Collaborative Research: Inflation Dynamics During the Financial Crisis

从宏观经济的角度来看，2007-09年金融危机和随后的缓慢复苏中最令人困惑的一个方面是通胀的相对稳定。尽管2008年下半年产出大幅收缩，过去几年资源持续处于高位，但美国只经历了适度的通缩压力——无论是在批发还是零售层面——这种行为很难用总通胀决定的标准模型来解释。本项目研究2008年下半年金融危机最严重时期金融环境的大规模紧缩在多大程度上可能有助于解释为什么通胀在金融危机和随后的复苏期间没有像预期的那样下降。潜在的假设是，一些公司可能会战略性地设定低价来吸引顾客，这是“顾客-市场理论”所暗示的行为，该理论强调价格设定是一种投资形式，可以建立未来的顾客基础。根据这一理论，企业在正常情况下会相对于竞争对手保持较低的价格，以吸引新客户。反过来，这意味着较低的流动溢价，因此较低的现金流。然而，在广泛的金融危机时期，当外部融资变得昂贵甚至无法获得时，财务脆弱的公司将通过提高价格来牺牲他们的客户群，以稳定当前的收入和增强他们的流动性头寸。因此，在严重的金融危机时期，尽管经济活动下降，出现了很大程度的资源利用不足，但通货膨胀实际上可能会上升。面对不利的需求冲击，通胀和产出向相反方向移动的可能性，也挑战了有关金融危机期间货币政策适当行动路线的传统思维。为了从经验上检验这一假设，该项目将企业层面的财务数据与从劳工统计局（BLS）收集的机密价格数据计算出的相应企业层面的通货膨胀率合并在一起。初步结果表明，金融因素对2007-09年金融危机期间的通胀动态有显著影响。与这一假设相一致的是，财务状况“较弱”的公司在2008年下半年提高了价格，而财务状况“较强”的公司则降低了价格，以应对总需求的下降。迄今为止，这一机制对总通货膨胀动态的影响在很大程度上仍未得到探索，本研究提供了一些第一个系统的证据，证明金融状况如何通过激励企业以保持其流动性和资产负债表能力的方式设定价格来影响企业的定价行为。就这一渠道在数量上的重要性而言，本项目的分析对衡量金融危机期间破坏性大规模通货紧缩的可能性具有指导意义。它还对货币政策的实施具有重要意义，因为它认为政策制定者在决定名义利率的未来走向时，应该考虑到企业资产负债表质量变化与它们的定价行为之间的相互作用。除了开发匹配算法，将劳工统计局的研究人员数据库与外部企业层面的数据集联系起来，该项目还提供了一种基于GPS坐标的当地银行信贷供应状况的新测量方法，该方法可能与未来的劳工统计局研究相结合，为金融状况和企业行为之间的空间相互作用开辟了一套全新的研究可能性。

From a macroeconomic perspective, one of the most puzzling aspects of the 2007-09 financialcrisis and the ensuing slow recovery is the relative stability of inflation. Despite a large and sharpcontraction in output in the second half of 2008 and the persistently high level of resource slackover the past several years, the U.S. experienced only a modest degree of disinflationarypressures-both at the wholesale and retail levels-behavior that is difficult to explain withcanonical models of aggregate inflation determination. This project studies the extent to whichthe massive tightening of financial conditions at the height of the crisis in the latter part of 2008may help explain why inflation did not fall as much as expected during the financial crisis and thesubsequent recovery. The underlying hypothesis posits that some firms may strategically set lowprices to attract customers, behavior implied by the "customer-markets theory," whichemphasizes the idea that price setting is a form of investment that builds the future customer base.According to this theory, firms will in normal times maintain low prices relative to theircompetitors in order to attract new customers. In turn, this implies low current mark-ups andhence lower cash flows.However, in times of widespread financial distress-when external finance becomes prohibitivelycostly or even unavailable-financially vulnerable firms will sacrifice their customer base byraising prices to stabilize current revenues and bolster their liquidity positions. Thus in periods ofacute financial distress, inflation may actually rise despite a drop in economic activity and theemergence of a significant degree of resource underutilization. The possibility that inflation andoutput move in opposite directions in response to an adverse demand shock also challengesconventional thinking regarding the appropriate course of action for monetary policy duringfinancial crises.To examine this hypothesis empirically, the project merges firm-level financial data with thecorresponding firm-level inflation rates computed from the confidential price data collected bythe Bureau of Labor Statistics (BLS). Preliminary results indicate that financial factors had asignificant effect on inflation dynamics during the 2007-09 crisis. Consistent with the hypothesis,financially "weak" firms raised prices in the latter part of 2008, whereas their financially "strong"counterparts lowered prices in response to a drop in aggregate demand. To date, the implicationsof this mechanism for aggregate inflation dynamics have remained largely unexplored, and thisstudy provides some of the first systematic evidence on how financial conditions influence theprice-setting behavior of firms through the incentives to set prices in a manner that preserves theirliquidity and balance sheet capacity.To the extent that this channel is quantitatively important, the analysis in this project has a directbearing on gauging the likelihood of destructive large-scale disinflations during financial crises. Italso has important implications for the conduct of monetary policy by arguing that policy makersshould take into account the interaction between changes in the quality of the firms' balancesheets and their price-setting behavior when deciding on the future course of nominal interestrates. In addition to developing the matching algorithm that links BLS' researcher databases withoutside firm-level data sets, the project also provides a new measure of local bank credit-supplyconditions based on the GPS coordinates that may be used in conjunction with future BLSstudies, opening up a whole new set of research possibilities on the spatial interaction betweenfinancial conditions and firm behavior.