Investments play a crucial role in economy and finance. Investments in tangible and intangible assets are a necessary condition for structural adjustment and economic growth. They provide the enhancement of existing basic funds and industries and the creation of new ones. The role of investment is increased many times at the current stage. In this respect, the role of the evaluation of the efficiency of investment projects, which allows for the realization of the most effective projects in the context of scarcity and limited investment resources, increases. Since virtually most investment projects use debt financing, the study of the influence of capital structure and debt financing on the efficiency of investment projects and determining the optimal capital structure is especially important at the present time. This is why in spite of the fact that a lot of different types of investment models have been developed—stochastic, dynamics [
1], investment banking valuation models [
2], etc.—the main problem, which has been discussed during the last few decades, is the impact of debt financing on the efficiency of investment projects and on the investment decisions of companies.
Below we review some approaches to these problems.
1.1. The Literature Review
Lang, L.E.; Ofek, E.; and Stulz, R. [
3] used so-called “Tobin’s q ratio” and considered companies with low Tobin’s q ratios as well as high ones. Note that the q ratio, or Tobin’s q ratio, equals the market value of a company divided by its assets’ replacement cost. The equilibrium takes place when market value equals replacement cost. The q ratio expresses the relationship between market valuation and intrinsic value. It is a means of estimating the fact of whether a given business or market is undervalued or overvalued.
The authors have shown that there is a negative correlation between future growth and leverage level at the company level and, for diversified companies, at the level of business segments. This negative correlation between growth and leverage level holds for companies with low Tobin’s q ratio, but not for high-q companies or companies in high-q industries. Therefore, for companies known to have good investment opportunities, leverage does not reduce growth, but it is negatively correlated to growth for companies whose growth opportunities are either not recognized by the capital markets or are not valuable enough to overcome the effects of their debt load.
Whited [
4] studied the influence of debt financing on companies’ investment decisions with pharmaceutical firms in India for 11 years, from 1998 to 2009.
To study the impact of debt financing on the firms’ investment decisions, Whited used pooling regression as well as random and fixed effect models. Leverage level, retained earnings, Tobin’s q, sales, Return on Asset, cash flow and liquidity were considered as independent variables and investment as the dependent one. Whited considered three types of companies, depending on their size: small companies, medium companies and large companies. He showed that there is a significant positive correlation between leverage level and investment for large companies, while for medium companies a negative correlation between leverage level and investment took place.
Kang [
5] studied the connection between leverage level and investment decisions. “Interdependent tax models” were used to try to explain the specifics of company leverage levels by analyzing the interdependency between financing decisions and investment. These models account for the so-called “investment effect”: the influence of investment on debt tax benefit and financial risk. One of the questions is how “investment effect” influences bond financing decisions and hence the leverage level. Different “Interdependent tax models” lead to different connections between investments and leverage levels.
Some authors mentioned a positive connection via the fact that the financial risk and hence the cost of bond financing decrease with an increase in investment at a given leverage level. A negative connection has been mentioned by DeAngelo and Masulis, 1980 [
6] and Dotan and Ravid, 1985 [
7]. The first authors concluded this since the tax benefits of debt compete with those of capital investment. The second authors refer to the fact that financial risk and thus the cost of bond financing will increase with investment increase.
The impact of investment increase on financial risk may depend on company-specific factors, like company-specific technology (Dammon and Senbet, 1988) [
8]. An analysis of the impact of corporate and personal taxes on a firm’s optimal investment and financing decisions under uncertainty is provided in this paper. By endogenizing firms’ investment decisions, it extends the DeAngelo and Masulis capital structure model. The authors’ results indicate that the existing predictions about the relationship between investment-related and debt-related tax shields must be modified in cases where investment is allowed to adjust optimally. The authors show that increases in investment-related tax shields due to changes in the corporate tax code are not necessarily associated with reductions in leverage level at the individual company level. Companies with higher investment-related tax shields (as cross-sectional analysis shows) need not have lower debt-related tax shields (normalized by expected earnings) unless all companies utilize the same production technology. Differences in production technologies across companies may thus explain why the empirical results of recent cross-sectional studies have not conformed to the predictions of DeAngelo and Masulis [
6]. A model of company leverage level choice is formulated in this paper, in which corporate and differential personal taxes exist and supply-side adjustments by companies enter into the determination of equilibrium prices of debt and equity. The presence of corporate tax shield substitutes for debt such as depletion allowances, accounting depreciation and investment tax credits are shown to imply a market equilibrium in which each company has a unique interior optimum leverage level decision. The optimal leverage level model yields a number of interesting predictions regarding cross-sectional and time-series properties of firms’ capital structures. Below we discuss some portfolio investment models as well as behavioral aspects of investors, which play an important role in investments.
Among those portfolio investment models is the well-known Black–Litterman model, which was created by Fischer Black and Robert Litterman in 1992 [
9]. They developed the model to address the problems that institutional investors have encountered the during application of modern portfolio theory in practice. Starting with an asset allocation based on the equilibrium assumption, the model then modifies that allocation by accounting the investors’ opinions with respect to future asset performance.
Anthony Loviscek [
10] has applied the Black–Litterman model of modern portfolio theory to well-known index mutual funds?one guided by the classic 60%/40% stock/bond allocation and one based on an all-equity allocation. The period under their study is from 2000 to 2020. Although statistical evidence supports that the efficacy of a precious metal allocation is elusive, the results suggest an average allocation of about 2% for “buy-and-hold” investors who seek one. He has shown that, from 2003 to 2010 and from 2016 to 3Q2020, the allocations were in the range of 5–10%. Other periods, however, register only a little more than 0%.
Nusret Cakici and Adam Zaremba [
11], using data on 65,000 stocks from 23 countries, reconsidered the performance of the Fama–French factors in global markets. As their results showed, the value, profitability and investment factors are far less reliable than what is commonly thought. Their performance depends strongly on the geographical regions and periods examined. Moreover, most factor returns are driven by the smallest companies. Virtually no value or investment effects are present among the big companies representing most of the total market capitalization worldwide. These results cast doubt on the five-factor model’s applicability in international markets, citing that the smallest companies are typically not invested in by major financial institutions.
A growing number of investors want to use firm sustainability information in their investment decision processes to avoid risk, satisfy their own asset preference, or find a new alpha-generating factor. Not too many users of environment, social and governance (ESG) data understand how ESG ratings change over time. Bahar Gidwani [
12] used the CSRHub data set to show that ESG ratings regress strongly toward the mean. These ratings include both data from 640 sources and from most commercial ESG ratings firms. The observed regression persists during nine years within the ratings data, for a sample set of more than 8000 firms. Newly-rated firms show even more reversion than “seasoned” firms. Firms can only rarely maintain an especially high or low ESG rating. Investors and firm managers should understand that ESG ratings are likely to change toward the mean. This however does not mean that a good firm is getting worse or a bad one is getting better.
Keith C. Brown, W. V. Harlow and Hanjiang Zhang [
13] have developed statistics (holdings-based) to estimate the volatility with time of investment style characteristics of funds. They found that funds with lower levels of style volatility significantly outperform funds with higher levels of style volatility on a risk-adjusted basis. The authors have shown that style volatility has a distinct impact on fund performance in the future compared to expenses of funds or past risk-adjusted returns, with the level of indirect style volatility being the primary determinant of the overall effect. It was concluded that deciding to maintain a less volatile investment style is an important aspect of the portfolio management process.
Some behavioral aspects of investors have been considered by Marcos Escobar-Anel, Andreas Lichtenstern and Rudi Zagst [
14], who introduced a strategy generalizing the CPPI (Constant Proportion Portfolio Insurance) approach. The target of this strategy is to guarantee the investment goal or floor during participation in the performance of the assets and limiting the downside risk of the portfolio at the same time. The authors show that the strategy accounts for the following behavioral aspects of investors: a risk-averse behavior for gains, distorted probabilities recognition and a risk-seeking behavior for losses. The developed strategy turns out to be optimal within the Cumulative Prospect Theory framework by Tversky and Kahneman [
15].
1.2. Some Problems under the Evaluation of the Effectiveness of the Investment Projects
Some of the major problems under the evaluation of the effectiveness of investment projects are suggested as follows:
Which financial flows should be taken into account when calculating the parameters of efficiency of a project (NPV, IRR, etc.)?
How many discount rates should be used for discounting various cashflows?
How can these discount rates be accurately evaluated?
The first two problems are still under intensive discussion. Concerning the third issue, we need to note that, in the last decade, significant progress in the accurate determination of the cost of the equity and company weighted average cost, which just are the discount rates when evaluating the effectiveness of the project, has been achieved. The progress is mainly associated with the studies by Brusov, Filatova and Orekhova (BFO theory) [
16,
17,
18], in which a general theory of capital cost of the company and its capital structure was established, and the dependence of capital cost on leverage level and on the age of a company was found for the companies of arbitrary age. The main difference between their theory and Modigliani-Miller theory is that the former one removes the assumption of perpetuity for the companies under discussion, which leads to a significantly different new theory from the theory established by the Nobel laureates Modigliani and Miller [
19,
20,
21].
In modern conditions, the requirements for improving the quality of assessing the effectiveness of investments have increased. The modern investment models, which have been well-tested in real economic situations, have been developed by Brusov, Filatova and Orekhova [
16,
18]. They have created long-term as well as arbitrary duration models and have considered the effectiveness of the investment project from two points of view: from the equity holders and from the owners of equity and debt. NPV in each of these cases could be calculated by two different methods: with the division of credit and investment flows and using two different discounting rates, and without such a division and using a general discounting rate (for which WACC can, obviously, be chosen).
Applying their modern investment models on the evaluation of the dependence of the effectiveness of investments on debt financing of one telecommunication company in 2010–2012 from the point of view of optimal structure of investment, the authors showed that in 2012, the company lost 675 million USD on average, because the investment structure had been far from the optimal one. The ability to calculate the correct optimal capital structure is one important feature implied by the Brusov, Filatova and Orekhova modern investment models [
16,
17,
18].
1.3. The Discount Rates
As we mentioned above, one of the most important elements of calculating the effectiveness of investment projects is the assessment of the discount rate. In the case of long-term investment models without the division of credit and investment flows, the discount rate WACC has been calculated using the Modigliani-Miller formula [
19,
20,
21]
while in the case of arbitrary duration models, the Brusov–Filatova–Orekhova formula for WACC [
9,
10,
11]
has been used.
Here and below, WACC is the weighted average cost of capital; k0 is the equity cost at zero leverage (L = 0); kd is the debt cost; wd is the debt share; t is the tax on profit; n is the project duration; ke is the equity cost; and L is the leverage level.
In the case of long-term investment models with the division of credit and investment flows, the discount rate for discounting the investment flows (equity cost
ke) has been calculated using the Modigliani-Miller formula [
19,
20,
21]
while in the case of arbitrary duration models, equity cost
ke has been calculated from the formula
using the Brusov−Filatova−Orekhova value for WACC [
16,
17,
18].
The calculation methods of the discount rates (WACC, equity cost
ke) has been generalized in [
22] for the real conditions of the implementation of investment projects: for arbitrary frequency of payment of tax on profit.
In this paper, new modern investment models, both long-term and arbitrary duration, will be created, as close as possible to real investment conditions. They will account the payments of interest on debt and of tax on income a few times per year (semi-annually, quarterly, monthly), which are applied in real economic practice. Their verification will lead to the creation of a comprehensive system of adequate and correct assessment of the effectiveness of the company’s investment program and its investment strategy.