Monetary Economics: Short Run Correlations

Keke Sun
Jordi Sebastià
Pedro García Ares (2
)

(2
)

(2
)

1
Data Presentation
Section 1

2
 Data Presentation

Database
10.00

9.50
US economy data from Q1 1959
to Q4 2009 (billions of dollars) 9.00

204 quarters 8.50

Seasonally adjusted
8.00
GDP in real terms
7.50
St Louis Fed database
7.00
(1
)
6.50

6.00

5.50 (3
(2 )
)
5.00

4.50 (3
)
4.00

3.50

3.00
Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3
1959 1961 1963 1965 1968 1970 1972 1974 1977 1979 1981 1983 1986 1988 1990 1992 1995 1997 1999 2001 2004 2006 2008

Log M0 Log M2 Log GDP Log M1

3
Empirical Study
Section 2

4
 Methodology used

Study Simple Correlations

Study the moments of detrended data

1. Calculate the log of the variables

2. Calculate the trends thanks to the
HP filter
3. Calculate deviations from the trend
4. Show the correlations between the
variables

(1
)

(3
(2 )
)

(3
)

5
 Dynamic Correlations

Cyclical Behaviour of U.S. Monetary Aggregates

Cross Correlation of Real GDP with

x(t-8) x(t-7) x(t-6) x(t-5) x(t-4) x(t-3) x(t-2) x(t-1) x(t) x(t+1) x(t+2) x(t-3) x(t+4) x(t+5) x(t+6) x(t+7) x(t+8)

Nominal Money Stock

M0 0.136 0.190 0.240 0.253 0.210 0.072 -0.106 -0.306 -0.421 -0.445 -0.378 -0.269 -0.159 -0.086 -0.059 -0.043 -0.027
M1 0.212 0.212 0.212 0.183 0.120 0.018 -0.105 -0.239 -0.330 -0.368 -0.350 -0.281 -0.194 -0.139 -0.134 -0.125 -0.115
(1
M2 ) 0.055 0.020 -0.015 -0.052 -0.107 -0.200 -0.317 -0.405 -0.419 -0.335 -0.197 -0.072 0.007 0.081 0.110 0.135 0.175

Cross Correlation of the growth rate of Real GDP with

x(t-8) x(t-7) x(t-6) x(t-5) x(t-4) x(t-3) x(t-2) x(t-1) x(t) x(t+1) x(t+2) x(t-3) x(t+4) x(t+5) x(t+6) x(t+7)
(3 x(t+8)
(2 )
)
Nominal Money Stock
(3
)
M0 -0.005 0.018 0.060 0.096 0.120 0.099 0.065 0.023 -0.006 -0.037 -0.050 -0.054 -0.047 -0.044 -0.040 -0.029 -0.021
M1 0.054 0.074 0.103 0.124 0.14 0.13 0.107 0.047 -0.013 -0.076 -0.11 -0.118 -0.103 -0.096 -0.095 -0.081 -0.078
M2 0.056 0.136 0.229 0.314 0.387 0.418 0.394 0.312 0.195 0.065 -0.042 -0.134 -0.22 -0.269 -0.281 -0.273 -0.232

Source of Data: St Louis Fed

Taking absolute values and growth rates is not the same thing
We are interested in changes not in absolute values 6
 Dynamic Correlations (II)

Cross Correlations of Real GDP Cross Correlations of the Growth Rate of GDP

0.50 0.50

0.40 0.40

0.30 0.30

0.20 0.20

0.10 0.10

0.00 (1 0.00
)

x(t)

x(t+1)

x(t+2)

x(t+4)

x(t+5)

x(t+6)

x(t+7)

x(t+8)
x(t-8)

x(t-7)

x(t-6)

x(t-5)

x(t-4)

x(t-3)

x(t-2)

x(t-1)

x(t-3)
x(t+1)

x(t+2)

x(t+4)

x(t+5)

x(t+6)

x(t+7)

x(t+8)
x(t-8)

x(t-7)

x(t-6)

x(t-5)

x(t-4)

x(t-3)

x(t-2)

x(t-1)

x(t-3)
x(t)

-0.10 -0.10

-0.20 -0.20

-0.30 -0.30 (3
(2 )
-0.40 ) -0.40

-0.50 -0.50
(3
M0 M1 M2 )
M0 M1 M2

Cross Correlations of the Growth Rate of GDP

M0 and M1 are generally acyclical (according to Fiorito and Kollintzas between 0 and 0.20)
M2 is procyclical in the leads and countercyclical in the lags (more than 0.20 and less than -0.20)
7
 U.S. Recessions from Q1 1959 to Q4 2009

US Recessions according to the NBER

Federal Reserve began raising interest rates in 1959.

1960-61 The government switched from deficit to surplus

The government attempted to decrease the budget deficits of the Vietnam War
1969-70 The Federal Reserve raised interest rates with an expansion of inflation.

1973 oil crisis and 1973-1974 stock market crash

1973-75 High rising unemployment and inflation led by Vietnam War.

Federal Reserve raised interest rates dramatically to fight the inflation of the 1970s.
1980
1979 energy crisis

1981-82 Tight monetary policy in the United States to control inflation started the recession.

Raising inflation,
1990 oil price shock and the debt accumulation of the 1980s
1990-91
The collapse of the speculative dot-com bubble
The 9-11
2001
Subprime mortgage crisis led to the collapse of the United States housing bubble later contributing to a
global financial crisis
2007- ?

8
Idealized Business Cycles

Real GDP Business Cycles from 1959 to 2010

Speculations about the death of the Business Cycle

4.00%

3.00%

2.00%

1.00%

0.00%
Q4 1961

Q2 1964

Q4 1966

Q2 1969

Q4 1971

Q2 1974

Q4 1976

Q2 1979

Q2 2009
Q2 1959

Q4 1981

Q2 1984

Q4 1986

Q2 1989

Q4 1991

Q2 1994

Q4 1996

Q2 1999

Q4 2001

Q2 2004

Q4 2006
-1.00%

-2.00%

Trend of GDP Actual GDP

As we can notice, the US trend is between 0% and 1%

The amplitude of the cycles is not very big
Recession will mean to decline substantially behind the trend
9
 Cyclical Components GDP, M0, M1 and M2

7.00%

6.00%

5.00%

4.00%

3.00%

2.00%

1.00%

0.00%
Q1 1959

Q3 1966

Q1 1974

Q3 1981

Q1 1989

Q3 1996

Q1 2004

Q3 2006
Q3 1961

Q1 1964

Q1 1969

Q3 1971

Q3 1976

Q1 1979

Q1 1984

Q3 1986

Q3 1991

Q1 1994

Q1 1999

Q3 2001

Q1 2009
-1.00%

-2.00%

-3.00%

-4.00%

-5.00%

-6.00%

M1 GDP M0 M2

M1 is the more unstable and volatile measure

It seems that monetary policy became more active since the 80´s 10

Business cycles also are less big since them

 Cyclical Components GDP and M0

4.00%

3.00%

2.00%

1.00%

0.00%
Q1 1964

Q1 1969

Q3 1976

Q3 1981

Q1 1989

Q1 1994

Q1 1999

Q3 2006
Q1 1959

Q3 1961

Q3 1966

Q3 1971

Q1 1974

Q1 1979

Q1 1984

Q3 1986

Q3 1991

Q3 1996

Q3 2001

Q1 2004

Q1 2009
-1.00%

-2.00%

-3.00%

-4.00%

-5.00%

M0 GDP

There is no clear evidence that M0 leads the cycle

11
M0 behavior is generally stable
 Cyclical Components GDP and M1

7.00%

6.00%

5.00%

4.00%

3.00%

2.00%

1.00%

0.00%
Q1 1959

Q1 1964

Q1 1969

Q3 2006
Q3 1961

Q3 1966

Q3 1971

Q1 1974

Q3 1976

Q1 1979

Q3 1981

Q1 1984

Q3 1986

Q1 1989

Q3 1991

Q1 1994

Q3 1996

Q1 1999

Q3 2001

Q1 2004

Q1 2009
-1.00%

-2.00%

-3.00%

-4.00%

-5.00%

-6.00%

-7.00%

M1 GDP
M1 is very volatile
There is no clear evidence that M1 leads the cycle
12
M1 moves contemporaneously with the cycle (specially before 1980)
 Cyclical Components GDP and M2

4.00%

3.00%

2.00%

1.00%

0.00%
Q1 1959

Q1 1964

Q1 1969

Q3 2006
Q3 1961

Q3 1966

Q3 1971

Q1 1974

Q3 1976

Q1 1979

Q3 1981

Q1 1984

Q3 1986

Q1 1989

Q3 1991

Q1 1994

Q3 1996

Q1 1999

Q3 2001

Q1 2004

Q1 2009
-1.00%

-2.00%

-3.00%

-4.00%

-5.00%

M2 GDP
As we noticed before, M2 is the more comprehensive measure of the money stock
It seems to lead the cycle
13
Cyclical components of M2 fall before GDP falls; this pattern seems to disappear after 1982
Milton Friedman and Anna J. Schwartz Work
Section 3

14
 Milton Friedman and Anna J. Schwartz findings

Some facts about the study of the Monetary History of the United States (1867-1960)

Changes in the behavior of the money stock have been closely associated with changes in economic
activity, money income, and prices
The interrelation between monetary and economic change has been highly stable
Monetary changes have often had an independent origin; they have not been simply a reflection of
changes in economic activity
Bank failures and the massive withdrawals of currency from the financial system that followed,
significantly shrank the money supply (the total amount of currency and outstanding bank deposits),
which greatly exacerbated the economic contraction

15
 Milton Friedman and Anna J. Schwartz findings (II)

There is a close relationship Chart from the Money and Business Cycles paper by Milton and Anna J. Schwartz (1963)
between the variability of money
and of net national product: the
two curves parallel one another
with a high degree of fidelity

16
 Milton Friedman and Anna J. Schwartz findings (III)

The correlation is generally highest Table from the Money and Business Cycles paper by Milton and Anna J. Schwartz (1963)
when the standard deviations are
compared synchronously; it is
generally lowered if standard
deviations for money are compared
with either later or earlier standard
deviations for NNP though, for the
earlier period

The correlation is highest when

money leads one year for three of
the four lengths of moving standard
deviations.

The correlations rise steadily as the

number of terms in the moving
standard deviations is increased.

17
 Conclusions

Comparisons
Monetary disturbances are the main source of business cycle fluctuations (“Money matters”)
The 1929 crash is a clear example
The need of a stable monetary policy that must be maintained
A monetary
Milton History of the High degree of economic stability has also displayed a high degree of stability of the year-to-year stock of
Friedman U.S. money
and Anna The crisis periods have showed appreciably greater instability of the year-to-year change in both money and
Schwartz income
findings The price level and the monetary aggregates are procyclical
Money and Close relationship between the variability of money and of net national product
Business Cycles Increasing and high long term correlation between GDP and the monetary aggregates
paper
The monetary base or M1 leads the cycle

Cross Importance of using growth rates instead of absolute values

Correlations Decreasing and slow correlation (mainly in the case of M2)
table
Monetary Aggregates are not always procyclical

Our results M2 shows some evidence that leads the cycle by a couple of quarters
Cyclical M2 has a positive correlation with future GDP and negative with past GDP
Component Cyclical components of M2 fall before GDP falls but this pattern seems to disappear after 1982
Graphs
No evidence that either the M0 or M1 leads the cycle
Both are generally procyclical and the monetary base seems to lag the cycle slightly

Problem
Reverse Causality : Monetary Aggregates – GDP?
Apart from M0, other monetary aggregates are endogenous
Endogeneity 18
Credit Endogeneity, banks liquidity needs, unstable money demand, stability of the money multiplier
This positive correlation might simply be due to endogenous monetary policy