1、This PDF-file contains a composition of slides that were used for courses taught in our MIEPP-program. In total, we present slides from 6 different courses to you. We thus hope to give you a short impression of the diversity of our curriculum with regard to content as well as methods. Should you req
2、uire further information on the courses, please consult the MIEPP-webpage and do not hesitate to contact the particular instructors directly with specific requests. Sample Course Slides Sample Course Slides No. 1 The following slides are taken from the course: International Macroeconomics Subsection
3、: Exchange Rates The course instructor is: Prof. Philipp Harms Harms/ I nternational Macroeconomics 6-1 6. Exchange Rates 6.1. Motivation 6.2. Basic concepts and definitions 6.3. The real exchange rate and purchasing power parity 6.4. Interest parity conditions 6.5. The monetary model of exchange ra
4、te determination 6.6. The Dornbusch overshooting“ model 6.7. Fundamental determinants of the real exchange rate 6.8. Summary and outlook Harms/ I nternational Macroeconomics 6-2 The monetary model of exchange rate determination Structure and assumptions: Small open economy with perfect international
5、 capital mobility : Real interest rate determined by world capital markets: (6.8) Flexible exchange rate ( = Central bank does not intervene on foreign exchange markets). Classical dichotomy: Real output, consumption etc. are exogenous. FtHt rr=Harms/ I nternational Macroeconomics 6-3 The monetary m
6、odel of exchange rate determination Structure and assumptions: Real money demand is given by (6.9) with mD,Ht: Log. of domestic nominal money demand pHt: Log. of domestic price level yHt: Log. of domestic real income (exogenous!) iHt: Nominal interest rate paid on assets held in domestic currency be
7、tween period t and t+1. , 0 - Permanent money market equilibrium: HtHtHtHDt iypm =,HtHStHDt mmm =,Harms/ I nternational Macroeconomics 6-4 The monetary model of exchange rate determination Structure and assumptions: Nominal money supply is exogenous. Rational expectations: Expectations about x = Con
8、ditional expected value based on all information available at t (including the model). jttejt xx+=EEt : nominal exchange rate Et : expectations operator Harms/ I nternational Macroeconomics 6-5 The monetary model of exchange rate determination Structure and assumptions (contd.): (Absolute) purchasin
9、g power parity (PPP): (6.10) Interpretation: If measured in the same currency, price levels are identical across countries. Rationale: Price indices are based on identical goods baskets and the law of one price holds for all goods. (More about this later) FttHtHtFttt PEPPPEQ = 1Harms/ I nternational
10、 Macroeconomics 6-6 The monetary model of exchange rate determination Structure and assumptions (contd.): Absolute Purchasing Power Parity (PPP) (contd.) Writing the PPP condition in logs yields (6.11) Differencing (6.11) yields the relative version of PPP (6.12) FtHtt ppe =( )FtFttHtHttttt ppppee =
11、 + 111 EEEHarms/ I nternational Macroeconomics 6-7 The monetary model of exchange rate determination Structure and assumptions (contd.): The Fisher equation: Nominal interest rate i as sum of real interest rate r and expected inflation: (6.13a) (6.13b) Combining (5.8), (5.12) and (5.13) yields the U
12、IP (6.14) HtHttHtHt ppri+=+1EFtFttFtFt ppri+=+1EtttFtHt eeii+=+1EHarms/ I nternational Macroeconomics 6-8 The monetary model of exchange rate determination Determining the exchange rate Using (6.9), (6.11 ) and (6.14) yields (6.15) Forward-iteration of (6.15) yields (6.16) 1E11 1 += ttFtHtFtHtt eiyp
13、me TtTTFsHsFsHststtst eiypme E1limE111 += = Harms/ I nternational Macroeconomics 6-9 The monetary model of exchange rate determination Determining the exchange rate (contd.) Note: The “bubble” term is zero unless the nominal exchange rate explodes/implodes ( grows infinitely large in absolute value)
14、. In case of a speculative bubble, this is exactly what happens: Value of asset (here: currency) grows only because market participants expect to grow in future. TtTTFsHsFsHststtst eiypme E1limE111 += = Fundamental value of the nominal exchange rate Bubble“ term Harms/ I nternational Macroeconomics
15、6-10 The monetary model of exchange rate determination Determining the exchange rate (contd.) Assuming that the nominal exchange rate does not explode, we can set the last term on the RHS of (6.16) equal to zero, which yields (6.17) FsHsFsHststtst iypme += =E11 1Harms/ I nternational Macroeconomics
16、6-11 The monetary model of exchange rate determination Interpretation The current nominal exchange rate depends on current and expected future values of . . the money supply (+) . the level of real income (-) . the nominal foreign interest rate (+) . the foreign price level (-) Future values are “di
17、scounted” using the factor /(1+). Harms/ I nternational Macroeconomics 6-12 The monetary model of exchange rate determination Interpretation (contd.) Why does the future matter for the current exchange rate? Consequence: Expectations about higher nominal exchange rate in future result in higher nomi
18、nal exchange rate in current period. HtHtHtHt iymp += Price level increases in the nominal interest rate (Effect of interest rate on real money demand). FtHtt ppe = PPP: Nominal exchange rate determined by domestic price level. 1+= tttFtHt eeii E UIP: Nominal interest rate increasing in expected dep
19、reciation of domestic currency. Harms/ I nternational Macroeconomics 6-13 The monetary model of exchange rate determination Merits of the monetary model: What have we learned? Key message: Current nominal exchange rate depending on current and expected future values of exogenous variables (money sup
20、ply, output, interest rates etc.) : Importance of expectations in determining exchange rates Exchange rate as the price of an asset (domestic currency) : Like other asset prices, exchange rate determined by expectations about the future. Harms/ I nternational Macroeconomics 6-14 The monetary model o
21、f exchange rate determination Figure 6.10: Worldwide turnover on forex markets and worldwide exports of goods and services Source: BIS, World Bank (WDI) 0500010000150002000025000198919901991199219931994199519961997199819992000200120022003200420052006200720082009Worldwide value of exports (billions o
22、f US dollars) 0500100015002000250030003500400045001989 1992 1995 1998 2001 2004 2007 2010Worldwide turnover on foreign exchange markets per day (billions of US dollars) Harms/ I nternational Macroeconomics 6-15 The monetary model of exchange rate determination Taking the monetary model to the data W
23、ith two (structurally identical) countries, the fundamental value of the nominal exchange rate is given by To save notation, we define the monetary fundamentals ( )+=+ += jtjtjtjttjjt yymme E1110+ + jtjtjtjtjt yymm Harms/ I nternational Macroeconomics 6-16 The monetary model of exchange rate determi
24、nation Taking the monetary model to the data (contd.) We assume that the monetary fundamentals follow an AR(1): It can be shown that in this case Interpretation: Changes in nominal exchange rate driven by changes in fundamentals. Unless = 1: Nominal exchange rate less volatile than fundamentals. 11
25、+ += ttt ( )( )tttt ee += + 11 111Harms/ I nternational Macroeconomics 6-17 The monetary model of exchange rate determination Figure 6.11: Monetary Fundamentals and the Nominal Exchange Rate Source: IMF (IFS) and own calculations -25%-20%-15%-10%-5%0%5%10%15%20%1975 1979 1983 1987 1991 1995 1999 200
26、3 2007Percentage change of monetary fundamentals (money supply and GDP)Depreciation of the Swiss Franc against the US dollarHarms/ I nternational Macroeconomics 6-18 The monetary model of exchange rate determination Shortcomings of the monetary model: What seems wrong and what is missing ? The monet
27、ary model can be interpreted as a nominal extension of the intertemporal model presented in Sections 3 and 4. Taking the real magnitudes determined in the intertemporal model as given, it is used to determine the nominal exchange rate. Hence, it can be accused of making the same (heroic) assumptions
28、 as the intertemporal model: Perfect capital mobility and risk-neutrality Permanent PPP Monetary neutrality Harms/ I nternational Macroeconomics 6-19 The monetary model of exchange rate determination Shortcomings of the monetary model: What seems wrong and what is missing ? (contd.) The monetary mod
29、el suggests that the exchange rate is at most as volatile as monetary fundamentals. However, observed nominal exchange rates are more volatile than monetary fundamentals (excess volatility“) The monetary model is based on the notion that PPP holds at every point in time. However, we observe persiste
30、nt deviations from PPP. Sample Course Slides No. 2 The following slides are taken from the course: Monetary Theory and Policy Subsection: Welfare Cost of Inflation The course instructor is: Dr. Leopold von Thadden Motivation Money demand specifications and empirical evidence Welfare cost of inflatio
31、n CommentsI Motivation! Some features of fiat money regimes:private opportunity cost of holding real balances (m) depends on thelevel of the short-term nominal interest rate (i)social opportunity cost of providing money is essentially zero, ie thecentral bank can make the amount of money in circulat
32、ion arbitrarilylarge (and thereby depress i)the wedge between the private and social cost at positive interest ratescreates an inefficiencymoreover, in equilibrium, the nominal interest rate, the inflation rateand the quantity of money in circulation are linked to each other! These features give ris
33、e to two types of (closely linked) policy questionswhich have been addressed in monetary economics for centuries:1) From a positive perspective: How large is the welfare cost of inflation?2) From a normative perspective: What is the optimal rate of inflation?2 / 33Motivation Money demand specificati
34、ons and empirical evidence Welfare cost of inflation CommentsI Motivation! The MIU-model offers a transparent and tractable general equilibriumperspective, ie it is a widely used starting point to address these questions,qualitatively and quantitatively(in particular, see Lucas, 2000):! Why is the M
35、IU model a good model to analyze these questions? Because:The costs of inflation (via the interest rate) are clearly captured by theexpressionum(ct, mt) = it1 + ituc(ct, mt) (1)Steady-state superneutrality implies that we dont have to worry aboutthe real side of the economy when comparing steady sta
36、tes with differentinflation and interest ratesThe flow objective u(c, m) offers a simple and direct metric for welfarecomparisons3 / 33Motivation Money demand specifications and empirical evidence Welfare cost of inflation CommentsI MotivationQuestion 2: What is the optimal rate of inflation?! In th
37、e basic MIU model the answer to this question is ratherstraightforward, since monetary policy instruments (i or ) do not affect thesteady-state value of c, only of m! Money growth rule (via ) : Optimality requires u(c,m) = u(c,m)m m = 0,or u(c,m)m = 0, implying in eqn (1)i = 0! This is the Friedman
38、rule which implies (using 1 + i 1 + r + pi)pi x0r,i.e. the (long-run) optimal inflation rate is a rate of deflation approximatelyequal to the return on capital and government bonds (Friedman, 1969)4 / 33Motivation Money demand specifications and empirical evidence Welfare cost of inflation CommentsI
39、 MotivationFriedman rule: an important qualificationThis reasoning is not yet a basis for policy advice, since the interestrate i (and indirectly pi) is the only distortionary policy instrumentRecall from the model set-up: government has access to lump-sumtransfers/taxes , making distortionary seign
40、iorage revenues costlyThe discussion about the optimality of the Friedman rule becomes onlymeaningful when we introduce distortionary taxesThen, as argued by Phelps (1973), the Friedman rule may well break down! We will return to Question 2 in detail in Part II which covers theFriedman vs. Phelps de
41、bate, assuming distortionary taxation5 / 33Motivation Money demand specifications and empirical evidence Welfare cost of inflation CommentsI MotivationLet us return toQuestion 1: How large is the welfare cost of inflation?! Lucas (2000) uses the MIU model to provide quantitative guidance forthe like
42、ly range of welfare costs of inflation! The methodology proposed by Lucas has been influential since it exploitsthe general equilibrium dimension of the MIU model, as opposed to partialequilibrium estimates that were traditionally used (Bailey 1956)To reproduce the findings from Lucas, we proceed in
43、 2 steps:Step I): Two types of money demand specifications and discussion of theempirical evidenceStep II): Partial vs. general equilibrium measures of welfare costs of inflation6 / 33Motivation Money demand specifications and empirical evidence Welfare cost of inflation CommentsII Money demand spec
44、ifications and empirical evidence! Any welfare measure of the costs of inflation depends critically on theform of the money demand equation and the sensitivity of moneydemand to the opportunity cost of holding real balances! In the basic MIU model these issues are directly related to thespecificatio
45、n of the utility function which drives the shape ofum(ct, mt) = it1 + ituc(ct, mt)!The empirical literature commonly distinguishes between log-logspecifications as opposed to semi-log specifications of money demand7 / 33Motivation Money demand specifications and empirical evidence Welfare cost of in
46、flation CommentsII Money demand specifications and empirical evidenceCES utility function between c and m :! Consider the following CES utility function which displays a constantelasticity of substitution between consumption and real balances:u(ct, mt) = ac1x0bt + (1x0a)m1x0bt 11x0b with: 0 0 (and b
47、 6= 1)(2)! This function satisfiesum(ct, mt)uc(ct, mt) = (1x0aa )(ctmt )band, when combined with the first-order optimality conditionum(ct, mt) = it1 + ituc(ct, mt),it gives rise to the money-demand functionmt = (1x0aa ) 1b x1( it1 + it)x01b x1ct (3)8 / 33Motivation Money demand specifications and e
48、mpirical evidence Welfare cost of inflation CommentsII Money demand specifications and empirical evidenceLog-log specification:! In the empirical literature it is common to rewrite equations of type (3), iemt = (1x0aa ) 1b x1( it1 + it)x01b x1ctby taking logs, leading tolog(m) = 1b log(1x0aa ) + log(c)x0 1b log( i1 + i )
