding that equity frictions are larger and more dispersed is likely to survive interest-rate adjustment.

73% of the variation in top-10% wealth shares relative to France, with the correct country ranking — but the level of inequality is over-predicted everywhere. In the data the top 10% hold 52.6% of wealth in France, 59.1% in Germany, 59.5% in Austria and 42.7% in the Netherlands; the model gives 59.0%, 62.7%, 63.9% and 49.8%. Austria and Germany, where outside equity is limited, have the highest wealth concentration; the Netherlands, where equity ownership is widely dispersed, has the most compressed distribution. The paper reads the over-prediction as itself informative: Bewley-type models are known to struggle to generate enough top wealth inequality, and this model combines that literature with the observation that a significant share of firm equity — even for publicly traded firms — is directly held by households, so the wealth distribution inherits the concentration of the firm size distribution.

Financial frictions alone explain 27% of the variation in top-10% wealth shares in the model; together with the estimated firm-productivity distribution the share explained rises to over 73%. The paper flags an important caveat on this decomposition itself: to the extent that financial market frictions influence which projects or firms get started in the first place, this counterfactual can be viewed as an upper bound on the contribution of financial markets to wealth inequality.

No — and this asymmetry is the paper’s distinctive result. Debt frictions carry an equality–efficiency trade-off; equity frictions do not. As predicted by Cagetti and DeNardi (2006), looser debt constraints let entrepreneurs invest more, earn higher profits and save more, so better debt access raises both output and wealth concentration; the paper illustrates that if German creditor rights worsened to the French level, the top-10% wealth share would decline by approximately 1.5 percentage points, alongside a decline in output. Tighter equity frictions also reduce output — if issuing outside equity were as costly in the Netherlands as in France, output would fall by 2.3% — but they increase rather than decrease top wealth inequality. The paper draws the implication explicitly: while lower debt frictions raise both output and inequality, lower equity frictions improve risk sharing and thereby generate higher output alongside lower inequality.

A precautionary-savings channel operating on the entrepreneur, plus a general-equilibrium wage channel operating on workers. Comparing two otherwise identical entrepreneurs facing different fixed IPO costs, the one facing the lower cost sells 40% of his company at t = 0 and can rapidly increase investment, both because he finances only 60% of any given investment and because the IPO proceeds supply additional cash. The private entrepreneur instead only slowly saves out of the borrowing constraint. Initially the public entrepreneur is wealthier, because access to outside equity raises firm value — but eventually the private entrepreneur’s wealth is higher, driven by precautionary savings: after 15 model periods she still runs a smaller firm, is no longer at the borrowing constraint, and is wealthier than the public entrepreneur. The reason she invests less is that her portfolio is much riskier — she owns 100% of the firm and would absorb 100% of the losses on exit, against 60% for the counterfactual entrepreneur — so she saves more, accumulates more wealth, and runs a smaller firm. On top of this, because each entrepreneur owns less capital when equity frictions are tighter, labour demand and the equilibrium wage are lower, workers are poorer and hold less wealth, and there is more inequality among workers as well.

To assess the aggregate and distributional consequences of deviations from Modigliani–Miller using a quantitative model matched to new ownership data, with the explicit modelling of equity frictions as the novel element. Relative to the finance literature, firms here are owned by risk-averse insiders, motivated by the documented importance of insider shares even in publicly traded firms. Relative to the entrepreneurship literature, the size of the entrepreneurial sector is endogenised by modelling the choice to issue outside equity in addition to debt. The paper positions itself as complementary to the law-and-finance literature: rather than constructing indices of corporate-governance rules, it conducts a quantitative evaluation of the effect of investor-protection quality — inferred from observable firm decisions — on aggregate outcomes. The closing framing is that explicitly considering firm ownership decisions and incorporating equity frictions matters for models of entrepreneurship and for macro models that aim to match the household wealth distribution.