CAJ | 학술논문

货币搜寻理论清晰刻画了货币交易媒介的本质功能，已成为研究通货膨胀福利成本等问题的重要方法。本文在对货币搜寻理论进行介绍的基础上，使用货币搜寻模型和福利三角形法对中国通货膨胀的福利成本进行了估计和比较分析。研究表明，基于货币搜寻模型对通货膨胀福利成本进行估计的结果更为可靠，与通货膨胀率为0相比，中国10％的通货膨胀率造成的福利成本约为收入的1．2％，高于采用福利三角形法得到的估计结果。
화폐수심이론청석각화료화폐교역매개적본질공능，이성위연구통화팽창복리성본등문제적중요방법。본문재대화폐수심이론진행개소적기출상，사용화폐수심모형화복리삼각형법대중국통화팽창적복리성본진행료고계화비교분석。연구표명，기우화폐수심모형대통화팽창복리성본진행고계적결과경위가고，여통화팽창솔위0상비，중국10％적통화팽창솔조성적복리성본약위수입적1．2％，고우채용복리삼각형법득도적고계결과。
Assessing the welfare costs of inflation requires change Most monetary models in macroeconomics such as a sound understanding of the benefits of monetary ex- money-in-utility model, cash-in-advance model and shopping-time model are reduced-form models. By this we mean that they make assumptions, such as putting mon- ey in the utility function or imposing cash-in-advance constrains, that are presumably meant to stand in for a role of money that is not made explicit; The search theory of money, which developed in the last 20 years from the pionee- ring works of Kiyotaki and Wright, provides microfoundations for monetary economics based on search theory, with explicit descriptions of meetings, specialization, information, and so on. However, the the first generation of search models of money were based on assumptions that were too restrictive to deliver useful insights for the analysis of mo- netary policy~ Due to the work of Shi, Lagos and Wright and Molico, we can apply the search-theoretic models of money to get better understanding of the welfare costs of inflation and other monetary issues. In this paper we apply both welfare triangle method and search-theoretic model to study the welfares costs of inflation in China from 1992 to 2010. When using the traditional welfare triangle method developed by Bailey to esti- mate the welfare costs of inflation, which consists of computing the area underneath a money demand function, the form of money demand function is of great importance. So we use both double-log money demand function and semi- log money demand function to estimate China＇s money demand function. Based on unit root test and cointegration test, we draw the conclusion that double-log money demand function is more suitable for the description of the be- havior of money demand in China. Then we calculate the area underneath the double-log money demand function and get the estimate of the welfare costs of inflation in China. We show the result that on average agents would give up less than 0.92 of percent income to avoid 10 percent inflation. Many works have shown that the search-theoretic models spell out explicitly the benefits of monetary exchange for society and they provide both microfoundations for the demand for real money balances and a better measure of welfare. After an introduction to search-theoretic models, we adopt the search framework of Lagos and Wright, in which trades take place under two different market structures consisting of a decentralized market and a competitive market, to estimate the welfare costs of inflation in China. According to the results, we find that going from 10 per- cent to 0 percent inflation is worth 1.2 percent of income, which is much higher than previous estimates by other papers and the result in this paper by welfare triangle method. Given the different results by traditional welfare triangle method and search-theoretic model, we conclude that in comparison with the traditional welfare triangle method, the search-theoretic model has more concrete microfoun- dations and provides new insights for evaluating the welfare costs of inflation. So we choose the 1.2 percent of in- come as the welfare costs of inflation in China. Since we suffer much more from inflation and we often experience high inflation since the initiation of reform and opening up, we put up policy that the central government should keep the target level of inflation as low as possible, and should take active measures to fight inflation when it seems to show up. Additional extensions are worth considering. For instance, one should take into account of the degree of mar- ket competition when estimating the welfare costs of inflation, since the welfare costs of inflation are different in en- vironments with perfect competition and monopolistic competiton. Moreover, in this paper we treat all the people as a represent agent, which ignores the heterogeneity across the population. In fact the redistributive effects of inflation are different across different people, so we can extend the model to study the redistributive effects of inflation so as to see whether inflation benefits or hurts the poor people.