View original document

The full text on this page is automatically extracted from the file linked above and may contain errors and inconsistencies.

FRBSF WEEKLY LEtIEA
October 2, 1987

How Reliable is Ml A?
M 1 grew at a very rapid 13 V2 percent average
annual rate in 1985 and 1986. Under normal
circumstances, such rapid growth would have
preceded a surge in ()utputand inflation, but
output growth has remained moderate and inflation relatively subdued. Prior to this episode, the
Federal Reserve had often relied heavily on M1
in the conduct of monetary policy. In 1987, for
the first time, the Fed decided not to establish an
M1 target range. The breakdown in the relationship between M1 and macroeconomic developments has since spurred the search for other
monetary aggregates that might provide useful
information about the future course of the
economy.
The authorization of interest-bearing checkable
deposits (NOW accounts) in the 1980s is a commonly cited reason for M1 's aberrant behavior.
As a result, some observers have argued that
M1A, which includes only non interest-bearing
currency and demand deposits, should be
affected less by deregulation than M1, which
includes NOWs. In fact, they have argued that
M1A behaves like M1 prior to deregulation, and
that M 1A may therefore be a rei iable alternative
to M1 as a monetary policy indicator.
This Letter discusses the conceptual basis for this
view and also looks at the available evidence.
We can find no compelling reason for expecting
M 1A to behave like M1 did prior to deregulation. Moreover, our empirical evidence supports
an even stronger inference: that movements in
M 1A provide little useful information about the
future course of the economy.

Is M1A like Ml used to be?
An important reason for the stable and s.imple
relationship between M1 and GNP that persisted
for a substantial portion of the post-war period
appears to be that interest on checkable deposits
was prohibited by law. This prohibition apparently induced individuals to keep most of their
savings-type balances in interest-bearing small
time and saving deposits, as well as open market
securities. In other words, the regulation of interest on checkable deposits effectively separated

transaction balances from those held for savings
purposes, and thus set M1 apart from the
broader monetary aggregates, M2 and M3.
The introduction of interest-bearing checkable
deposits ~ NOW accounts, which are included
appears to have changed all that.
in M1
Households use NOW accounts to store both
savings and transactions balances. As a consequence, variations in the growth rate of M1 are
likely to reflect not only planned changes in
transactions (spending), but also the diverse set
of factors that determine saving and portfolio
allocation - such as changes in interest rate differentials between alternative liquid assets,
uncertainty about future returns, and changes in
investors' preferences for various maturities and
liquidity characteristics. Such portfolio adjustments can interfere at various times with the
relationships between M1, income, and prices.
This conclusion about the source of problems
with M1 makes it tempting to look at narrower
monetary aggregates, which exclude interestbearing instruments, as monetary policy indicators. M 1A is a natural candidate since neither of
its components - demand deposits and currency in the hands ofthe public - bear interest.
One might conclude, then, that M1A primarily
contains transactions balances, and thus may
have a stable relationshipwith GNP and prices.
Unfortunately, such a conclusion does not
appear to be warranted by conceptual Or empirical considerations. NOW accounts have
attracted a substantial amount of household
transaction balances that are excluded from
M 1A. There is no reason to expect that the
behavior of M 1A, which contains most business
transaction accounts but only a fraction of those
held by households, would bear much of a
resemblance to the behavior of M 1, wh ich contained most business and household transaction
accounts, prior to deregulation.
Furthermore, interest-sensitive households are
more likely to have moved their transactions
accounts into NOWs. Consequently, the nature

FABSf
of household transactions accounts still held in
M 1A is likely to differ from those that were held
in M1 prior to deregulation. Deregulation may
have affected the behavior of demand deposits
held by firms as well, since it has provided firms
with a variety of close substitutes.

Indicator value of M1A
Even though conceptual considerations by themselves do not suggest thatM1A would be a reliable monetary policy indicator, we cannot rule
out the possibility on theory alone; For example,
if the portion of household transaction accounts
that remains inM1A bore some consistent relationship with that in NOW accounts, M1A might
prove useful for the Federal Reserve in conductingmonetary policy. The available empirical
evidence, however, suggests otherwise.
Fi rst, even though M 1A growth has been
noticeably less volatile than M1 growth over the
past few years, it has been fadrom stable. A
4 V2 percent average M 1A growth rate over
1983-84 (fourth quarter to fourth quarter) was
followed by an average rate of 10 percent in
1985~86. Real income grew strongly in 1983
and 1984, but at modest rates over the following
two years, when M1 A growth was rapid; nor did
inflation show any signs of picking up in
response to the acceleration in M 1A.
Thus, as shown in Chart 1, the velocity of M 1A
(measured as the ratio of GNP in current dollars
to M1A) increased in both 1983and 1984 (similar to its behavior over the ·1960s and 1970s) but
declined dramatically in 1985 and 1986. M1A's
velocity grew at an average annual rate of
approximately 3.5 percent over 1983-84, but fell
at a 4.5 percent average annual rate over
1985-86.While this behavior was less volatile
than that of M 1's velocity, the sharp fall in M1 A
velocity over the past two years represents a
major departure from past trends.
Even though M 1A by itself does. not appear to
contain very much information about future
movements in GNP and prices, it still is possible
that this aggregate contains information over and
above that contained in other variables. If this
were the case, a combination of M1A and these
other variables could be used as an indicator of
future developments in the economy.

Chart 1

Velocity

8

,..0'""

.....1

7
6

I
M1A

":
..

t'

't

:

,"

5
4

3
2-j-nC"T""fTTTT'""'""'TTTT1~rrn,,-r""""C"T""fT""""""-T"T"T""J"T""T"T"rrn~

\ ,?>(Q~,?>(Q(,?>(Q~,?>(Q~,?>(Q~,?>1~,?>1 (,?>1 ~,?>1~,?>1 ~,?>'O~,?>'O(,?>'O~,?>'O(Q

To test for this, we used a technique called vector.autoregression to examine the relationship
between real personal income, the price level,
the six-month commercial paper rate, the nonM1A components of M3, and M1A itself. First,
we used all these variables to predict M1 A. We
then looked at how M1A velocity would be
affected by an increase in M1 A that could not be
predicted on the basis of these variables. For
M1A to be a useful indicator for monetary policy, such a positive "surprise" in M1A should be
followed by an increase in income. Since
velocity is defined as the ratio of income to
money, the increase in income implies that
velocity should tend to return to its original
level.
We present the results of this exercise for two
different time periods in Chart 2 - the prederegulation period from the beginning of 1974
to mid-1979 and the post-deregulation period
from mid-1981 to the end of 1986. The chart
shows the average cumulative changes in M1A
velocity that occurred in response to an unpredictedJ -percent increase in M 1A.ltis useful to
keep in mind that over most of the 1970s, M1 A
was virtually identical with M1. Thus, a comparison of the two panels shows how M 1A today
corn pares with M1 prior to deregulation.
In the pre-deregulation period, velocity tended
to return to its original level shortly after a sur-

Chart 2
Response of M1A Velocity to a 1 Percent M1A Surprise

Percent

A.

January 1974 to June 1979

Percent

0.0

B. July 1981 to December 1986

0.0
-0.2

-0.2
-0.4
-0.6

-0.4

-0.8
-0.6
-1.0
-0.8

-+--~r----r-----.--~.,.-"----r~~-,

4

8

12

16

20

24

Months After 'Surprise'

prise increase in M1A. Thus, a permanent
1-percent change in M 1A would have been followed by an equal permanent increase in
income. By contrast, during the 1980s, the surprise in M 1A is followed essentially by no
change in income because a 1-percent increase
in M 1A leads to a 1-percent decrease in
velocity. This evidence suggests that, during the
1980s, changes in M 1A do not provide useful
information about future changes in income.

Conclusion
A breakdown in recent years in the relationship
between M1 and GNP has forced the Fed to rely
more heavi lyon its broader monetary aggre-

-1.2
4

8

12

16

20

24

Months After 'Surprise'

gates, M2 and M3, which include savings-type
balances. Nevertheless, there remain strong theoretical reasons to believe that a properly measured transactions aggregate would be a more
useful monetary policy indicator. Thisconsideration has stimulated analysis of alternative
transactions aggregates, such as M 1A. Unfortunately, our analysis suggests that M 1A does
not seem to bear a very close relationship with
macroeconomic developments. More study is
therefore needed to find a useful alternative to
M1 as the primary guide to conducting monetary policy.

johllP. judd and Bharat Trehan

MONETARY POLICY OBJECTIVES FOR 1987 AND 1988
On July 21, Federal Reserve Board Chairman Paul Volcker presented a mid-year report tothe
Congress on the Federal Reserve's monetary policy objectives for the remainder of 1987 and
1988. The report reviews economic and financial developments in 1987 and presents the
economic outlook heading into 1988. For single or multiple copies of the report, write to the
Public Information Department, Federal Reserve Bank of San Francisco, P.O. Box 7702, San
Francisco, CA 94120, or phone (415) 974-2246.

Opinions expressed in this newsletter do not necessarily reflect the. views of the management of the Federal Reserve Bank of San
Francisco, or of the Board of Governors of the Federal Reserve System.
Editorial comments may be addressed to the editor (Gregory Tong) or to the author .•.• Free copies of Federal Reserve publications
can be obtained from the Public Information Department, Federal Reserve Bank of San Francisco, .P.O. Box 7702, San Francisco
94120. Phone (415) 974-2246.

uo~6U!ljsom

oljoPI

ljo~n

U060JO

HomoH O!UJoJ!l0)

ouozP8

0POl\0U
o~sol8

O)SI)UOJ::J UOS

~O ~U08
aAJaSa~ IOJapa::J

~uew~Jodea
BANKING DATA-TWELfTH FEDERAL RESERVE DISTRICT
(Dollar amounts in millions)

Selected Assets and liabilities
Large Commercial Banks
Loans, Leases and Investments1 2
Loans and Leases 1 6
Commercial and Industrial
Real estate
Loans to Individuals
Leases
U.S. Trea.sury and Agency Securities
Other Securities2
Total Deposits
Demand Deposits
DemandDeposits Adjusted 3
Other Transaction Balances4
Total Non-Transaction Balances 6
Money Market Deposit
Accounts-Total
Time Deposits inAmounts of
$100,000 or more
Other Liabilities for Borrowed MoneyS

Two Week Averages
of Daily Figures

9/9/87
204,811
180,987
51,057
69,738
37,038
5,410
16,899
6,925
207,922
53,147
35,736
20,553
134,221

Change from, 9/1 0/86
Dollar
Percent?

Change
from

Amount
Outstanding

9/2/87
-

1,620
2,555
441
2,537
- 4,175
130
5,458
1,284
475
228
- 13,016
2,871
- 3,575

0.8
3.7
- 10.1
2.3
47.7
- 15.6
0.2
0.4
- 26.6
16.2
2.5

-

44,854

-

51

-

2,318

-

31,087
24,558

-

83
586

-

3,895
1,543

- 11.1
- 5.9

-

-

-

-

Period ended

Period ended

9/7/87

8/24/87

Reserve Position, All Reporting Banks
Excess Reserves (+ )/Deficiency (-)
Borrowings
Net free reserves (+)/Net borrowed (-)
1

0.7

685
735
123
104
91
5
73
23
388
199
638
262
74

-

45
6
39

186
24
162

Includes loss reserves, unearned income, excludes interbank loans

2 Excludes trading account securities
3 Excludes U.S. government and depository institution deposits and cash items
4 ATS, NOW, Super NOW and savings accounts with telephone transfers

S Includes borrowing via FRB, TT&L notes, Fed Funds, RPs and other sources
6 Includes items not shown separately
7 Annualized percent change

-

1.3

4.9

4)Joesel:J