جستجو برای بازده مالی و خصوصی سازی تأمین اجتماعی
|کد مقاله||سال انتشار||مقاله انگلیسی||ترجمه فارسی||تعداد کلمات|
|24489||2013||18 صفحه PDF||سفارش دهید||محاسبه نشده|
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : Review of Economic Dynamics, Volume 16, Issue 2, April 2013, Pages 253–270
I develop a general equilibrium model in which the quality of household financial decisions is endogenously determined by the incentives to exert effort in learning about financial opportunities. The model generates predictions for asset market participation and returns across households. Moreover, search for financial returns enables the model to generate a more skewed equilibrium wealth distribution. In this context, social security privatization affects household search effort, asset market participation and the competitiveness of the asset market. Privatization reduces average welfare and this reduction is somewhat magnified by the search friction. While some have suggested that household decision making could be important for the consequences of privatization, my analysis does not bear this out.
Standard models of household consumption and savings assume that all households earn the marginal product of capital on their savings. This assumption is at odds with the evidence on the actual financial arrangements of households, which shows that a substantial number of households seem to have difficulty choosing a portfolio. Evidence of these difficulties takes several forms. For example, households may not allocate any savings to equities or they may hold under-diversified portfolios ( Haliassos and Bertaut, 1995; Calvet et al., 2007 ). Surveys of financial literacy have also found that many house- holds do not understand some fundamental financial concepts such as the difference between bonds and stocks ( van Rooij et al., 2007 ). Other studies have found that those households that spend more effort planning for retirement reach retire- ment age with more wealth ( Ameriks et al., 2003; Lusardi and Mitchell, 2007 ). In addition, researchers have found that experimental subjects have difficulty making sound financial decisions even when there is a clear normative ranking of the available choices ( Choi et al., 2010 ). Many of these studies find that households with higher levels of income, wealth and education have more success in making sound financial decisions. One way of understanding this empirical evidence is to view managing a portfolio as an activity that requires effort, with the incentive to devote effort varying across households. For example, households with high levels of wealth have more to gain in absolute terms from improving the return on their portfolios. Alternatively, highly educated households may be better able to assess the various risks and trade-offs that arise in choosing a portfolio. In this paper, I develop a general equilibrium model of household saving behavior in which households must exert effort to learn about the available investment opportunities by searching for high returns. In the model, a household can raise the expected return on its portfolio by devoting more effort to search. The benefit of search exists because there is dispersion in the rates of return offered by financial intermediaries. When households are imperfectly informed, intermediaries can still attract savings even if they are not offering the highest return, but intermediaries that offer higher returns will attract more savings. Therefore, intermediaries face a trade-off between the number of savers they will attract and their profit margin. In the model, these competing forces balance in such a way that intermediaries choose to offer a range of returns, which gives rise to an endogenous distribution of offered returns that depends on the search and saving behavior of households. The search friction is meant to capture a relationship between the effort households exert managing their finances and the return earned on their savings. Search frictions are a convenient way of modeling these relationships, but in reality there are more forces at work. For instance, some households have an imperfect understanding of the terms of the financial contracts or the principles of portfolio management. The point of view I take here is that households could learn these things if they devoted more effort. There is a long tradition of literature that considers the role of information acquisition in financial markets. This in- formation acquisition is often formulated in terms of learning about the payoff of an asset at some cost. 1 In these models, trade takes place in centralized markets at known prices. Van Nieuwerburgh and Veldkamp (2010) use a model in this style to rationalize under-diversified portfolios. 2 Another line of literature considers the acquisition of information about prices and trading opportunities in asset markets affected by search frictions ( Duffie et al., 2005; Lagos and Rocheteau, 2009 ). This literature considers trading environments that resemble over-the-counter markets in which investors and dealers negotiate over prices. In this paper, I consider costly information acquisition in the context of a life- cycle savings model and explore the impact of the social security system on the incentives for households to expend effort in acquiring information in financial markets. The model presented here also features search friction, but prices are posted rather than negotiated, an assumption which seems more appropriate for retail financial markets. The model is based on a heterogeneous-agent life-cycle savings model in the style of Bewley (undated) , Huggett (1993) and Aiyagari (1994) . It is natural to model household financial decisions within this framework because financial choices are non-linear functions of household assets, which means the distribution of financial outcomes will depend on the distribution of wealth. I modify the Bewley–Huggett–Aiyagari framework to include a search friction in the asset market. In the model, intermediaries post rates of return on risk-free assets and households choose how much time to spend searching among the offers for a high rate of return. To generate dispersion in returns, I use insights developed in the literature on search and equilibrium price dispersion, in particular from the work of Butters (1977) and Burdett and Judd (1983) . 3 The other side of the asset market, in which intermediaries interact with production firms, is frictionless and does not play a major role in the analysis. The model generates predictions for three aspects of household saving behavior that are absent from the standard model with a frictionless asset market: the amount of time that households spend managing their finances, asset market participa- tion rates, and the distribution of returns. I compare the model’s predictions to the available data and the model performs well on many of these dimensions. The search friction also has implications for the distribution of wealth. Bewley–Huggett–Aiyagari models traditionally have had difficulty explaining the extreme skewness of the distribution of wealth. Cagetti and De Nardi (2006) , Campanale (2007) and Benhabib et al. (2011) have shown that heterogeneity in household savings technologies can generate a more re- alistic distribution of wealth. While preceding work has relied on exogenous variation in the rates of return that households earn on their savings, the model presented here offers an endogenous mechanism that produces heterogeneity in returns. The mechanism at work was first pointed out by Arrow (1987) and further developed by Peress (2004) : when households can pay to acquire information that will raise returns, wealthy households will acquire more information and earn higher re- turns, which leads to a more concentrated distribution of wealth. Indeed, the model predicts that wealthy and high-income households will be more likely to participate in the asset market and earn higher returns conditional on participation. This prediction is the result of a scale effect: as a household accumulates wealth, its incentive to search increases and it will earn higher returns on average. I am able to quantitatively explore the role of Arrow’s mechanism in shaping the distribution of wealth. I find that the search friction does produce additional skewness in the distribution of wealth, raising the share of wealth held by the top quintile by 4.5 percentage points. I compare these predictions to those of a model with a fixed cost of asset market participation. That model does not generate as much wealth inequality because there is not heterogeneity in returns among those who participate in the market. In an application of the model, I analyze the consequences of social security privatization in an environment in which households have difficulty allocating savings to the best investment opportunities. Many proposals for social security reform give individual households a larger role in managing their social security savings, but the empirical household finance liter- ature raises questions about how well-prepared households are to take on this increased responsibility. To my knowledge the literature on social security reform has not formally analyzed how household financial decision making frictions affects the welfare implications of privatization. 4 There are two views on how household financial decisions might affect the consequences of introducing private social security accounts. One perspective is that some households will make poor choices for their private social security accounts and will have few savings with which to retire. 5 Another perspective is that fixed costs in portfolio management exacerbate the distortions caused by the social security system because they create an economy of scale in saving. If the system were privatized, the benefit of private saving would rise for two reasons. First, there is the usual reason that it is more important to save to provide for one’s own retirement. Second, the accumulation of private savings would lead some households to overcome the fixed costs of managing their portfolios and realize a higher return on their savings. 6 The model is able to generate both of the effects mentioned above. On the one hand, some households will have poor financial outcomes either due to a lack of effort or due to bad luck. On the other hand, privatization increases the size of the household’s portfolio and strengthens the incentive to search, which raises their expected returns above what they would have earned if they did not adjust their search effort. In the context of this model, social security reform can affect welfare through an additional channel: the competitiveness of the asset market. Privatization leads most households to increase their search effort as they hold more wealth. This increase in search effort has two partially offsetting effects on the distribution of offered returns. For those households who were already participating in the asset market, they are now more likely to have multiple offers and the higher degree of competition motivates firms to make more attractive offers. However, there are some households that were previously non-participants that enter the market after the reform. As these new participants are on the margin of non-participation, they choose relatively low levels of search effort, which motivates firms to make less attractive offers. The implications of privatization for the competitiveness of the asset market are therefore ambiguous. To explore these effects, I simulate a partial privatization of the social security system in which benefits and payroll taxes are reduced by roughly 50% and I compute a 150-year transition path from the initial steady state to the privatized steady state. I find that social security privatization leads households to devote more effort to search on average and leads firms to make more competitive offers, but the magnitude of the latter effect is small. The partial privatization produces an average welfare loss equivalent to 1.02% of consumption. To assess the impact of the search friction on this result, I also perform the same experiment in a benchmark economy that is identical except that households earn the marginal product of capital on their savings without having to exert any effort. The average welfare loss from privatization in the benchmark economy stands at 0.88% of consumption. In the long run, after the transition costs have been paid, privatization leads to welfare gains. Here the average welfare gain is smaller when the search friction is included, standing at 6.2% of consumption rather than 6.7% in the benchmark model. The purpose here is not to evaluate social security reform per se, but to evaluate the importance of asset market frictions to the analysis. My interpretation of the findings is that the incorporation of frictions into household financial decision making does not fundamentally alter the analysis of social security privatization and so the analysis with the standard model is robust in this regard. The next section introduces the model and Section 3 discusses its calibration. Section 4 describes the model’s steady state and compares the model’s prediction to the data. Section 5 presents the privatization experiment and Section 6 concludes.
نتیجه گیری انگلیسی
In this paper, I have argued that several features of the data on household savings behavior can be understood in terms of the incentives to learn about investment opportunities and I have proposed a model of savings in which households must search for returns in the asset market. The model accurately predicts that households will spend more time managing their finances and be more likely to participate in asset markets as they approach retirement. The model also predicts that wealthy and high-income households will spend more time managing their finances, be more likely to participate in the asset market and earn higher returns conditional on participation. With regard to social security privatization, the analysis is not an assessment of the desirability of social security re- form so much as an assessment of the importance of asset market frictions in this debate. The key message is that the introduction of a friction in the asset market as modeled here makes privatization less attractive, but only slightly so.