Asset Markets, Exchange Rates and Monetary Policy: Empirical and Theoretical Investigation

摘要题目：资产市场、汇率与货币政策：实证与理论研究。主要研究者：Charles Engel。第一个项目调查开放经济体中最优货币政策的性质，特别关注汇率失调的作用。在过去的二十年里，理论宏观经济学在基于优化家庭和企业微观经济行为的凯恩斯模型中，对通胀和失业的成本（或“产出缺口”——实际产出与充分就业产出水平之间的缺口）有了更好的理解。这项工作帮助政策制定者了解了影响通胀目标与产出缺口之间权衡的因素。恩格尔（Engel，《美国经济评论》，2011）最近的一篇论文使用了一个简单的凯恩斯主义框架来展示货币失调如何影响政策选择。在那篇论文中，当名义汇率的变动导致一个国家相对于另一个国家的消费者价格与这些国家生产和销售商品的相对成本不一致时，货币就会失调。凯恩斯模型的延伸表明，当经济体在全球高度一体化时，减少这些货币失调可能是货币政策的一个重要目标。该分析的一个缺点是，它从合作的角度考察了政策制定，即假设货币政策制定者在全球范围内进行合作，以实现所有国家家庭的最高福利。这个项目着眼于同样的问题，但当政策制定者不合作时。在合作下，任何货币失调都不利于世界福利，因为这意味着价格不能有效反映成本。然而，从非合作的角度来看，这些货币失调可能会改善一个国家的福利。汇率可以影响就业水平和通胀，但问题是，除了对产出和通胀的影响之外，汇率如何影响最优政策目标。例如，当生产者以进口消费者的货币定价时，对于任何给定的外国销售水平，以本国货币表示的收入将越低，本国货币越弱。不合作的政策制定者是否更喜欢某种程度的低估或高估？最终，当政策制定不合作时，全球福利是否会付出重大代价？第二个项目研究固定和浮动名义汇率国家的实际汇率行为（当使用汇率以共同货币表示价格时，每个国家的相对消费者价格水平）。这项研究将使用31个欧洲国家的消费者价格水平数据，研究欧元区国家实际汇率调整的决定因素，并将其与那些不使用欧元且实行浮动汇率的国家进行比较。一个普遍的信念——虽然没有得到现代宏观经济理论的充分支持，但在政策制定界得到广泛认同——是，与采用固定汇率或使用同一种货币的国家相比，浮动汇率能更快地调整任意两个给定国家之间的相对价格水平。事实上，当名义汇率浮动而商品价格调整缓慢时，实际汇率可能无法有效调整，因为名义汇率受到许多金融市场因素的影响，而这些因素可能导致一国相对于另一国的消费者价格在较长时间内上涨或下跌。也就是说，欧元/英镑名义汇率的波动可能导致实际的升值或贬值，而这与决定相对价格的商品的供求因素无关。实际价格和汇率的行为与有效实际汇率调整的预测进行了比较。第三个项目提出了一个汇率和短期存款或债券回报的模型。一个能够以本国货币借款的国家可能会通过贬值使其货币贬值，以减少其国际债务的实际价值。这种货币政策行为相当于部分违约，因为贷款人得到的是已经贬值的货币。在该模型中，借款人会因这种行为受到国际贷方的惩罚，即对未来的贷款收取惩罚性利率。这里的目的是模拟市场上普遍持有的观点，即一些货币，如美元，是“安全港”。在经济不确定时期，美元往往会升值。从规避风险的投资者的角度来看，在全球动荡时期将资产存放在美国可能是有道理的，但标准模型认为，这应该会推高这些资产（股票、债券和房地产）的美元价值，但不一定会增加美元的价值。这里的想法是，在动荡时期，一些国家的货币政策制定者可能更倾向于让本币贬值，而投资者认为美联储（fed）不太可能这么做。该模型或许还可以解释有关汇率和利率在短期和长期的行为的一些困惑。项目四提出了基于短期资产边际流动性价值的汇率新模型。这是为了解释2008年末和2009年初美元的升值，但也可能揭示正常时期回报率的决定因素。在危机期间，银行和其他金融机构发现，它们需要持有更多以美元计价的短期资产。银行需要持有流动资产，以弥合储户需求与贷款期限结构之间的不匹配，并为贷款组合的违约风险建立安全网。通常情况下，隔夜贷款市场是流动性的一个重要来源，但在危机期间，由于贷款机构开始怀疑银行必须提供的抵押品的价值，这一市场枯竭了。因此，对流动美元资产的普遍需求推高了美元的价值，尽管危机起源于美国。该计划是对这种流动性需求的决定因素进行实证测量，然后测试这些决定因素在危机期间和正常时期对货币价值的影响。

AbstractTitle: Asset Markets, Exchange Rates and Monetary Policy: Empirical and Theoretical Investigation Principal Investigator: Charles Engel The proposed work consists of four separate projects, though they are interrelated. The first project investigates the nature of optimal monetary policy in open economies, with a particular focus on the role of exchange-rate misalignment. Over the past two decades, theoretical macroeconomics has developed a better understanding of the costs of inflation and unemployment (or the "output gap" -the gap between actual output and the full-employment level of output) in Keynesian models based on optimizing microeconomic behavior of households and firms. This work has helped policymakers understand the factors that influence the tradeoffs involved in targeting inflation and the output gap. A recent paper, Engel (American Economic Review, 2011), used a simple Keynesian framework to show how currency misalignments can influence policy choices. In that paper, a currency is misaligned when the movements in nominal exchange rates lead consumer prices in one country relative to another to be out of line with the relative costs of producing and selling goods in those countries. That extension of the Keynesian model shows that reducing these currency misalignments can be an important target for monetary policy when economies are very integrated globally. A shortcoming of that analysis was that it examined policymaking from the cooperative standpoint - that is, assuming monetary policymakers cooperate globally to achieve the highest possible well-being of households in all countries. This project looks at the same question, but when policymakers do not cooperate. Under cooperation, any currency misalignment is detrimental to world welfare because it means that prices do not efficiently reflect costs. However, from a non-cooperative standpoint, these currency misalignments might improve one country's welfare. The exchange rate can influence the employment level and inflation, but the question is how the exchange rate might affect optimal policy goals beyond their effects on output and inflation. For example, when producers set prices in the importing consumer's currency, its revenue expressed in its own currency will be higher for any given level of foreign sales the weaker the domestic currency. Might the non-cooperative policymakers prefer some level of under- or over-valuation? Ultimately is there a significant cost to global welfare when policy is set non-cooperatively? The second project studies the behavior of real exchange rates (the relative consumer price levels in each country, when using the exchange rate to express prices in a common currency) in countries with fixed and floating nominal exchange rates. The study will use consumer price level data for 31 European countries to look at the determinants of real exchange rate adjustment for the countries with in the eurozone compared to adjustment in those countries that do not use the euro and have floating exchange rates. A common belief -not well supported by modern macroeconomic theory, but widely held in policymaking circles- is that floating exchange rates allow faster adjustment of relative price levels between any two given countries than if they have fixed exchange rates or share a common currency. In fact, when nominal exchange rates float but goods prices adjust slowly, the real exchange rate might not adjust efficiently because the nominal exchange rate is influenced by many financial market factors than can lead consumer prices to rise or fall for extended periods of time in one country relative to another. That is, drift in the euro/pound nominal exchange rate may cause real appreciation or depreciation that is unrelated to the supply and demand factors for goods that ought to determine this relative price. The behavior of actual prices and exchange rates are compared to predictions of efficient real exchange rate adjustment. The third project proposes a model of exchange rates and returns on short-term deposits or bonds. A country that can borrow in its own currency may debase its currency through depreciation in order to reduce the real value of its international debt. Such monetary policy behavior amounts to a partial default, because the lender is repaid with currency that has become devalued. In the model, the borrower is punished by international lenders for this behavior by being charged a penalty rate of interest on future loans. The aim here is to model the widely held view in the markets that some currencies, such as the U.S. dollar, are "safe havens". In times of economic uncertainty, the dollar tends to appreciate. From the standpoint of a risk averse investor, it might make sense to park assets in the U.S. during times of global turbulence, but standard models say that should drive up the dollar value of those assets (stocks, bonds, property) but not necessarily increase the value of the dollar. The idea here is that during times of turbulence, monetary policymakers in some countries may be more prone to devalue their currencies, while investors believe the Federal Reserve is less likely to do that. The model perhaps can also explain some puzzles regarding the behavior of exchange rates and interest rates in the short run and long run. The fourth project proposes a new model of exchange rates based on the marginal liquidity value of short-term assets. This is meant to explain the appreciation in the dollar in late 2008 and early 2009, but may also shed light on the determinants of returns during normal times as well. During the crisis, banks and other financial institutions found that they needed to hold greater volumes of short-term dollar denominated assets. Banks need to hold liquid assets to bridge mismatches between depositor demands and the maturity structure of loans, as well as a safety net against default risk on the loan portfolio. Normally the overnight loan market provides an important source of liquidity, but that market dried up during the crisis as lenders began to doubt the value of collateral that banks had to offer. So, a general demand for liquid dollar assets drove up the value of the dollar, even though the crisis originated in the U.S. The plan is to derive empirical measures of the determinants of this liquidity demand, and then test the effects of those determinants on currency values both during the crisis and during normal times.