Business cycle. Giovanni Di Bartolomeo Sapienza University of Rome Department of economics and law

Transcription

1 Sapienza University of Rome Department of economics and law Advanced Monetary Theory and Policy EPOS 2013/14 Business cycle Giovanni Di Bartolomeo

2 US Real GDP Real GDP (in 2000 dollars)

3 Long-run growth Secular long-run growth, or long-run growth, is the sustained upward trend in aggregate output per person over several decades A country can achieve a permanent increase in the standard of living of its citizens only through long-run growth. So a central concern of macroeconomics is what determines long-run growth The word secular is used to distinguish long-run growth from the expansion phase of business cycles, which lasts less than 5 years on average. Secular long-run growth refers to the growth of the economy over several decades

4 Economic Growth and Fluctuations Figure 2 shows the long-term growth trend and cycles

5 Total Output Business Cycle Phases of the Business Cycle Expansion Recession Expansion Peak Trough Secular growth trend 0 Jan.- Mar Apr.- June July- Sept. Oct.- Dec. Jan.- Mar Apr.- June July- Sept. Oct.- Dec. Jan.- Mar Apr.- June

6 Business Cycle (NBER) Recession = two consecutive quarters (six months) of negative growth in real GDP Peak = highest point before a recession Trough = the lowest point at the end of a recession and before an expansion Expansion = the period between the end of a recession and the next peak (real GDP grows) Recovery = the very beginning of an economic expansion Boom = an extremely fast increase in output, usually near the end of an expansion Depression = a very long and low recession

7 Business Cycle Dates Recessions in the US The US has had 13 Recessions since the great depression Peak Trough Contraction (peak to trough) Duration (In Months) Expansion (Previous trough to this peak) August 1929 March May 1937 June Feb 1945 Oct Nov 1948 Oct July 1953 May Aug 1957 April April 1960 Feb Dec 1969 Nov Nov 1973 March Jan 1980 July July 1981 Nov July 1990 March March 2001 Nov December 2007 June Average Cycle (Peak from previous peak)

8 Business cycle facts All business cycles have features in common in terms of the cyclical behavior of economic variables: What direction does a variable move relative to aggregate economic activity? Pro-cyclical: in the same direction Countercyclical: in the opposite direction A-cyclical: with no clear pattern What is the timing of a variable's movements relative to aggregate economic activity? Leading: in advance Coincident: at the same time Lagging: after

9 Correlation Variables X and Y are said to be positively correlated if X is likely to increase when Y increases X is likely to decrease when Y decreases Variables X and Y are said to be negatively correlated if X is likely to decrease when Y increases X is likely to increase when Y decreases Positive or negative correlation can be illustrated on a time series plot (plot X and Y against time) or on a scatter plot (plot X against Y for different observations)

10 Two graphical ways of illustrating correlation time series plot scatter plot

11 Leading and Lagging variables x is a leading variable, (changes in GDP follow changes in x) x helps to predict GDP y is a lagging variable, (changes in y follow changes in GDP) GDP helps to predict y If a variable neither leads nor lags, it is called a coincident

12 Index of leading economic indicators GDP index of leading economic indicators is a function of the following variables: Average workweek Initial unemployment claims New orders for consumer goods Vendor performance Plant and equipment orders Building permits An increase in unfilled durable orders Stock prices (S&P 500) Money supply Index of consumer expectations The government might use Index of leading economic indicators to predict recessions

13 Index of leading economic indicators The Index of Leading Economic Indicators: (a) is positively correlated with GDP (b) leads GDP (c) is slightly more volatile than GDP

14 Growth rates of Real GDP, Consumption Percent change from 4 quarters earlier Average growth rate Shaded areas indicate recessions Real GDP growth rate Consumption growth rate

15 GDP, Consumption and Investment Percent change from 4 quarters earlier Investment growth rate Real GDP growth rate Consumption growth rate

16 Natural log of per capita real GNP Actual GDP and trend (secular log-run growth)

17

18 Deviations from trend in real GNP Business cycle

19 Deviation (%) of GDP from a HP trend The recessions are pretty easy to spot! gen-57 gen-67 gen-77 gen-87 gen-97 gen

20 Unemployment Rate Unemployment Unemployment rises (falls) in recessions (expansions) While the average unemployment rate (excluding recessions) has been around 5% since 1957, the average unemployment rate during recessionary periods averages around 7% gen-57 gen-67 gen-77 gen-87 gen-97 gen

21 Output gap and productivity The correlation between productivity and real GDP is high. This indicates that productivity is highly pro-cyclical Cyclical Variation Productivity (x100) (Real) Output Gap

22 GDP and labor productivity Labor productivity = output / employment: (a) is positively correlated with GDP (corr. coeff. = 0.83) (b) coincides with GDP (c) is less volatile than GDP (std(c)=0.63*std(gdp))

23 GDP and employment Employment: (a) is positively correlated with GDP (corr. coeff. = 0.81) (b) lags GDP (c) is less volatile than GDP (std(c)=0.60*std(gdp))

24 Okun s Law Percentage change in real GDP Y Y u Okun s Law: the negative relationship between GDP and unemployment Change in unemployment rate

25 Some facts about business cycle GDP growth averages percent per year over the long run with large fluctuations in the short run. Consumption and investment fluctuate with GDP, but consumption tends to be less volatile and investment more volatile than GDP. Correlation of consumption and the output gap: Correlation of investment and the output gap: Unemployment rises during recessions and falls during expansions. Correlation of unemployment and GDP is about 0.67 Okun s Law: the negative relationship between GDP and unemployment.

26 Business cycle facts (quantities) Output movements persist (but not periodic) Industries move together Consumption, investment and imports are pro-cyclical They move with the business cycle and this is highlighted by positive correlations Unemployment is countercyclical, but productivity is procyclical

27 US Nominal and Real GDP Real GDP (in 2000 dollars) Nominal GDP

28 Percentage change from 12 months earlier Two measures of inflation in the U.S. 15% 12% 9% 6% 3% 0% -3% GDP deflator CPI

29 GDP and price level Note that the correlation may change w.r.t. the period considered. Prices: (a) are negatively correlated with GDP (corr. coeff. = -0.26) (b) coincide with GDP (c) are less volatile than GDP (std(c)=0.58*std(gdp))

30 Inflation gen-58-2 lug-65 gen-73 lug-80 gen-88 lug-95 gen GDP (Deviation) Inflation Rate Notice that inflation tends to decline during recessions and increase during expansions.

31 The return on a 90 day T-Bill Jan-57 Jan-62 Jan-67 Jan-72 Jan-77 Jan-82 Jan-87 Jan-92 Jan-97 Jan-02 Jan-07 Interest rates tend to decline during recessions.

32 Cyclical Variation Output gap and real short interest rate The correlation between short term interest rates and Real GDP is approximately This indicates that real interest rates are (mildly) procyclical Real Fed Funds Rate (Percent) Real Fed Funds Rate (Real) Output Gap

33 GDP and money supply Note it also depends on the money aggregate considered. Money supply: (a) is positively correlated with GDP (corr. coeff. = 0.36) (b) leads GDP (c) is slightly less volatile than GDP (std(c)=0.77*std(gdp))

34 Output gap and velocity The correlation between velocity and GDP is about This indicates that velocity is highly pro-cyclical US Percent Velocity Output Gap

35 Business cycle facts (prices) Inflation is pro-cyclical. Real wages are mildly pro-cyclical or a-cyclical. Short term interest rates are pro-cyclical. Money and velocity are pro-cyclical.

36 Business cycles: Main stylized facts The goal of any business cycle model is to explain as many facts as possible Variable Correlation Leading/Lagging Consumption Pro-cyclical Coincident Unemployment Countercyclical Coincident Real wages Pro-cyclical Coincident Interest Rates Pro-cyclical Coincident Productivity Pro-cyclical Leading Inflation Pro-cyclical Lagging

37 Appendix Descriptive statistics tools Averages: Mean Median Mode Measures of Dispersion: Variance Standard Deviation Covariance and Correlation between different variables (or same variable across different points in time!)

38 Averages from data Suppose you have some data on a particular random variable, x, i.e. x i = {x 1, x 2, x 3,, x n }, i = 1,,n, are the n observations forming your random sample E.g. height of people in this room at the very moment. Different Measures of Averages Mean: 1 n 1 E x x i 1 i ( x1 x2... x n ) n n Weighted Mean: 1 n 1 x w x w x w x... w x i 1 n n i i n n where w i are the weights on each observation, i = 1,,n

39 Other measures of central tendencies Median: is the value m, such that: Prob(X m) ½ And Prob(X m) ½ in other words: the median value in a data sample for a random variable, is the middle ranked observation, i.e. the particular observation where at least 50% of all other observations are smaller than it. Mode: is the value of x i occurring most frequently in the sample.

40 Standard deviation and variance The variance is a measure of dispersion in the sample. 1 Var( x) E x x i 1 i n 1 1 n 2 2 x i 1 i n n 1 n 2 2 Variance of x is represented by the symbol s 2 x for the population, and s 2 x for a sample. Standard Deviation is the (positive) square root of the variance, and is represented by symbol s x for the population, and s x for a sample.

41 Covariance and correlation For a sample of two or more variables, we can compute the covariance and correlation between different variables. We can also compute the covariance or correlation of a single variable over time Correlation for a population, Cov( x, y) E x x y y 1 n x i 1 i x yi y sxy n 1 Cor ( x, y) xy s x xy ss y

42 Covariance and correlation Covariance for a population equals Cov( x, y) E x x y y s xy The sample counterpart equals: 1 n Cov( x, y) 1 x y s i n 1 i x i y xy Similarly, the population (sample) correlation for two variables (x,y) is defined as the covariance divided by the (sample) standard deviation of each variable: s xy sxy Cor ( x, y) xy (sample: rxy ) ss ss x y x y

43 Scatter diagrams A scatter diagram plots the value of one variable on the x-axis and the value of another variable on the y-axis. A scatter diagram can make clear the relationship between two variables. The following three scatter diagrams are examples of variables that move in the same direction, in opposite directions, and in no particular relationship to each other.

44 Some references Mankiw, N. Gregory, 1990, A Quick Refresher Course in Macroeconomics, Journal of Economic Literature Kydland, F. and Edward Prescott, 1990, Business Cycles: Real Facts and a Monetary Myth, Quarterly Review, Federal Reserve Bank of Minneapolis Romer, Christina, 1999, Changes in Business Cycles: Evidence and Explanations, Journal of Economic Perspectives Basu, S. and A. Taylor, 1999, Business Cycles in International Historical Perspective, Journal of Economic Perspectives Zarnowitz, V., 1999, Theory and History Behind Business Cycles: Are the 1990 s the Onset of a Golden Age?, Journal of Economic Perspectives