LITERATURE REVIEW AND HYPOTHESES DEVELOPMENT

Financing of R&D Expenditures

Internal Financing (Cash Flows)

In the absence of market imperfections, the level of internal capital (cash flows) should not influence investment decisions, as firms are able to obtain the necessary funds from the capital markets at a cost similar to that of the cost of internal capital to finance positive NPV projects.

With respect to investment in R&D projects, the problem of asymmetric information between insiders and outside investors may be worse be­cause of the potentially high sunk costs and high future returns this type of projects may generate, firms may be reluctant to reveal information to outside investors. Moreover, firms need to protect the knowledge related to these projects from their competitors.

The idea that internal capital and investment expenditures should be correlated positively has been recently contested by Cleary et al. (2007) who show both theoretically and empirically that the relationship between internal funds and investment is actually a U-shaped. Three key assumptions drive this result: external capital is costly, this cost is determined endogenously (i.e. higher cost for higher amounts of debt borrowed), and investment is scalable. The bottom-line of the model is the hypothesis that firms with large financial gaps find it feasible to finance large rather than small investments. We argue that U-shaped curve is even more relevant for R&D expenditures since the risk associated with R&D is higher and, in turn, the returns to R&D range from negative to a huge payoff. Assuming that internal funds are not sufficient to finance all investment’s requirements, and then decreasing in cash flow will cause the firm to decrease its investments to avoid high new debt financing and high risk. R&D investment will probably be first investment item to be cut in such a situation due to its high-risk characteristic. On the other hand, at low or negative levels of internal cash flows investors are likely to be repaid a lesser amount since the firm may be unable to pay all of the promised payment.

H1: At high levels of internal funds R&D ex­penditures increase monotonically; at low levels of internal funds R&D expenditures decrease monotonically.

External Financing: Debt vs. Equity

The second aspect of the relationship between financing and R&D is whether firms use debt or equity as an external source of financing for their R&D expenditures. Core of this aspect is the company’s capital structure. The literature for choice of capital as the mix of debt and eq­uity starts with Modigliani and Miller (1958 and 1963) who examine the financing decision of a firm separately from its investment decision. There have been several other theories focusing on debt equity choice. The tradeoff theory arrives at an optimal capital structure by balancing the tax advantages with the agency and bankruptcy costs. Some others explain that leverage depends on market frictions, such as informational sym­metries between insiders (managers) and outsiders (investors or lenders). These asymmetries cause some firms not reach to sufficient capital to fund all their good projects (financial constraints) due to agency costs (Jensen & Meckling, 1976; Jensen, 1986), signaling costs (Stiglitz & Weiss, 1981), and source of capital as a market friction recently examined by Faulkender and Petersen (2006), who find that firm with credit ratings as an indication of ability to access to debt market use significantly more debt.

The literature has examined the relationship between debt financing and R&D. Studies gen­erally argue that debt financing is a sub-optimal alternative for R&D because in most cases R&D expenditures convert into firm-specific assets, which have a low liquidation value in the event of bankruptcy.

H2: R&D expenditures are positively associated with higher levels of equity financing.

The Role of Market Timing

The timing of issuing new shares may also matter for financing investment proj ects along with equity financing. In particular, companies may issue new shares when the market price of their shares is viewed as abnormally high, because this reduces the costs of equity. Graham and Harvey (2001) conduct a large-scale survey among American and Canadian executives. The results show that these executives confirm the importance of mispricing in decisions regarding new equity issuances. This behavior is more likely to occur if firms believe that the high stock price is a temporary phenom­enon, which means they should take action as soon as possible. For example, Kim and Weisbach (2008), using an extensive dataset of 17,000 IPOs and nearly 13,000 seasoned equity offerings from companies in 38 countries during the period 1990-2003, find evidence that firms with high market-to-book ratios keep more of their equity offerings in cash than firms with low market-to- book ratios do.

A number of papers find empirical support for the fact that companies take advantage of mispricing opportunities on their investment decisions. Barro (1990) using US and Canadian data, and Galeotti and Schiantarelli (1994), based on US firm-level data, show that lagged values of real stock market prices significantly influence investments even after controlling for Q and other cash flow variables. Baker et al. (2003) using a large sample of US firms covering the period 1980-1999, find evidence that investments of equity-dependent firms (i.e. firms that need equity capital to finance marginal investments), are more sensitive to stock prices than they are for non-equity dependent firms. Campello and Graham (2007), based on a sample of 1,943 US firms covering the 1971-2003 period, show that financially constrained so-called “non-tech” firms issue equity due to mispricing and subsequently some of it to invest, while the rest is saved in their cash accounts. This observation turns out not to be true for unconstrained “non-tech” or “tech” firms.

None of the above studies have explicitly looked at the extent to which companies take advantage of mispricing opportunities to finance R&D expenditures. We contribute to the literature by looking at this issue in this chapter by testing the following hypothesis.

H3: R&D expenditures are positively associated with higher past valuation of equity.

Financing of R&D and Firm Characteristics

Some of the above-discussed papers point out that financing patterns of investment may differ due to firm-specific characteristics. One important firm characteristic that has been discussed in some of these papers is whether or not firms are financially constrained, i.e. whether or not they are unable to finance their investment projects from internal capital. As has been discussed in Campello and Graham (2007), financial constraints are important with respect to firms’ need for external financing and/or the past valuation of their shares.

Another potentially important firm characteris­tic that may influence financing patterns of R&D expenditures is the extent to which firms make such expenditures. This may be particularly relevant for explaining debt versus equity financing of these expenditures. At low levels of these expenditures, the expectation would be that firms do not change their financing mix for small increases in R&D. On the other hand, the situation can be different for firms that expend considerable resources for R&D expenditures. This argument is consistent with Myers’s (1984) relaxation of strict financing order in the case where firms have an abundance of future investment opportunities. At some point, additional resources devoted to finance R&D will be viewed as additional risk. Then the main ques­tion would be whether the firm makes up enough revenues or has enough cash flow to pay for the costs of additional resources. Lenders may view the additional expenditures as risky as well and either limit additional funds and/or charge more for these funds. Moreover, R&D expenditures convert into firm-specific assets, which have a low liquidation value in the event of bankruptcy. The prudent course of action may therefore be to use equity financing and, thus, avoid additional required payments. Therefore, in our analysis of financing patterns of R&D expenditures we take into account the firm-level size of these expenditures. Thus, with respect to the role of firm-specific characteristics and how they help explaining financing patterns of R&D expendi­tures, we test the following hypotheses:

H4: The positive associations between R&D expenditures and equity financing, and past valuation of equity are stronger for finan­cially constrained firms.

H5: The positive associations between R&D expenditures and equity financing, and past valuation of equity are stronger for firms with high levels of these expenditures.

Financing of R&D and the National Context

Differences in financing patterns of R&D expen­ditures of companies may also occur because of differences in the national context. In particular, we focus on two specific elements of this context: the nature of national financial markets and patent rights protection.

In the literature, financial markets around the world have been characterized as market-based or bank-based. In market-based financial markets, stock markets play an important role in channel­ing sources to investment, whereas in bank-based markets, banks are the most important channel for financial intermediation (Demirguc-Kunt & Levine, 2001; Demirguc-Kunt & Maksimovic, 2002; Levine, 2002). The type of financial markets firms have to deal with may have consequences for the financing patterns of investment expenditures. In particular, in the context of R&D investments, the hypothesis that firms use more equity as op­posed to debt financing for this type of project may be more valid in a market-based financial market than in a bank-based financial market. First, in bank-based financial markets, strong relation­ships between banks and firms reduce the agency problems of debt and asymmetric information. This means that in bank-based financial markets debt may be used relatively more for financing of R&D expenditures than in market-based financial markets. Second, in market-based financial mar­kets firms have better access to equity, increasing the use of this type of external finance vis-a-vis debt finance.

The other important country specific factor is patent rights protection. Firms should conduct more R&D if they feel there is less chance that their R&D work will be stolen or imitated. The literature documents a strong relationship between patent rights protection scheme and R&D invest­ment. Varsakelis (2001) finds that countries with strong patent protection invest more in R&D. Therefore, the analysis in this chapter controls the effect of patent rights. Ginarte and Park (1997) first constructed an index of patent rights protection for 1960-1990 for 110 countries, and then Park (2008) updated the index to 2005 and extended it to cover 122 countries.

The other possibility is to have a nonlinear effect of patent rights on R&D. Allred and Park (2007) find that there is a significant and negative effect on firm-level R&D for low level of pat­ent rights and the positive and significant effect after some threshold level. While this evidence is relevant for developed countries, Allred and Park do not detect a significant effect of patent rights in developing countries. Summarizing the above discussion with respect to the role of the national context and how this helps explaining financing patterns of R&D expenditures, we test the following hypotheses:

H6: The positive association between R&D expen­ditures and higher levels of equity financing is stronger for firms in market-based econo­mies as compared to bank-based economies.

H7: There is a nonlinear relationship between R&D expenditures and patent rights espe­cially for firms in market-based economies as compared to bank-based economies.