توسعه مالی و ارزیابی دارایی ها: حالت خاصی از دارایی واقعی

This paper studies the impact of financial development on asset valuation. We model the agency theoretic perspective of risk-averse investors and financiers in a general equilibrium setting under the framework of rational expectations (i.e., symmetric information). We focus on real estate, as it constitutes a special case of complete market contracting where adverse selection and moral hazard are easily mitigated. Our results illustrate an increase in pareto-efficiency, as financial architecture advances from: (i) banks to capital markets; and (ii) plain vanilla debt to an innovative one with participation clauses. This is attributed to the reduction in agency costs and cross-sectional risk-sharing, leading to an increase in the value of property.Our results predict that an optimal financial system will orient itself towards efficient financial contracts, irrespective of its source of origination. We also rationalize the co-existence of banks and capital markets, and generalize our results under a set of restrictive conditions.

“Mortgage market development is likely to be a key factor in overall financial market development. In particular, an efficient mortgage market will act as a positive externality for the other capital markets, creating pressure for higher efficiency in these markets. On the other hand, a poorly functioning mortgage market is likely to ‘pollute’ other financial markets with its inefficiency.” (Jaffee and Renaud, 1998, p. 75) The ongoing US sub-prime mortgage crisis and the Asian financial crisis of a decade ago illustrate how fragile financial systems devastate a country’s (or a region’s) real estate sector and thus its economy.1 These two crises originated from contrasting financial systems, i.e. a market-based and a bank-based respectively.2 Nonetheless, these crises bring to focus the vital linkage between the financial system of a country and the value of its assets, especially its real estate (see Glaeser, 2000). These crises reignite the debate on the design of efficient financial intermediation to mitigate the vulnerability of the macroeconomy to risk.3,4 This paper aims to explain efficient financial intermediation in the context of the evolution of the financial system, and factors that shape its architecture. Debates on financial system development were initiated by Gerschenkron (1962), who inferred empirically that banks′ prominence in economic development stems from economic backwardness. In recent years, there has been increasing momentum towards a capital market based system, especially in the emerging economies of Latin America and Eastern Europe (see Allen and Gale, 2000). However, it is still unclear as to how financial systems in these countries will evolve and how it will impact their welfare. As Levine (1997, pp. 702–703) points out, “We do not have adequate theories of why different financial structures emerge or why financial structures change. We need models that elucidate the conditions, if any, under which different financial structures are better at mitigating agency costs.” This is precisely the focus of our paper. It is almost an article of faith that the primary reason for the existence of banks is the mitigation of the twin issues of information asymmetry and moral hazard.5 This paper, however, presents a special case of complete market contracting that does not require the financier to have any information-processing or monitoring advantage.6 Financing of real estate constitutes an exceptional situation for two reasons. First, real estate lenders (the principals in a debt contract) can decipher any proprietary (ex-ante) information held by borrowers (the agents in a debt contract) by trading financial claims over a multi-period horizon. This is deduced from the literature on multi-period insurance contracting (see Hosios and Peters, 1989). In the “real world”, lenders also have access to information on ex-post risk and return on various classes of properties to help them underwrite their facilities appropriately. Adverse selection, stemming from ex-ante information asymmetry, is reduced further by releasing funds in the escrow process, when the title of the specific property is exchanged for cash. Second, real estate lenders also reduce moral hazard, stemming from ex-post change in borrower behavior, by mandating the following in the mortgage covenant: (i) minimum maintenance on the property; (ii) payment of taxes; and (iii) adequate insurance coverage. We identify financial liberalization, financial deepening, risk management and financial innovation as the key transmission channels of financial system development, as financing advances from (i) banks to capital markets, and (ii) plain vanilla debt to innovative ones such as participating debt. Therefore, increased capital market sophistication and the presence of non-bank financiers in capital markets diminish bank lending. “Mortgage market development is likely to be a key factor in overall financial market development. In particular, an efficient mortgage market will act as a positive externality for the other capital markets, creating pressure for higher efficiency in these markets. On the other hand, a poorly functioning mortgage market is likely to ‘pollute’ other financial markets with its inefficiency.” (Jaffee and Renaud, 1998, p. 75) The ongoing US sub-prime mortgage crisis and the Asian financial crisis of a decade ago illustrate how fragile financial systems devastate a country’s (or a region’s) real estate sector and thus its economy.1 These two crises originated from contrasting financial systems, i.e. a market-based and a bank-based respectively.2 Nonetheless, these crises bring to focus the vital linkage between the financial system of a country and the value of its assets, especially its real estate (see Glaeser, 2000). These crises reignite the debate on the design of efficient financial intermediation to mitigate the vulnerability of the macroeconomy to risk.3,4 This paper aims to explain efficient financial intermediation in the context of the evolution of the financial system, and factors that shape its architecture. Debates on financial system development were initiated by Gerschenkron (1962), who inferred empirically that banks′ prominence in economic development stems from economic backwardness. In recent years, there has been increasing momentum towards a capital market based system, especially in the emerging economies of Latin America and Eastern Europe (see Allen and Gale, 2000). However, it is still unclear as to how financial systems in these countries will evolve and how it will impact their welfare. As Levine (1997, pp. 702–703) points out, “We do not have adequate theories of why different financial structures emerge or why financial structures change. We need models that elucidate the conditions, if any, under which different financial structures are better at mitigating agency costs.” This is precisely the focus of our paper. It is almost an article of faith that the primary reason for the existence of banks is the mitigation of the twin issues of information asymmetry and moral hazard.5 This paper, however, presents a special case of complete market contracting that does not require the financier to have any information-processing or monitoring advantage.6 Financing of real estate constitutes an exceptional situation for two reasons. First, real estate lenders (the principals in a debt contract) can decipher any proprietary (ex-ante) information held by borrowers (the agents in a debt contract) by trading financial claims over a multi-period horizon. This is deduced from the literature on multi-period insurance contracting (see Hosios and Peters, 1989). In the “real world”, lenders also have access to information on ex-post risk and return on various classes of properties to help them underwrite their facilities appropriately. Adverse selection, stemming from ex-ante information asymmetry, is reduced further by releasing funds in the escrow process, when the title of the specific property is exchanged for cash. Second, real estate lenders also reduce moral hazard, stemming from ex-post change in borrower behavior, by mandating the following in the mortgage covenant: (i) minimum maintenance on the property; (ii) payment of taxes; and (iii) adequate insurance coverage. We identify financial liberalization, financial deepening, risk management and financial innovation as the key transmission channels of financial system development, as financing advances from (i) banks to capital markets, and (ii) plain vanilla debt to innovative ones such as participating debt. Therefore, increased capital market sophistication and the presence of non-bank financiers in capital markets diminish bank lending “Mortgage market development is likely to be a key factor in overall financial market development. In particular, an efficient mortgage market will act as a positive externality for the other capital markets, creating pressure for higher efficiency in these markets. On the other hand, a poorly functioning mortgage market is likely to ‘pollute’ other financial markets with its inefficiency.” (Jaffee and Renaud, 1998, p. 75) The ongoing US sub-prime mortgage crisis and the Asian financial crisis of a decade ago illustrate how fragile financial systems devastate a country’s (or a region’s) real estate sector and thus its economy.1 These two crises originated from contrasting financial systems, i.e. a market-based and a bank-based respectively.2 Nonetheless, these crises bring to focus the vital linkage between the financial system of a country and the value of its assets, especially its real estate (see Glaeser, 2000). These crises reignite the debate on the design of efficient financial intermediation to mitigate the vulnerability of the macroeconomy to risk.3,4 This paper aims to explain efficient financial intermediation in the context of the evolution of the financial system, and factors that shape its architecture. Debates on financial system development were initiated by Gerschenkron (1962), who inferred empirically that banks′ prominence in economic development stems from economic backwardness. In recent years, there has been increasing momentum towards a capital market based system, especially in the emerging economies of Latin America and Eastern Europe (see Allen and Gale, 2000). However, it is still unclear as to how financial systems in these countries will evolve and how it will impact their welfare. As Levine (1997, pp. 702–703) points out, “We do not have adequate theories of why different financial structures emerge or why financial structures change. We need models that elucidate the conditions, if any, under which different financial structures are better at mitigating agency costs.” This is precisely the focus of our paper. It is almost an article of faith that the primary reason for the existence of banks is the mitigation of the twin issues of information asymmetry and moral hazard.5 This paper, however, presents a special case of complete market contracting that does not require the financier to have any information-processing or monitoring advantage.6 Financing of real estate constitutes an exceptional situation for two reasons. First, real estate lenders (the principals in a debt contract) can decipher any proprietary (ex-ante) information held by borrowers (the agents in a debt contract) by trading financial claims over a multi-period horizon. This is deduced from the literature on multi-period insurance contracting (see Hosios and Peters, 1989). In the “real world”, lenders also have access to information on ex-post risk and return on various classes of properties to help them underwrite their facilities appropriately. Adverse selection, stemming from ex-ante information asymmetry, is reduced further by releasing funds in the escrow process, when the title of the specific property is exchanged for cash. Second, real estate lenders also reduce moral hazard, stemming from ex-post change in borrower behavior, by mandating the following in the mortgage covenant: (i) minimum maintenance on the property; (ii) payment of taxes; and (iii) adequate insurance coverage. We identify financial liberalization, financial deepening, risk management and financial innovation as the key transmission channels of financial system development, as financing advances from (i) banks to capital markets, and (ii) plain vanilla debt to innovative ones such as participating debt. Therefore, increased capital market sophistication and the presence of non-bank financiers in capital markets diminish bank lendingAbiad et al. (2008) refer to financial liberalization as a reduction in the role of government and an increase in the role of the market in allocating credit. In the empirical literature, the indicators often used for this are credit controls; interest rate controls; entry barriers for banks; regulations; and restrictions on international financial transactions. The Financial Liberalization Hypothesis, propounded by McKinnon, 1973 and Shaw, 1973, argues that the choice of investments made by banks is affected by government restrictions and therefore impacts on the whole economy. In contrast, a liberalized financial system, with no restrictions on direct ownership of assets, leads to market-determined interest rates resulting in efficient allocation of capital (credit). This implies that a liberalized financial system is in a better position to promote economic growth and development than a repressed one (see Ranciere et al., 2006). In other words, financial liberalization enhances social welfare. We narrow our focus to the government regulation of banks. It is a well known fact that banks in the United States are restricted from taking equity positions in properties (see Allen and Gale, 2000). According to Stulz (2000), allowing banks to hold equity positions in assets has pros as well as cons. A bank that takes an equity position in an asset (along with debt) cares more about overall asset value than one that does not. However, this exposes banks to systemic risks. In other words, it makes them more vulnerable to financial crisis that could devastate the entire economy. We initially consider the case of a partially liberalized commercial bank, which has no restrictions on its loan to value (LTV) ratio, but is prevented from holding equity positions in firms. We then extend our study to the case of universal banks and non-bank financiers, such as pension funds and insurance companies. Abiad et al. (2008) refer to financial deepening as the increase in the volume of credit being intermediated in financial markets. Many studies use the terms ′financial development′ and ′financial deepening′ interchangeably without distinguishing between the two. However, we consider financial development as much broader in scope encapsulating financial liberalisation, financial deepening, risk management, and financial innovation. We call these the four pillars of financial development, as they impact on the efficiency of the financial system. In a liberalized financial system, financial deepening occurs with an increase in funds, allowing a greater volume of investment to take place through capital markets. Economies that have financially deep markets have high capital market liquidity, which increases the intrinsic value of assets traded in it (see Levine and Zervos, 1998). According to Allen and Gale (1997), a well-developed financial system has a comparative advantage in providing cross-sectional risk-sharing, i.e., diversification of risk at a given point in time. This is due to the presence of financiers who are not constrained from taking equity position in firms (dispersed ownership).7 The Allen and Gale (1997) theory thus predicts that as a financial system moves towards the developed stage, risk management will gain greater significance through use of options, futures, and other derivatives. Hence financial innovation plays a very important role in the risk management process (strategy and tactics) of organizations. The theory is thus consistent with the fact that risk management techniques are more important in economies with developed financial systems than in rudimentary financial systems (see Levine, 2002). Pareto-optimal contracting through the use of risk management as well as financial innovations helps improve the efficiency of a financial system by reducing endogenous agency costs of debt (see Merton, 1995). When firms are debt financed, manager-entrepreneurs have an incentive to transfer downside risk (of the project) to the financiers while benefiting from the upside potential. This is a well-known problem of risk-shifting or asset substitution. A number of studies, such as Smith and Warner, 1979 and Barclay and Smith, 1995, have illustrated that risk management through the use of secured debt alleviates this issue.8 Other studies, such as Haugen and Senbet, 1981 and Green, 1984, have argued that participating/convertible debt and other forms of innovations also mitigate this issue by allowing the financiers to share in any windfall that the manager-entrepreneur receives. The second type of agency problem associated with debt financing is referred to as the under-investment problem (see Myers, 1977). Here property-owners are motivated to reject positive NPV investment proposals if the wealth enhancement associated with the property accrues mostly to financiers. Bodie and Taggart (1978) and others have argued that innovative instruments such as participating/convertible debt can be employed to mitigate this issue. An important reason for the interest in financial system development is that it influences real decisions (see Pang and Wu, 2009). We consider how the state of development of the financial system impacts on the borrower’s choice of mortgage financing source and hence on property valuation (see Fig. 2). Unfortunately, the existing literature does not take into consideration the variety of financing choices that an investor typically faces in a financial system. Capital markets provide more financing choices to an investor than do banks. We study the issue of how the scope of financing available to investors (under different stages of financial system development) affects their decision to invest in properties. This is an issue that is extremely important to investors, financiers, and policy makers. As commercial banks are constrained from investing in property, we demonstrate that availability of participating mortgages (explained below) is welfare-improving in a well-developed financial system. Full-size image (30 K)Participating loans encompass a range of financial instruments, secured by the real assets of the underlying project, where the financier receives a “proportion” of the payoffs in the operating or liquidating states (or both). In other words, the investor (i.e., the borrower) trades-off a contingent share in the project against either a reduction in the interest rates or an increase in the loan-to-value ratio. Thus, in the case of Participating Mortgages (PMs), the lender receives either a proportion of gross or net operating income, cash-flows after debt service, or proceeds from the sale of the property (see Ebrahim, 1996).9 Fig. 3 illustrates the various forms of a PM observed in practice. These involve various combinations of participation components such as (i) Shared Appreciation Mortgage (SAM), where the financier subsidizes the interest component in return for a share in the appreciation of the property; (ii) Shared Equity Mortgage (SEM), where the financier is the co-owner, and (iii) Shared Income Mortgage (SIM), where the financier subsidizes the interest component in return for a share in the income from operations.A PM is generally preferable to a convertible mortgage for the following reasons: (i) it is malleable in the three dimensions yielding the variants such as SAM, SEM and SIM as illustrated in Fig. 3; (ii) it allows an investor to retain control of the real estate venture even in the good state of the economy, unlike convertible mortgages where the conversion to equity dilutes these control rights (see Lee et al., 2009). This paper addresses the vital issue of real sector decision by developing a theoretical model in the context of a specialized banking system of United States or United Kingdom, and extends it to that of the universal banking system of Germany or Japan. We investigate the impact of financial system development on property valuation in a general equilibrium setting.10 We model the agency theoretic perspective of risk-averse investors (property owners) and financiers (banks/capital markets) in a framework of rational expectations, i.e., symmetric information, under the assumption of increased liquidity, as we move from banks to capital markets.11,12,13. We implicitly assume the existence of an information architecture as espoused in Levine et al. (2000). That is, an economy where property rights, foreclosure procedures (needed for real estate to serve as collateral), and accurate methods of valuing property are well established. We illustrate that property financing is undertaken in a pecking order of increasing pareto-efficiency, with reduction in the agency costs of debt, in a three-stage process where financial architecture advances from a partially liberalized commercial bank to a developed stage of capital markets. Four key results that contribute to the literature on financial system development and property valuation are derived in this paper. First, since commercial banks in a specialized banking system are constrained from investing in property, they are generally confined to underwriting plain vanilla (risk-free/risky) mortgages. We illustrate that in the rudimentary stage of a commercial bank, property financing is undertaken in a pareto-efficient, plain vanilla, default-free package that collateralizes the debt, thus reducing the risk-shifting issue. This result is in accordance with the prognosis of Smith and Warner, 1979 and Barclay and Smith, 1995. The bank-based equilibrium lies in the lowest rung of pareto-efficiency. In general, defaulting debt equilibrium may not be feasible, as it does not alleviate risk shifting. However, when the defaulting mortgage equilibrium is feasible, it is at best pareto-neutral to default free mortgage equilibrium.14 Thus, valuation of properties under pareto-neutral default-free and defaulting debt constitutes a dilemma for an appraiser, as they are contingent on equilibria. Second, a pareto-improvement of the first solution, commercial bank based equilibrium in a specialized system, is obtained by removing the constraint on ownership of property from financiers. This moves the equilibrium to a more efficient one, as it allows universal banks and non-bank financiers (like pension funds and insurance companies) to diversify cross-sectionally in property markets. This interior solution resolves the real estate version of the asset location puzzle as described by Geltner and Miller (2001).15 However, allowing capital market based financiers like pension funds, insurance companies, etc. (with increased liquidity stemming from financial deepening) leads to a decrease in interest rates and an increase in the price of property. This increase in the price of property can also be perceived as ensuing from an increase in its demand from two competing agents in our economy. Third, we illustrate a further pareto-enhancement of the above equilibrium under financial innovation by embedding the above default-free mortgage with options, in the form of a Participating Mortgage. This makes the financier’s earlier portfolio, of risk-free loan along with fractional purchase of property, redundant, yielding a corner solution, where the investor owns all property in the economy, while the financier owns a quasi-equity claim of the Participating Mortgage. The equilibrium under financial innovation further reduces agency costs embedded in capital market (non-bank) equilibrium by mitigating both the risk-shifting and the under-investment issues, in accordance with the prognosis of Green, 1984 and Haugen and Senbet, 1981. Participating mortgages are also allocatively efficient, as the loan-to-value ratio is higher. The reason for this is that the loan includes the price of the option to share in the cash flows from operations and/or appreciation in property. We also indicate the violation of the well-known Black and Scholes (1973) model when applied to pricing contingent claims on individual property. Fourth, our analysis identifies the crucial link between the value of a property and the pareto-efficient financing package as the intertemporal marginal rates of substitutions (IMRSs) of the property-owner and financier. These adjust to yield the unique mortgage and property pricing parameters, in contrast to the Fisher Separation Theorem (see Hirshleifer, 1958). Our results also differ with the well-known Capital Structure Irrelevance Hypothesis espoused in Modigliani and Miller, 1958 and Hellwig, 1981. We rationalize our theoretical difference with the above studies due to our methodology of segregating the welfare of the agent and principal of the various (default-free/defaulting) mortgage contracts under a framework of risk-aversion (i.e., a non-linear valuation scheme), where value-additivity espoused in the above theories fails to hold, in accordance with the prognosis of Varian (1987). Our results yield implications for financial system development. First, our analysis predicts that an optimal financial system will skew itself towards efficient financial contracts irrespective of its origination (from a specialized banking system or a universal banking system). A financial system in its infancy will be commercial bank dominated, and increased financial development diminishes bank lending. Second, we also rationalize the co-existence of banks and financial markets in a well-developed financial system, irrespective of the type of system. While our model demonstrates that a borrower chooses either bank financing or capital market financing, we could envision financing options lying along a continuum, with plain vanilla defaulting mortgage (by banks) and default-free participating mortgage (by capital market financiers) as the polar extremes. The key determining factors of financing choices are the quality of the underlying collateral and the risk preferences of the agents in the economy. This paper is organized as follows. Section 2 illustrates the theoretical underpinnings of the model for a market-based financial system, while Section 3 evaluates the model solutions and extends it to a bank-based financial system. Due to space limitations, we do not include the Appendix containing all proofs. This is available on request. Finally, Section 4 provides the concluding remarks.

The efficiency of the financial system is of interest to investors, financiers, and policy makers. Many emerging economies have been moving towards a financial system skewed towards capital market financing in recent years without a clear consensus that such systems are welfare-improving. Hence, it is important that growth and development economists address this issue to advise policy makers, as it significantly impacts the country’s economy (see Blejer, 2006). This paper investigates the impact of financial development on real estate valuation in a general equilibrium framework. We utilize the principle of rational expectations and model the agency-theoretic perspective of risk-averse investors (property owners) and financiers (banks/capital markets). In contrast to previous research, we consider a complete market setting where financiers possess no inherent information-processing or monitoring advantages. We demonstrate that property financing is undertaken in a pecking order of increasing pareto-efficiency (with reduction in the agency costs of debt) in a three-stage process, as financial architecture advances from a partially liberalized commercial bank to a developed stage of capital markets. The primary solution is obtained in the rudimentary stage of commercial banks (in a specialized banking system), where risk-management in the form of secured debt alleviates risk-shifting. This is in accordance with the prognosis of Smith and Warner, 1979 and Barclay and Smith, 1995. It also leads to the pareto-optimality of default-free mortgages over defaulting ones. A pareto-improvement of the first solution is obtained by removing the constraint on asset ownership by universal banks, pension funds, insurance companies, etc. This solution is due to cross-sectional risk-sharing and resolves the real estate version of the asset location puzzle (see Geltner and Miller, 2001 and Dammon et al., 2004). A further pareto-enhancement of this equilibrium is obtained under financial innovation by embedding the above default-free mortgage with options, in the form of a participating mortgage, to mitigate risk-shifting and under-investment issues. This is in accordance with the prognosis of Green, 1984 and Haugen and Senbet, 1981. Our analysis reveals that the critical link between the value of a property and the pareto-efficient financing packages is the intertemporal marginal rates of substitutions (IMRSs) of the borrower and lender. These IMRSs adjust in such a way as to yield a unique equilibrium interest rate, participation rate, loan amount, and property value, in contrast to the Fisher Separation Theorem (see Hirshleifer 1958). Here, our results differ from the well-known Capital-Structure Irrelevant Hypothesis (see Modigliani and Miller, 1958). Our distinct interior solutions stem from the fact that, under a framework of risk-aversion (i.e., a non-linear valuation scheme), value additivity assumed in the Irrelevant Hypothesis does not hold, as agents adjust their intertemporal marginal rate of substitution to own a property and to trade default-free mortgage claims against it with the lender (see Varian, 1987). Further analysis also reveals that, in general, defaulting mortgage equilibrium may not be feasible, as it does not alleviate risk-shifting. However, when the defaulting mortgage equilibrium is feasible, it is at best pareto-neutral to default-free mortgage equilibrium. Thus valuation of properties under pareto-neutral (default-free and defaulting) mortgages constitutes a dilemma for an appraiser, as these are contingent on equilibria. Our results produce implications for financial system architecture and development. Our analysis predicts that the optimal financial system will align itself towards efficient financial contracts, irrespective of its source of origination. A financial system in its infancy will be bank-dominated, and increased financial development diminishes bank lending. Greater development in a financial system is manifested through the following: (i) Financial Liberalization: In a liberalized financial system, there are no government imposed restrictions on LTV ratio, and financiers are not prohibited from direct ownership of property. But in reality, restrictions on banking activities lead to the development of capital markets. Hence, we demonstrate how regulatory constraints, aimed principally at banking scope, affect the evolution of a financial system. Our results reinforce the empirical findings of Cho, 1988, Chari and Henry, 2008 and Ghosh et al., 2008. (ii) Financial Deepening (increase in the volume of credit being intermediated in a financial market): Our model demonstrates that stock market liquidity enhances the supply of funds, which is a parameter of financial deepening. This improves the loan-to-value ratio of the mortgages and subsequently the allocative efficiency of capital welfare of agents in the economy. Our results support the empirical findings of King and Levine, 1993 and Levine et al., 2000. (iii) Risk Management and Financial Innovation: In well-developed financial systems, innovations in security design provide a richer set of risk management tools (see Levine, 2002). These help to reduce macroeconomic volatility and thus diminish asset price volatility, increasing the resilience of the economy to shocks. Here, our results differ from those of Modigliani and Miller, 1958 and Hellwig, 1981, who argue that optimal security design and financial innovation are irrelevant. Thus, the welfare relevance of capital markets should grow through time as a financial system develops. Our analysis is in conformity with those of Weinstein and Yafeh, 1998 and Allen and Gale, 2000. However, it differs to those of Rajan and Zingales, 2001 and Tadesse, 2002, who provide a strong case for bank financing. We also rationalize the co-existence of banks and financial markets in a well-developed financial system. Whilst in our model, the borrower either chooses bank financing or capital market financing, we could envision financing options lying along a continuum, ranging from plain vanilla defaulting mortgage (by banks) to default-free participating mortgage (by capital market financiers). The key determining factors of financing choices are the quality of assets a borrower can pose as collateral and the risk profile of agents in the economy. This general result of contingency of pareto-improvement of equilibria on the quality of collateral is consistent with Shleifer and Vishny (1992). Thus, for low-quality collateral (such as mobile homes), where the terminal payoffs are zero for some states of the economy, risky mortgage financing with features of preferred stock (if feasible – under Lemma 2) may be the only option available, as explicated in our corollary. Furthermore, agents with low risk aversion may be indifferent to bank and capital market financing, as demonstrated in the trivial solutions explicated in the Appendix (available upon request from the corresponding author). Although our measures of financial liberalization, financial deepening, and financial innovation are not perfect, the results do provide clear insights with sensible policy implications. Improving the functioning of financial markets is critical for boosting long-term economic growth. Thus, policy makers should focus on strengthening the overall efficiency of the financial markets. However, as a future policy measure, it is important that the banks and capital markets are integrated. This should be supported by a sound financial sector and appropriate macroeconomic policies, in order to maximize the benefits and minimize the risks associated with financial liberalization. In short, the bottom line is to strengthen the architecture of the financial system to reduce the frequency and impact of financial crises.