Firms differ along many dimensions. There are many small and few large ones. Some firms are very R&D-intensive, while others are not. Small firms serve local markets, while large multinationals operate globally with affiliates in many countries. The ownership of entrepreneurial firms is concentrated, where the interests of managers and owners are largely congruent. Large firms are typically characterised by dispersed ownership, are run by professional managers, and are subject to oversight and control by a board. Given these differences, one can infer that different firms respond in different ways to taxes.
Firm heterogeneity is important to policymakers since business taxation is more harmful to some firms than to others. For this reason, governments might want to offer targeted tax relief to some types of firms that would otherwise find it particularly costly. For example, most countries provide R&D tax incentives to stimulate innovation and growth, which is purposely designed to favour genuinely innovative firms in order to encourage effective R&D. Similarly, protecting the national tax base in the presence of intensifying international tax competition calls for relatively lower taxes on mobile firms, which are often more R&D-intensive as well, provided that location decisions or profit-shifting are indeed responsive to tax.
Financial constraints tend to be concentrated among young, innovative firms where large investment and financing needs coincide with little own funds. These firms tend to earn a higher return on capital since their difficulty in accessing external funds leaves them with profitable but unexploited investment opportunities. Large, cash-rich firms obviously face no such difficulty and may even suffer from overinvestment. Rather than treating all firms uniformly, tax policy should be concerned with supporting capital reallocation from less profitable to more profitable firms, by encouraging dividend distribution in large firms and giving selective tax relief to small-size, high-growth companies. In all these cases, a targeted approach would by definition treat some firms different from others.
A first step is to find out which types of firms are most sensitive to business taxes. Recent research has already uncovered important dimensions of the heterogeneous tax sensitivity of firms.
- For example, Chetty and Saez (2005) found that the 2003 US dividend tax cut produced a very heterogeneous response in dividend payments, depending on the availability of cash reserves, differences in ownership structures, and managerial incentives in corporate firms.
- Zwick and Mahon (2017) have investigated the effects of bonus depreciation in the US and found that small firms respond much more strongly than big firms do. Importantly, they emphasised that differences in financial constraints determine the magnitude of the investment response to tax changes.
- Yagan (2015) investigated the response of C- versus S-corporations in the US to the 2003 dividend-tax cut, where C-corporations are subject to dividend taxation and S-corporations are not. He found no effect of the dividend-tax cut on investment of C-corporations relative to unaffected S-corporations. This ‘non-result’ on the effect of the dividend tax on investment could emerge, for example, if marginal investments are funded out of retained earnings and riskless debt as in ‘new-view’ models of dividend taxation. Such an argument typically applies to cash-rich firms, which have large internal funds to finance investment.
To tie in to this, in a recent paper (Egger et al. 2018) we develop an analytical framework that systematically relates a firm’s investment to its organisational and financial regime. The theoretical analysis explains that entrepreneurial and managerial firms operate under very different financial constraints and, even within each group, differ by the availability of their own funds relative to investment needs. Entrepreneurial firms may be short of their own funds so that investment is ‘cash-constrained’, while others are effectively unconstrained, or lightly constrained for tax reasons only. The tax bias in favour of debt leads firms to raise external debt until they exhaust their debt capacity and become ‘tax-constrained’. We estimate the unobserved threshold level of own funds relative to total assets that separates the two groups of entrepreneurial firms. We classify firms as entrepreneurial (61% of all firms in our sample) if the largest owner holds a share larger than 50%. Empirical results do not change if we set the ownership threshold at 25%.
Managerial firms are run by professional managers and are governed by a board. They operate under entirely different incentive constraints but are also heterogeneous among themselves. We subdivide them into three groups. First, we classify cash-rich firms as ‘dividend paying’ and unconstrained. This group corresponds to new-view models, which imply that dividend taxes have no effect on investment. The remaining two parts are cash-poor and pay no dividends, at least no dividends beyond an absolute minimum. To take account of further heterogeneity, we divide those firms into ‘cash-constrained’ and ‘severely constrained’ firms. Again, the two threshold levels of own funds separating these three groups are unobserved and estimated.
Figure 1 Financial regime and firm-level investment
Our econometric results derive from a panel of 35,092 firms from 17 different European countries for the years 2004-2012 (see Table 3 of Egger et al. 2018). Figure 1 illustrates that, empirically, investment always depends on the presence of own funds, although to varying degrees. Our theoretical analysis implies that even cash-rich firms, which should have no problem in financing investment and raising external debt, end up being lightly constrained for tax reasons. For ‘tax-constrained’ entrepreneurial firms, an increase in the cash-ratio by 1% would raise investment (as a share of total assets) by 0.117%. In contrast, the impact of a higher cash-ratio on constrained entrepreneurial firms is more than double. The pattern is much the same across managerial firms when comparing dividend paying and non-dividend paying, constrained firms. However, we have also identified a severely constrained group, which not only pays no dividends, but also has exhausted its debt capacity to raise additional external debt. An increase of the cash-ratio by 1% would boost the investment rate by 1.3%, which is more than ten times the effect on dividend-paying managerial firms! Clearly, financial heterogeneity influences the growth potential of different firms to a great extent.
The estimates in Figure 2 illustrate the heterogeneous tax sensitivity of firm-level investments. The first, light-shaded bar in each group refers to corporate taxation, the second, dark-shaded bar to dividend taxation. Corporate taxes affect investment negatively across all firm-types. The effect is larger for entrepreneurial firms where the estimated coefficients are between 40% and 70% higher than for managerial firms. Even within the set of entrepreneurial firms, the differences across firm types are substantial. An increase in the effective marginal corporate tax of 1% leads to a decline in investment by 0.84% for cash-constrained entrepreneurial firms and by 0.68% for tax-constrained firms. Investment of managerial firms decreases by roughly 0.5% assuming the same policy shock. Given an average size of investment of 15-17% (as measured by the relative change of fixed assets between the beginning and the end of the accounting year), these effects are economically large. An increase in corporate taxation by 10% would cut the investment of cash-constrained entrepreneurial firms in half.
Figure 2 Heterogeneous tax effects on firm-level investment
Dividend taxation only has significantly negative effects on the investment of cash-constrained entrepreneurial firms and non-dividend paying managerial firms. This result is perfectly in line with theory. Cash-constrained firms have to reduce dividends to a minimum to relax the financing constraint and scale up investment. When owners demand a given minimum level of dividends net of tax, firms must compensate for a higher dividend tax and pay a larger gross dividend, which drains internal funds and restricts investment. Our econometric estimates suggest that a 1% increase in dividend taxation reduces investment by around 0.11-0.17% for cash-constrained entrepreneurial and managerial firms, respectively. Cash-rich firms are unconstrained and pay dividends in excess of the minimum. When the dividend tax rises, they simply pay out a lower net of tax dividend, so that retained earnings and investment remain unaffected, as the new view of dividend taxation would predicts.
To sum up, our analysis finds considerable heterogeneity in the tax sensitivity of firm-level investment. For example, the impact of corporate taxation is up to 70% higher for entrepreneurial firms than for managerial firms. Dividend taxation negatively affects investment of financially constrained firms but entails no significant impact on cash-rich firms. These differences are tightly linked to the existence of financial constraints affecting specific firm types. Policy might provide targeted tax relief to the most constrained firms, if other policies are not successful or impractical in solving the problems with access to external capital.
Chetty, R, and E Saez (2005), “Dividend Taxes and Corporate Behaviour: Evidence from the 2003 Dividend Tax Cut”, Quarterly Journal of Economics 120, 791-833.
Egger, P, K Erhardt and C Keuschnigg (2018), “Heterogeneous Tax Sensitivity and Firm-Level Investments”, CEPR Discussion Paper no. 13341.
Yagan, D (2015), “Capital Tax Reform and the Real Economy: The Effects of the 2003 Dividend Tax Cut”, American Economic Review 105, 3531-3563.
Zwick, E, and J Mahon (2017), “Tax Policy and Heterogeneous Investment Behavior”, American Economic Review 107, 217-248.