框架效应与贷款决策:一项金融素养的实验研究

Framing Effect and Loan Take-up:An Experimental Study on Financial Literacy

过度借贷是一个常见的非理性决策。近年来,网络借贷市场的激烈竞争降低了申请贷款门槛,导致金融素养较低的人群开始参与网络借贷,进一步加剧了过度借贷现象。本文采用实验室实验方法,给被试者提供本质上完全等价的贷款产品,同时外生地变化贷款成本形式(绝对数值形式的利息相比于百分数形式的利率、单期利率相比于多期复利),研究贷款成本的展示形式对贷款意愿是否存在影响。结果表明,相比于月利率,展示月利息使得被试者的贷款接受率显著上升21.3个百分点;相比于年化复利,展示月利率使得被试者的贷款接受率显著上升7.91个百分点。这两种框架效应可分别由金融素养的知识层面和思维层面所解释。本文对网络小额贷款场景下的过度借贷提出了一种新的解释——贷款成本的框架效应,并采用实验经济学方法对此提供了可靠的证据。本文研究结论提示,加强金融教育和制定相关法规双措并举,有助于人们做出更审慎的贷款决策。

Over-borrowing is a common irrational decision.In recent years,fierce competition in the online loan lending market has lowered the threshold for applying for loans,leading people with low financial literacy to start participating in online borrowing,further exacerbating the phenomenon of over-borrowing.This article examines a behavioral bias strongly associated with financial literacy—the framing effect.The framing effect can cause the market to have an incentive to manipulate the way information is presented,which implicitly manipulates consumers'decision-making.Therefore,understanding the framing effect in loan decision-making can help policy makers make more specific regulations on the displaying form of loan content and protect the rights and interests of financial consumers more effectively.This paper uses a laboratory experiment which provides subjects with economically equivalent loan products,and at the same time exogenously change the form of loan costs to study whether the displaying form of loan costs affects loan take-up decisions.Existing literature generally designs loan decisions as intertemporal risk decisions in the laboratory(Ginéet al.,2010;Wu Zuguang et al.,2012;Baland et al.,2017).On this basis,we added conditions such as loan cost,consumption value,future income,and overdue fine to simulate the characteristics of consumption loans.This experiment adopts a between-subject experimental design,and the subjects randomly enter into one of Group A(monthly interest amount),Group B(monthly interest rate),and Group C(annual compound interest rate).The information content of the three experimental groups is economically equivalent,and the only difference lies in the displaying form of the"loan cost".Taking the monthly interest rate of 0.5%as an example,Group A sees"the monthly interest rate is 0.5%",Group B sees"the monthly interest amount is 2.5 experimental coins(assuming the loan amount is 500 experimental coins)",Group C sees"the annual compound interest rate is 6.17%".Holding other variables constant,the loan take-up rate in Group B(monthly interest amount)is 21.30 percentage points higher than that of Group A(monthly interest rate)(which is 21.30/43.42=49.06%higher than that of Group A),and it is statistically significant.Holding other variables constant,the loan take-up rate in group C(annualized compound interest)is 7.91 percentage points lower than Group A(monthly interest rate)on average(which is 7.91/43.42=18.22%lower than that of Group A),and it is also statistically significant.This paper further explores the channels of the two framing effects and finds that they result from different channels of financial literacy.In a broad sense,financial literacy includes not only people's mastery of financial knowledge,but also people's ability to process economic information and make wise financial decisions(Lusardi and Mitchell,2014;Liu,2018).After the loan take-up decisions,we test the financial literacy of the subjects.More than 90%of the subjects correctly answer the questions about the conversion of interest rates and interest amount.We also test the subjects'calculating ability to proxy the heuristics in financial decision-making.Subjects with weaker calculating ability are more prone to the framing effect of interest amount and interest rate.The financial literacy quiz also finds that 71.92% of the subjects have the bias of underestimating exponential growth.The stronger the exponential growth bias,the more prone to the framing effect of single-period interest rate and multi-period compound interest.Therefore,the framing effect of interest amount and interest rates can be explained by the heuristics of financial literacy,while the framing effect of single-period interest rate and multi-period compound interest can be explained by lacking financial knowledge.This paper proposes a new explanation for over-borrowing in the context of online consumption loan-the framing effect of loan costs,and provides clean and reliable evidence for this using the method of experimental economics.Secondly,this paper divides financial literacy into knowledge and heuristics,and distinguishes the different causes of the two framing effects.This paper suggests that we need the combination of financial education and financial policy to manage the over-borrowing problem in the Internet age.To alleviate the framing effect of single-period interest rate and multi-period compound interest,government can provide more financial education on the concept of compound interest to the public.To alleviate the framing effect of interest amount in absolute values and interest rate in percentage points,government can directly make clear regulations on the displaying form of loan information,such as adjusting the displaying form of loan costs from single-period interest amount to single-period interest rate,or further adjusting to multi-period compound interest rate.