آیا یک مدل اسلامی امور مالی تعاونی مسکن می تواند وضعیت اقتصادی افراد محروم را بالا ببرد؟

A formal home loan is onerous to subprime borrowers in efficient markets. This can deter homeownership for financially strapped individuals, leading to a market failure. This paper proposes a special form of cooperative mortgage financing (practiced in Oman) to overcome this market failure. We integrate the literature of mortgage design with that of informal savings schemes (i.e., ROSCAs/ASCRAs) to illustrate that this mode of financing dissipates credit risk better than the formal mode of financing. It is also resilient to volatility of interest rates and allows prepayments without any additional charges. Finally, we verify the assertions of Besley et al. (1994) and Hart and Moore (1998) that cooperative mortgages are pareto-superior to formal mortgages in special cases.

The profound argument made by Stiglitz (1994) is that market failure is a fundamental cause of poverty and financial market failures, which mainly arise from market imperfections, asymmetric information and the high fixed costs of small-scale lending, limit the access of the poor to formal finance, thus pushing the poor to the informal financial sector or to the extreme case of financial exclusion. In addition, it is argued that improving the access of the poor to financial services enables these agents to build up productive assets and enhance their productivity and potential for sustainable livelihoods (World Bank, 2001). Hence the bottom line argument is that improving the supply of financial services to the poor can directly contribute to poverty reduction (Jalilian and Kirkpatrick, 2002) (Green et al., 2005, p. 19) Housing plays a vital role in the economy (see Sheng, 1997). This is due to its following attributes. First, a home is both consumption good as well as an investment (see Malpezzi, 1990). The investment aspect of homeownership helps to increase wealth (i.e., reduces poverty—see Buckley, 1994, Englehardt, 1994, Sheng, 1997 and Haurin et al., 2002). Second, homeowners support their neighborhood more than renters, as they participate in crime prevention and support public schools. They are better citizens and vote at a higher rate (see Haurin et al., 2002). Homeownership fosters investment in local amenities and social capital, thus enhancing the status and quality of the community (see DiPasquale and Glaeser, 1999). Policy makers therefore have an obligation of ensuring access to this indispensable asset through an efficient financial intermediation system.1 The ongoing subprime mortgage crisis (emanating from the U.S.) constitutes a financial market failure.2 This is because financing a home for a subprime borrower poses a dilemma in an efficient capital market (see Fama, 1970 and Fama, 1991). The ingenious mortgage bankers figured out an innovative way (which turned out to be disastrous as explained below) to get financially strapped individuals to qualify for a home loan using lax underwriting standards.3 This innovative way was through the use of “exploding” adjustable rate mortgages (ARMs) with unusually low introductory (i.e., “teaser”) rates, which reset to higher rates at the expiration of the introductory ones (see Gapper, 2007).4 This was supposed to facilitate access to a home for borrowers and to help them establish some credit history before qualifying for refinancing with a fixed rate mortgage. The implicit assumption was that the eventual appreciation of the home would bail out the borrower prior to the expiration of the “teaser” rate as he/she would be able to refinance and not be exposed to the shock of higher mortgage payments. Unfortunately, the opposite happened, and not only did payments increase drastically (with the termination of the teaser rate) but home prices fell too. This made it difficult for borrowers to keep up with their payments. Also, they could not refinance (or sell) their homes, as their values were significantly below their mortgage balance (leaving them with negative equity). They had no option but to default. It is estimated that more than 2.4 million American families have lost their homes through foreclosures (see Economist, 2007a, Ip and Paletta, 2007 and Mason and Rosner, 2007). This number is expected to go up to 9 million, as a second wave of defaults (stemming from the above exploding ARMs, along with negatively amortizing ARMs, which increase principal at the end of the low teaser, or optional payment periods) work their way through the system (see Wachter, 2008 and Ward, 2009). This crisis has drastically impacted the global economy as elaborated below.5 The repercussions of the subprime “woes” are being felt (in both the real, as well as the financial sectors of the economy) globally. In the case of the real sector, the crisis has led to: (i) an increase in supply of homes for sale (due to repossessions), thereby depressing their prices and negatively impacting the construction sector, sales of durable goods, and thus the manufacturing sector (see Economist, 2007b, Economist, 2009a and Spector, 2007); and (ii) an economic contraction (in the U.S.) impacting the economies of its trading partners through decreases in trade, investment and remittances, thus leading to a backlash against globalization (see Economist, 2009b). In the case of the financial sector, the reduction in value of underlying collateral of mortgages (i.e., homes) has resulted in: (i) a loss in market value of more than $290 billion of bonds associated with subprime mortgages, devastating the capital base of major financial institutions on both sides of the Atlantic (see Economist, 2008 and Barkley, 2008); (ii) the failure of more than 40 subprime lenders (see Authers, 2007); (iii) the inability of the U.S. government sponsored agencies (such as Fannie Mae and Freddie Mac) to provide some relief (in the crisis), as they themselves have been placed under conservatorship (see Crutsinger, 2008); (iv) the scrutiny of the remaining subprime lenders from state and federal regulators (see Ip and Paletta, 2007); (v) the tightening of credit to firms (in other industries, hedge funds, private equity groups, etc.) is anticipated to lead to a severe recession in the U.S. and a decline in value of American assets (see White et al., 2007);6 and (vi) the spreading of systemic problems from the U.S. to overseas, increasing capital market volatility, and “crimping” world growth (see Economist, 2007b and Gapper, 2007). The International Monetary Fund estimates the total losses to reach $2.2 trillion (see Wolf, 2009).7 The purpose of this paper is to present a novel way of home financing, using leverage endogenously among underprivileged aspiring homeowners (via a housing finance cooperative) to avert subprime like financial market failure. This is radically different from the formal exogenous form of financing via an intermediary (such as a bank, Savings and Loan Association or a mortgage company—see Jaffee and Renaud, 1998). Our goal in calling for the formulation of a specialized institution (“circuits”) catering to help aspiring homeowners is in contrast to the trend towards integrating formal mortgage underwriters with capital markets (see Diamond and Lea, 1992 and Jaffee and Renaud, 1998).8 This is because the formal system is onerous to the underprivileged, especially in efficient financial markets where subprime borrowers are subject to a high cost of funding, as they are perceived to be more risky (as discussed above). Furthermore, inflationary shocks (or volatility of interest rates) on formal mortgages create a tilt in real payments, which makes aspiring homeowners ineligible by the income requirement of financial institutions (see Buckley, 1994). It is therefore not surprising that formal intermediaries are not used by more than 70–80% of homeowners in the developing world (see Okpala, 1994 and Ferguson, 1999). The rationale behind the low underwriting rate of formal intermediaries (in the developing economies) is attributed by Jaffee and Renaud (1998) to the high costs of lending, especially when property rights, foreclosure procedures (needed for real estate to serve as collateral) and accurate methods of valuing property are not well established. Another strand of the literature (stemming from housing micro-finance) holds two factors primarily responsible for deterring (the underprivileged) from gaining access to formal mortgage finance (see Ferguson, 1999). One is the lack of affordability to legal buildings, as most of the dwellings in low-income settlements do not comply with building regulations, nor do they have formal land titles. This deters formal institutions from lending to these households. The second is the instability of income to secure repayment. Since the goal of this paper is to establish a basic framework to increase the affordability of legal (formal) buildings, we assume that the government has already laid the necessary infrastructure of the following: law and regulation, information, risk pricing, payment and settlement, and financial stability (see Renaud, 2009). This assumption stems from the research that organizations and structure of the financial system plays a crucial role in the quality and rate of economic growth (see Goldsmith, 1969 and Renaud, 2009). We also assume that prospective homeowners have stable incomes. This assumption is consistent with Levenson and Besley (1996), who find increasing ROSCA participation associated with income stability. We refrain from delving in the micro-financing of progressive housing (where households acquire land through purchase or invasion, thereby improving the structure and legal tenure incrementally and lobby for basic services), as the cost of micro-financing is higher than conventional banks, and thus still burdensome to the underprivileged (see Ferguson, 1999). Furthermore, this system of micro-financing is not sustainable in the long run due to high rates of default, as the mortgages are priced in an ad hoc manner (see Lee, 1995). Thus, after establishing a basic framework to increase affordability of a formal home, we plan to extend it to the issue of income instability in future research.9 The motivation behind our call for the establishment of cooperative financing institutions stems from the ad hoc practice of clans in Oman to fund the purchase of homes of their poor brethrens with gratuitous (interest-free) loans (termed as qard hasan).10 These are conducted along the lines of an Islamic endowment or trust (termed as waqf; plural: awqaf) which blends features of philanthropy with social service. The seed funding for this institution emanates from the cash contributions of well-to-do clan members from supplementary charitable sources of infaq (voluntary charity for a specific purpose) or sadaqah (voluntary charity) (see Cizakca, 2000).11Bremer (2004) classifies this as a rebirth of the waqf model and explains it as follows: Charities have played many critical functions in Islamic societies and have contributed to making these societies more just and fair through a number of mechanisms, in addition to the obvious one of providing service to the poor. Over and above their role in delivering services, Islamic charities served as a mechanism for narrowing social distances and reducing inequalities. Charities have particularly, served as a bridge between the haves and have-nots. They have provided a means by which the wealthier elements of society interact with poor individuals, come to know them as individuals, and recognize their obligations to assist them in combating poverty, its causes and effects. This linkage helps to keep low-income groups from becoming isolated from the social mainstream, strengthening the overall social order. Charities, particularly awqaf, provided a source of support for institutions and interest groups independent of, and sometimes in opposition, to the state. Islamic charities historically have played an additional role in society, that of promoter of decentralized economic development. Whether the charity is a waqf in the medieval Levant establishing commercial centers or building a khana for traveling business people, or an Indonesian zakat-funded charity teaching business management skills in today's Indonesia, Islamic charities have been actively engaged in economic development for centuries. In this respect, they reflect the blending of the religious and the secular, the social and the economic, that is the key characteristic of the Islamic idea. (Bremer, 2004, pp. 5–7) She explicates the last sentence further in a footnote as follows: This combination can be found present in the West in urban development-oriented civil society, as well, such as pro-poor non-profits that address inner-city economic development and civic business associations that promote the development of their respective cities. Generally, however, the mix of economic development and social service with charity is much more developed in the tradition of Islamic charities than in the more “purely” charitable tradition of Western society. (Bremer, 2004, fn. 4, p. 7) The above stated practice of Omani clans is also in accordance with the prognosis of King and Levine (1993) and Levine (1997), as it facilitates in mitigating risk and reducing transaction costs for the underprivileged masses. It also has the potential of resolving the market failure which has provoked the ongoing subprime crisis. However, it has not been scrutinized until now by academics, as the data on this form of financing (conducted privately among clan members) is not in the public domain and hard to obtain. The reason for this is that tribal elders do not want to disclose their economic power. Furthermore, commercial banks also find it hard to compete against this practice and try to restrain it. They have not succeeded, as the average Omani is extremely loyal to his/her clan and is not willing to divulge any information. We have been quite fortunate to learn of this practice through our interaction with some Omanis, who have disclosed it for the purpose of this study. Thus, the specific issues of concern of this paper are as follows: (i) Can a specialized circuit (in the form of a financial cooperative—using endogenous leverage) alleviate credit risk better than the formal mode of prime sector financing (using exogenous leverage)? (ii) Can it alleviate inflation risks better than its formal counterpart? (iii) Can it be pareto-optimal over its formal counterpart, thereby uplifting the economic status of the underprivileged and stimulating economic development? The three interrelated issues (stated above) encompass the optimal pricing of loans. This is a formidable task, as capital structure (i.e., choice of debt-equity) constitutes a major puzzle in finance (see Harris and Raviv, 1991). Typically, financial institutions price their loans in an ad hoc manner using credit rationing (in the form of initial loan-to-value (LTV) ratio and income ratio—see Jaffee and Stiglitz, 1990). This is not efficient, as empirically demonstrated by the literature on banking crises and real estate cycles (see Lee, 1995, Herring and Wachter, 1999 and Malpezzi and Wachter, 2002). Furthermore, it exacts a huge toll on the macroeconomy, as Renaud (2003) and Hoshi and Kashyap (2004) estimate the costs of real estate crises in Indonesia, Thailand, Japan and United States to be roughly 65%, 45%, 20% and 3% of the GDP respectively.12 We investigate the above interrelated issues by blending two streams in the literature. First, we focus on Security (Mortgage) Design, which espouses that risk management through the optimal employment of secured debt and debt maturity reduces agency costs and enhances firm value (see Stanton, 1998 and Eisdorfer, 2008).13 This view is reinforced in Ebrahim and Mathur (2007), who hypothesize that real estate mortgages have to be priced meticulously by adequately collateralizing them with the underlying tangible assets and income of the borrower. This condition is more stringent than the ad hoc credit rationing ones, and ensures that the financier is not exposed to the risk of homeownership. That is, the mortgage is nearly risk-free.14 This condition follows from their (two period) model, that under rational expectations (symmetric information), a collateralized risk-free loan is pareto-superior over its risky counterpart. The rationale behind this result is as follows. In a world of symmetric information, a financier would rationally price risky mortgage to incorporate deadweight costs of default in the form of high interest rates to be transmitted to the borrower (prospective homeowner). This reduces the welfare of the homeowner. If the risky interest rate is high enough, prospective homeowners forgo a risky mortgage in lieu of a risk-free one (with a conservative debt ratio and low cost of financing). This has credence in the real world, as Singapore (which has the highest homeownership rate—of 84% in the world) does not allow excessive risk taking via mortgages on public housing units (see Edelstein and Lum, 2004). Second, we focus on financial cooperatives, as they have played a crucial role in the economic development of Germany, United Kingdom, United States and many industrialized countries (see Shay, 1992). These originate from Accumulating Savings and Credit Associations (ASCRAs), which are interrelated to Rotating Savings and Credit Associations (ROSCAs) (see Grossman, 1992 and Bouman, 1995).15,16,17.Besley et al. (1993) find ROSCAs to be appealing, as they provide a pareto-superior solution to the problem of purchasing an indivisible (lumpy) good (such as a home). This is attributed by them to the improvement in social welfare stemming from intertemporal trade due to the mobilization of savings (under ROSCAs), which otherwise would have been idle under autarky. In contrast, Besley et al. (1994) establish that ROSCAs (in general) are less flexible and therefore less efficient than formal credit markets. However, there are special cases where a random ROSCA may increase welfare over a formal credit market due to the element of chance. Besley et al. (1994) therefore predict the decreasing role of ROSCAs with increasing economic development. It is thus a puzzle to see intermediaries with ROSCA roots (like Building Societies, Credit Unions and Mutual Savings Banks), which not only proliferate, but also compete head on with for-profit intermediaries (such as Commercial Banks, Stock Savings and Loans, etc.) in highly developed economies such as the Germany, U.K. and U.S. This is attributed in the literature to the (i) Inability of consumers to evaluate the quality of goods (or service) promised (or delivered) (see Hansmann, 1980); (ii) Inability of consumers to observe (or measure) the output (or benefit) (see Easley and O’Hara, 1983); (iii) Adaptability to a changing economic environment (see Emmons and Mueller, 1997); (iv) Narrowly defined activity where members have homogeneity of opinion (see Hart and Moore, 1998); (v) Mitigation of adverse selection (due to availability of adequate information on borrowers–see Buijs, 1998; and Smets, 2000); (vi) Alleviation of moral hazard (in the form of timeliness of payment and reduction of default–see Buijs, 1998; and Smets, 2000); and (vii) Relatively lower administrative and transaction costs (see Buijs, 1998). A housing finance cooperative involves a group of people, who form a society to enable them to raise funds endogenously (among themselves).18 It serves as a specialized mutual savings bank for facilitating the purchase of a lumpy good (i.e., a house) for the members of the cooperative. The member (for whom the house is being purchased) repays the principal along with lending an additional amount (in lieu of an interest payment in a formal mortgage) to the cooperative. This simultaneous action allows members to offset the cost of borrowing with the benefit of lending, thus yielding a facility with a zero interest rate (assuming negligible administrative costs).19 The intuition behind the offsetting of interest rates is akin to the “barn raising” example of nineteenth century frontier farmers in the United States, as described in Besley et al. (1993). This was accomplished by forming a specialized ROSCA requiring in-kind contribution of member farmers in one region to help each other to build a barn. This specialized ROSCA was dissolved after every member in the group had a barn. A housing finance cooperative can thus be construed as a special form of ASCRA, which is distinct from a ROSCA. This is because in this scheme of affairs one group of members do not benefit at the expense of the others. It is the cooperative which is the sole beneficiary or the benefactor (based on the net present values (NPV) of cash flows). We integrate the above two streams of literature thereby assuming the existence of an information architecture, where property rights, foreclosure procedures (needed for real estate to serve as collateral) and accurate method of valuing property are well established (see Levine et al., 2000 and Renaud, 2009). We initiate our study with a simple framework and extend it to study inflationary shocks, as well as prepayment options. Our efforts yield four key results described as follows. First, we optimally price both the formal (fixed rate—prime sector) mortgage, as well as the cooperative home mortgage. The term “price” in our paper is used in a broad sense (consistent with Baltensperger, 1978) to include not only the interest rate (as in the formal mortgage) but also the loan-to-value ratio, as well as the tenure of the facility (see also Eisdorfer, 2008). Our pricing algorithm is more scientific than the ad hoc credit rationing constraints used currently by banks. Risk control at the micro-level is important to arrest the volatility at the macro-level, in accordance with the prognosis of Sheng (1997) and Renaud (2009). Second, we realize that the lien profile of a mortgage issued by a housing finance cooperative is linear, in contrast to the concave lien profile of a formal intermediary. This helps in reducing the tenure of the endogenous mortgage, building up an “equity cushion” faster, alleviating credit risk and the overall cost of financing.20 Third, a housing finance cooperative is also able to control interest rate risk better than its formal counterpart. The endogenous use of leverage ensures that any increase in the cost of borrowing is offset by the benefit of lending. Inflationary shocks thus impact on the tenure of the cooperative facility instead of pricing out the prospective homeowner by increasing the front-end costs of owning a home (as in a formal prime sector mortgage). Finally, we acknowledge that the gains from intertemporal trade make home financing through cooperatives pareto-superior [pareto-inferior] over its formal prime sector counterpart, depending on the characteristics of a home, that of a borrower and the underwriting standards adopted. This verifies the claim of (i) Besley et al. (1994) that efficiency of a cooperative is mixed when contrasted with formal credit markets; and (ii) Hart and Moore (1998) that a cooperative works well when it is focused on a limited scope of activities. This result does not incorporate the lower transaction costs of a cooperative stemming from its organization as a non-profit entity. Furthermore, we do not even incorporate (i) the relatively low default costs (stemming from the mitigation of adverse selection and moral hazard) in contrast to the high ones (stemming from subprime borrowers) (see Buijs, 1998 and Smets, 2000); or (ii) the prepayment advantage of a cooperative mortgage. If we were to do so, the results would overwhelmingly tilt in its favor. Nonetheless, it is a responsibility of a cooperative manager to structure its portfolio by catering to the disadvantaged (with low income), aspiring to purchase a home (with low initial value and medium to high risk) and using the following underwriting constraints: medium income multiplier and high confidence level. This would suffice in internally generating a capital surplus critical for sustaining its growth and ultimately improving the status of the underprivileged. We thus conclude that an adaptation of an “informal” financial system of an ASCRA (blending philanthropy with social service), which is focused in a narrowly defined activity of funding homes (by mitigating adverse selection, moral hazard, administrative costs and transaction costs) can play a vital role in economic development by providing access to financial services to those shunned by the formal system. This result is consistent with Emmons and Mueller (1997), Buijs (1998), Hart and Moore (1998) and Smets (2000). It also supports the following assertion of Callier (1990): Informal finance persists and often flourishes because it resolves important problems that are handled poorly or not at all by most formal financial systems in developing countries. (Callier, 1990, p. 273) The paper is organized as follows. Sections 2 and 3 illustrate the theoretical underpinnings behind the design of both formal and cooperative mortgages to contrast their efficiencies. Finally, Section 4 presents our concluding remarks.

The formal housing finance system is onerous to the financially strapped individuals in an efficient capital market, as it prices the incremental risk of the subprime borrower. This has a capacity to cause a market failure, as the high cost of funding can deter homeownership. This paper proposes the establishment of a special circuit in the form of a cooperative to overcome this market failure, by integrating the two streams of the literature (comprising of mortgage design and informal saving schemes). Intuitively, we propose a system akin to the “barn-raising” one in nineteenth century frontier U.S., where the in-kind contribution of prospective homeowner offsets the cost of borrowing such that the net contractual interest rate is zero. This form of financing is a special case of ASCRA and is practiced in a limited way by clans in Oman to help their destitute clan members. It is consistent with the prognosis of King and Levine (1993) and Levine (1997), as it facilitates in mitigating risk and reducing transaction costs for the underprivileged masses. We implicitly assume that the cooperative has access to seed funding for its incorporation through either a charity or a mutual savings institution (such as a building society/credit union/mutual savings bank) or governmental agency or a non-governmental agency (NGO) or a supra national agency like the World Bank. Once it is established, it needs to be managed carefully for it to be self-sustaining.43 We assume the existence of an information architecture and optimally price mortgages (in the spirit of Baltensperger, 1978) to contrast the one made by the housing finance cooperative with that of a formal intermediary in the prime sector. This is accomplished in a more scientific way instead of using the ad hoc credit rationing constraints currently used by banks. Our efforts yield the following four key results. First, a cooperative home mortgage is allocatively more efficient than a formal prime sector mortgage, as the loan amount is marginally higher. This is despite the fact that both are subject to similar asset value constraints. However, a cooperative mortgage involves marginally lower total monthly payments with less tenure. This is also in contrast to the fact that both are subject to similar income constraints. A cooperative mortgage is thus less onerous than a formal mortgage, as its linear profile necessitates principal payments in contrast to a formal mortgage which necessitates principal plus interest. Second, the linear lien profile of a cooperative home mortgage also makes it less prone to defaults in contrast to the concave profile of a formal mortgage. This stems from the fact that a cooperative mortgage builds equity faster for the homeowner, leaving a larger safety net for the financier. Thus, we conclude that a cooperative mortgage is a better alternative than a formal mortgage, as it has the potential of reducing macroeconomic volatility in accordance with the prognosis of Sheng (1997) and Renaud (2009). Third, a housing finance cooperative performs better than its formal counterpart during periods of volatile interest rates (stemming from changes in inflationary expectations). This is attributed to the endogenous use of leverage, where the volatility in interest rates marginally impacts on the LTV and the tenure instead of pricing out the prospective homeowner due to the increase in affordability (in case of the formal mortgage). Finally, the overall efficiency of a cooperative vis-à-vis a formal prime sector mortgage is contingent on the underlying characteristics of a home, that of the borrower and its underwriting standards. For some values of these parameters, a cooperative constitutes a special case reported in Besley et al. (1994) and Hart and Moore (1998), which is pareto-superior to the formal mortgage. However, for other values of the above parameters, a cooperative is pareto-inferior to a formal mortgage. It should be noted that our analysis does not incorporate the relatively low administrative, default and transaction costs embedded in the cooperative one in contrast to the subprime one (see Buijs, 1998 and Smets, 2000).44 Furthermore, there is no prepayment cost in a cooperative mortgage as opposed to a formal mortgage, where it results in a higher interest rate or a higher initiation fee (see Hall, 1985). If we were to incorporate these lower costs, a cooperative would still dominate in terms of its efficiency.45 This competitive advantage of cooperatives (in the form of credit unions) in the U.S. (despite their handicap described below) has subjected them to intense pressure from the banking industry, which threatens to contain its growth through legal and political means (see Wysocki, 2006). Nonetheless, a diligent manager of the cooperative should underwrite a portfolio which internally generates a capital surplus crucial for sustaining its growth. This implies: (i) catering to the disadvantaged (with low income) who aspire to purchase a home (with low initial value and medium to high risk), and (ii) using the following underwriting constraints: medium income multiplier and high confidence level. The competitive advantages of the housing finance cooperative (as elaborated above) outweigh its disadvantages ensuing from its (i) illiquidity (as cooperatives are forced to hold the mortgages until maturity, in contrast to the formal intermediaries, who can securitize them in the secondary markets), and (ii) inability to raise funds quickly due to inordinate demand (as charity funded cooperatives do not have the same ability as formal intermediaries in raising funds and are compelled to do so by appealing to their constituents, and meticulously managing its portfolio). Nonetheless, a cooperative serves as an exemplary special circuit that does not depend on government subsidies, and manages available resources more efficiently with reduced risk (in contrast to its formal counterpart). We therefore recommend it to be adopted globally to help the disadvantaged gain from its economic benefits (see Renaud, 2009). The ensuing benefit of homeownership will help in elevating the economic status of the underprivileged, fostering investment in local amenities and social capital, thus enhancing the quality of the community and stimulating economic development (see Malpezzi, 1990, Buckley, 1994, DiPasquale and Glaeser, 1999 and Haurin et al., 2002).