Sapienza University of Rome Department of economics and law Advanced Monetary Theory and Policy EPOS 2013/14 Macro theory: A quick review Giovanni Di Bartolomeo giovanni.dibartolomeo@uniroma1.it
Theory: Big view Keynesian Cross Theory of Liquidity Preference IS curve LM curve IS-LM model Phillips curve Business Cycle fluctuations AD curve AS curve Model of AD/AS
The downward-sloping AD curve The aggregate demand curve shows the relationship between the price level and the quantity of output demanded. It can be derived from the IS/LM model P A B AD Y
Intuition: Deriving the AD curve r LM(P 2 ) Intuition for slope of AD curve: r 2 LM(P 1 ) P (M/P ) LM shifts left r r 1 P Y 2 Y 1 IS Y I Y P 2 P 1 A B AD Y 2 Y 1 Y
Shifting the AD curve An increase in the money supply shifts the AD curve to the right P C D AD 2 AD Y
Intuition: Monetary policy and the AD curve The Fed can increase aggregate demand: r r 1 LM(M 1 /P 1 ) LM(M 2 /P 1 ) M LM r I shifts right r 2 P Y 1 Y 2 IS Y Y at each value of P P 1 C D AD 2 Y 1 Y 2 AD 1 Y
Neoclassical view The economy works well on its own Invisible hand : the idea that if there are free markets and individuals conduct their economic affairs in their own best interests, the overall economy will work well Wages and prices adjust rapidly to get to equilibrium Equilibrium: a situation in which the quantities demanded and supplied are equal Changes in wages and prices are signals that coordinate people s actions Result: Government should have only a limited role in the economy
Quantitative theory in the AD/AS model Neo-classic economists: Money is a veil and monetary policy useless P LRAS In the long run, this raises the price level P 2 P 1 An increase in M shifts AD to the right. AD 1 AD 2 but leaves output the same. Y Determined by real factors Y
Do markets work?
The effects of a negative demand shock AD shifts left, depressing output and employment in the short run P LRAS Over time, prices fall and the economy moves down its demand curve toward natural employment P B A SRAS P 2 C AD 1 AD 2 Y 2 Y Y JM Keynes
Keynesian view The Great Depression: Classical theory failed because high unemployment was persistent Keynes: Persistent unemployment occurs because wages and prices adjust slowly, so markets remain out of equilibrium for long periods Indeed Keynes (and Kalescki) argues that prices and wages may do not adjust at all Anyway the long run is very far: In the long run we are all dead (J.M. Keynes) Conclusion: Government should intervene to restore full employment
Keynesian world (fixed prices) Keynesian economists: Money is not neutral (it affects output) P LRAS In the short run when prices are sticky an increase in aggregate demand P SRAS AD 2 AD 1 causes output to rise. Y 1 Y 2 Y Y
Samuelson and Solow in the 1960s Between (fully) flexible and fixed price models P AS B P 2 P 1 A AD 2 JFK Y 1 Y 2 AD 1 Y
AD/AS and the policy menu P The Phillips Curve is just an alternative way of describing the Aggregate Supply Curve AS Policy menu 106 B 6 B 102 A AD2 2 A AD1 0 7,500 8,000 Y 0 4 7 u=7% Y=8000 u=4% Y=7500 u Unemployment rates associated to output values (Okun Law)
The estimated Phillips curve: - a u + C Inflation rate ( ) Paul Samuelson 10 8 C 6 4 2 In the interview, Robert Solow said Paul Samuelson asked me when we were looking at these diagrams (of inflation and unemployment) for the first time, Does that look like a reversible relation to you? What he meant was Do you really think the economy can move back and forth along a curve like that? And I answered Yeah I m inclined to believe it, and Paul said Me too 1967 1968 1966 1965 1962 1964 1963 1961 -a 0 1 2 3 4 5 6 7 8 9 10 Unemployment rate (u) Robert Solow
Government loss function: L = b 2 +u 2 Inflation rate ( ) L 3 >L 2 >L 1 >L 0 10 8 L 3 6 L 2 4 L 1 First best [,u] =[0,0] 2 L 0 0 1 2 3 4 5 6 7 8 9 10 Unemployment rate (u)
Optimal policy (Tinbergen-Theil approach ) Inflation rate ( ) optimal policy rule 10 8 L 3 6 E L 2 First best 4 2 L 1 L 0 A B 0 1 2 3 4 5 6 7 8 9 10 Unemployment rate (u)
Formal representation of the problem The Government (Central bank) s flexible-target problem Min L = βπ 2 + u 2 s.t. π = αu + C How can we solve it? (many ways) By Lagrangian Min *π,u,l+ L = βπ2 + u 2 + l αu + C π Or by substitution Min *π,u,l+ L = βπ2 + C α π α 2 (find π and then u by PC) or Min *π,u,l+ L = βπ2 + αu + C 2 (find u and then π by PC) Or
Optimal policy rule (derivation) The Government (Central bank) s problem Min *π+ L = βπ 2 + u(π) 2 s.t. u(π) = C α π α The operative instrument is π (it chooses M, moving the AD, to get a certain P (i.e. π) on the AS, Y and u follows) First order condition 2βπ + 2u C(π) u = 0 i.e. optimal policy rule (Government minimize the cost): βπ 1 1 u = 0 π = u α αβ By using the optimal monetary policy rule and the PC we obtain the equilibrium for and (a linear system of two equations in two unknowns)
In the 1970s something changes Stagflation; both are u high (out of PC?) Inflation rate ( ) 10 8 1974 19801981 1979 1978 1975 6 4 2 0 1977 1973 1969 1971 1970 1968 1972 1967 1966 1965 1962 1964 1961 1963 1976 1 2 3 4 5 6 7 8 9 10 Unemployment rate (u)
The 1970s oil shocks P LRAS The oil price shock shifts SRAS up, causing output and employment to fall In absence of further price shocks, prices will fall over time and economy moves back toward natural level of employment P 2 P 1 Stagflation B A Y 2 Y SRAS 2 SRAS 1 AD Y
Optimal policy and an (observed) shock Inflation rate ( ) 10 C+e 8 6 4 2 0 1974 C 1977 1973 1969 1971 A 1970 1968 1972 1967 1966 1965 1962 1964 1961 1963 optimal policy rule 19801981 1975 π = 1 αβ u 1979 1978 1976 - a u + C + e 1 2 3 4 5 6 7 8 9 10 Unemployment rate (u)
Friedman s 1968 AEA Presidential Address Milton Friedman argued the following Theory (his theory) predicts no stable relationship between unemployment and inflation According to his theory, policymakers do face a shortterm tradeoff between unemployment and inflation due to the private sector s failure to quickly adapt to changing environments long term costs to policymakers of exploiting short term tradeoff. If policymaker generates temporarily low unemployment by inflating, in future, higher and higher unemployment rates will be associated with each level of inflation. (Phillips curve shifts out) This argument was formalized and refined by Lucas
From the short to the long run ( M > 0) An attempt to increase Y over LRAS P LRAS P 2 P C A B SRAS AD 2 AD 1 Y Y 2 Y
Expectations: Phillips Curve and SRAS Do shocks only matter? How does the economy moves from the short to the long run position? The role of expectations!!! SRAS curve: Output is related to unexpected movements in the price level e SRAS: Y Y + ( P - P ) + Phillips curve: Unemployment is related to unexpected movements in the inflation rate Phillips curve: e e n - a( u - u ) + e
The Phillips Curve and SRAS equivalence e SRAS: Y Y + ( P - P ) + e n Phillips curve: - a( u - u ) + e See next slide e Y Y + ( P - P )- Y -Y ( P - P - P + P )- e -1-1 N e - L ( u -u ) ( - ) - P e LP - N u - u + e N - a u - u + e
Note Y P and L P are potential output and (full) employment, thus if Y=Y P and L=L P, then u = LF L LF = 0. But in the long run there are some distortions thus the long rate (natural) output (Y or Y N ) and employment (L N ) are lower than potentials, and u N = LF L N LF Consider the Okun law: Y = θl, then We have that i.e. > 0. Y Y = θ L L N = θ L LF L N + LF = = LFθ L LF LF L N LF LF Y Y = LFθ u u N
Key Questions How are expectations formed? How fast do they adjust? Phillips curve: e n - b( u - u ) + Two theories Adaptive expectations (Friedman, ) Rational expectations (Sargent, Lucas, )
Friedman and Phelps The Phillips curve states that depends on cyclical unemployment: the deviation of the actual rate of unemployment from the natural rate supply shocks It can also depend on The expected inflation rate, e (for instance, in wage bargaining operators are interested in the real wage so AS should depend on the expected real wage) Note that as long as is the operational instrument of the central bank, one can think that e should also depend on the expectations about monetary policy! Friedman assume that e = -1 (lagged inflation)
Friedman Phillips Curve Friedman adjustment based on past price to form current expectations, e = -1, it follows 1 - a n ( u - u ) + e - Consider an increase of money growth, not all producers adjust prices (imperfect information, no one knows if the increase in demand is relative to his product or to all), output and prices increase short run But afterward, one that everybody realizes that the observed increase in the demand was general all will adjust prices long run As results output will finally not increase and inflation increase will be instead permanent
Friedman and adaptive expectation Friedman adjustment based on past price to form current expectations P LRAS SRAS 2 SRAS 1 P 3 C B P 2 P 1 A AD 2 AD 1 Y Y 2
Expectation-driven shifts in the Phillips curve People adjust their expectations over time, so the tradeoff only holds in the short run e n - b( u - u ) + e 2 e 1 + e + e B A C E.g., an increase in e shifts the short-run P.C. upward. So if the central bank tries to increase, in the long run the effect will be offset by an equal change in e n u u
Policy menu in the long-run AS/AD Model Phillips Curve P LRAS LR Policy menu P 2 B P 1 AD 2 A AD 1 0 Y N Y 0 Natural rate of unemployment u
Short- and long-run Phillips curves Inflation rate ( ) Long run Phillips curve 10 8 1974 19801981 1979 1978 1975 6 4 2 0 1977 1973 1969 1971 A 1970 1968 1972 1967 1966 1965 1962 1964 1961 1963 1976 Short rune Phillips curves 1 2 3 4 5 6 7 8 9 10 Unemployment rate (u)
The Lucas critique The Phillips curve states that depends on cyclical unemployment: the deviation of the actual rate of unemployment (u) from the natural rate (u N ) supply shocks, e Thus, - a u + C implies C= a u N + x + e Estimations from the past give unbiased values for a and u N, but what does it occur if the shifter contains a policy related term x( )? Estimating the theoretical equation (bold are estimated parameters by OLS): - a u - u N ) + e We obtain a biased estimation as u N is correleted with
Neo-classical economics New Classical Theories were an attempt to explain the apparent breakdown in the 1970s of the simple inflationunemployment trade-off predicted by the Phillips curve They argue that traditional models have assumed that expectations are formed in naïve ways But naïve expectations are inconsistent with the assumptions of microeconomics/rationality Expectations are rational (RE) so adjustment is immediate No effect for monetary policy (unless surprise but then usefulness or even counter-productive) But we do not need it as markets are efficient (on average)
Neo-classical economics Micro-foundation No Lucas critique (based on deep parameters, which do not depend on the policy regime) Assumptions Perfect markets (flexible prices) Rational expectations hypothesis (REH) Expectations are rational so adjustment is immediate RE disequilibrium exists only temporarily as a result of random, unpredictable shocks No effect for monetary policy (unless surprise, but then it is usefulness or even counter-productive) But, on average, all markets clear and there is natural employment. No need for government stabilization
Rational expectations Naïve expectations: inconsistent with rationality as they are associated with systematic errors The REH assumes that people use available information efficiently, including how the economy works (Muth). People know the true model of the economy and that they use this model to form their expectations of the future. By true model we mean a model that is on average correct in forecasting inflation People can make mistakes, but they do not make systematic forecasting errors The forecast errors of expectations will be random with a mean of zero, unrelated to those made in previous periods, revealing no discernible pattern, and have the lowest variance compared to other forecasting methods
From New Classicals to RBC and NK theory The New Classical Economics challenged Keynesian theory, and stimulated the development of Real Business Cycle (RBC) and New Keynesian (NK) Theory RBC theory accepts the REH, but views cycles arising in frictionless, perfectly competitive economies with complete markets. It argues that cycles arise through the reactions of optimizing agents to real disturbances, such as random changes in technology or productivity NK theory accepts the REH, but emphasizes the importance of imperfect competition, costly or impeded price adjustments, and externalities. It argues that nominal shocks are the predominant cause of business cycles