Full text of Economic Commentary (Federal Reserve Bank of Cleveland) : MIA - M.I.A.?, July 1, 1987

View original document

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

which goes far beyond the period usuChart 5 Currency IDeposit Ratios
ally identified with initial flows from
demand deposits to OCDs.
0.7~-------------1
Putting this evidence together, an
interesting possibility emerges. As
households switched transactions
accounts from regular demand deposits
to the new interest-bearing type, the
demand deposits component was transformed. Both the turnover rates and
currency/deposit ratios suggest that
demand deposit behavior has changed
dramatically since deregulation.
Within this scenario, we would
expect to see the velocity of MIA rising
faster during the early 1980s than did
0.2'------,1,.,.965~---~19~7~5- •••••.
-~1~9~85=--1 Ml velocity before deregulation. But, as
noted earlier, the rate of MIA velocity
growth from 1982 through 1984 was
SOURCE: Board of Governors of the Federal
Reserve System.
roughly comparable to that of prederegulation Ml. This corresponds
with the period in which the velocity of
Ml was experiencing sharp declines.
shift as the deposit measure increases
(and vice versa).
It seems possible that the relative staChart 5 shows currency/deposit ratios
bility of MIA velocity in the early 1980s
for demand deposits and total transacmerely reflected a coincidence of offsettions deposits. Like the turnover rates
ting forces on the rate of velocity
in chart 4, a striking feature of chart 5
growth. While the changing composiis that total transactions deposits
tion of demand deposits would have
appear to be behaving more like pretended to raise the average growth rate
deregulation demand deposits than-do
of MIA velocity through higher turndemand deposits themselves.
over rates, this tendency was offset by
The ratio of currency to demand dethe velocity-depressing effects of disinposits shows a protracted rise, which
flation. Thus, it is not the composition
begins at about the same time as the
of transactions deposits that matters
introduction of NOW accounts, but
most, but the relationship of those deposits to nominal GNP.

Federal Reserve Bank of Cleveland
Research Department
P.O. Box 6387
Cleveland, OH 44101

Material may be reprinted provided that the
source is credited. Please send copies of reprinted
materials to the editor.

July 1,1987

Federal Reserve Bank of Cleveland
Conclusion
In 1981, the nationwide introduction of
NOW accounts caused a large shift of
funds from demand deposits to OCDs,
distorting the measured growth rates
of both Ml and MIA. Since then, no
other regulatory changes have had
such distortive effects. Nevertheless,
the velocity of Ml has departed from its
previous growth trend to such an
extent that the FOMC chose not to set
an explicit target for Ml in 1987.
The apparent stability of MIA's velocity from 1982 through 1984 led some
observers to suggest that this narrower
measure of transactions money could
be substituted for Ml as a target. However, data on demand deposit ownership
shares, turnover rates, and currency/
deposit ratios suggest that the observed
stability of MIA's velocity in the early
1980s may represent a coincidence of
offsetting forces.
In the long run, the behavior of MIA
velocity could be expected to be quite different from that of MI prior to the 1980s.
The same factors that have affected
Mls behavior have also affected MIA,
diminishing its usefulness as a potential
policy target. The recent decline in
MIA velocity provides preliminary evidence that its velocity may not, in fact,
follow a growth pattern as predictable
as Ml 's previous velocity trend.

BULK RATE
U.S. Postage Paid
Cleveland,OH
Permit No. 385

Address Correction Requested: Please send
corrected mailing label to-the Federal Reserve
Bank of Cleveland, Research Department,
P.O. Box 6387,Cleveland, OH 44101.

ISSN 0428·1276

ECONOMIC
COMMENTARY
Chart I Velocity of MI and MIA
Log scale

1970

1975

1980

1985

SOURCES: Board of Governors of the Federal
Reserve System; and U.S. Department of
Commerce.

For many years, monetary policy has
been implemented largely through the
pursuit of monetary aggregate targets.
The Federal Open Market Committee
(FOMC), the policy making arm of the
Federal Reserve System, sets target
ranges for the growth of various monetary aggregates, which are intended to
be consistent with the broader objectives of policy.
While the Federal Reserve has maintained the need for multiple monetary
targets, business and research economists have considered the Ml aggregate
to be the most important of these various monetary targets. The Federal
Reserve did not set a target range for
Ml in 1987, however, citing "uncertainties about its underlying relationship to
the behavior of the economy and its
William T. Gavin is an economic advisor at the
Federal Reserve Bank of Cleveland, currently on
leave of absence at the U.S. Department of State.
Michael R. Pakko is an economic analyst at the
Federal Reserve Bank of Cleveland.
The views stated herein are those of the authors
and not necessarily those of the Federal Reserve
Bank of Cleveland or of the Board of Governors of
the Federal Reserve System.

sensitivity to a variety of economic and
financial circumstances .... "1
The uncertainty about Mls behavior
is often described in terms of a breakdown in the growth trend of its
velocity-the ratio of nominal GNP to
Ml (see chart 1). Ml velocity rose at
roughly a 3 percent annual rate for
most of the post-World War II era, fluctuating slightly in response to changes
in nominal interest rates. Since 1982,
however, the velocity of Ml has shown
much greater volatility and has, on
average, declined at a 3.2 percent
annual rate.
In light of Mls weakened status,
economists inside and outside the Federal Reserve System have searched for
an alternative policy target. One proposed solution is for the Federal
Reserve to target an aggregate that
would exclude interest-bearing checking accounts from the present definition of Ml (see table 1). The Federal
Reserve reported statistics for this
monetary measure from 1980 until
1983, referring to it as MIA.2
From 1982 through 1984, the velocity
of MIA seemed to follow a growth
trend similar to that which had previously characterized Ml velocity, providing support for the idea of an MIA
target. During 1985 and 1986, however,
Ml and MIA each grew much faster than
expected given the rates of inflation
and economic growth, resulting in
unanticipated velocity declines for both
measures. Despite this departure, support for greater reliance on MIA in the
conduct of monetary policy has persisted.' In this Economic Commentary,
we examine the behavior of Ml and
MIA in the 1980s and discuss some
1. See "Monetary Policy Report to Congress,"
Federal Reserve Bulletin, vol. 73, no. 4 (April
1987),pp. 239·254.
2. See Thomas D. Simpson, "The Redefined
Monetary Aggregates," Federal Reserve Bulletin,
vol. 66, no. 2 (February 1980),pp. 97-114.

MIA - M.I.A.?
by William T. Gavin
and Michael R. Pakko

Table I The Composition of MI
and MIA
Levels in
Dec. 1986*

Currency and
Traveler's Checks

$189.9

+ Demand Deposits

308.3

MIA

498.2

+ Other Checkable

Deposits

232.3
730.5

"Billions of dollars, seasonally adjusted.
SOURCE: Board of Governors of the Federal
Reserve System.

issues relevant to the possibility of
replacing Ml with MIA.
Most analysts who question the use
of Ml as a policy target have focused
on the contamination of Ml by savingsrelated balances in interest-bearing
checking accounts. However, we suggest that the characteristics of demand
deposits have also been altered by
deregulation; specifically, that demand
deposits are now dominated by commercial accounts. All else being equal,
this change would tend to raise the
growth rate of MIA velocity above that
of pre-deregulation Ml.
The similarity of MIA velocity growth
in the 1980s to pre-1980 Ml velocity
growth may, therefore, reflect a coincidence of offsetting influences. Thus,

3. See John Paulus, "Monetarism: If It Ain't
Broke, Don't Fix It," Economic Perspectives, Morgan Stanley, May 7,1986, pp. 1·9;and more
recently, Michael R. Darby, Angelo R. Mascaro,
and Michael L. Marlow, "The Empirical Reliability of Monetary Aggregates as Indicators: 19831986,"Research Paper No. 8701,U.S. Department of the Treasury, 1987.

Chart 2 Growth of Demand
Deposits and Other Checkable
Deposits
Billions of dollars, quarterly

25
20

Chart 3 Demand Deposit
Ownership
Ratio of nonfinancial business to
household deposits

changes

2.1

Other checkable
deposits

Demand deposits

-25L_~-:1~97~9:--..L.~;;Io.;-.L...1o~1
SOURCE: Board of Governors of the Federal
Reserve System.

the source of the declines in MIA velocity in 1985 and 1986 is likely to be ~he
same as for the earlier breakdown in
Ml's velocity: a fundamental realignment of the relationship between transactions deposits and nominal GNP in
the new disinflationary environment.
MI vs.MIA
Ml was previously considered the most
important of the monetary targets for a
variety of reasons. For many analysts,
Ml was preferable on theoretical
grounds, as the Federal Reserve's .
attempt to construct a comprehenslye
measure of assets that were held pnmarily for transactions. To others, Ml
seemed to be the most controllable of
the targeted aggregates. Finally, many
economists preferred Ml because It
seemed to be most predictably related
to economic activity.
Since the deregulation of deposit
rates in the early 1980s, it is increasingly difficult to argue that Ml represents a theoretically pure measure of
transactions balances. It appears likely
that at least a portion of the funds in
new interest-bearing transactions
accounts represent savings. Furthermore, many money market funds ~nd
accounts allow limited check-drafting
privileges, making it probable that .
some transactions-related funds are in
these non-Ml instruments.
Proponents of MIA as a policy target

4. See "Remarks on Monetary Policy" by Paul A.
Volcker in the Federal Reserve Bulletin, vol. 68,
no. 11 (November 1982), pp. 691-692.
5. For a detailed analysis of the flows among
alternative accounts, see Thomas D. Simpson
and John R. Williams, "Recent Revisions in the
Money Stock," Federal Reserve Bulletin, vol. 67,

have not generally claimed that MIA
provides a comprehensive me~s':lre of
transactions money, but that It IS preferable to Ml because it excludes
accounts contaminated by savings balances. Furthermore, because MIA is a
subset of the relatively controllable Ml,
it might also be more controllable than
either the broader aggregates or proposed weighted-average aggregates,
The most important rationale for an
MIA target, though, is that its relationship to economic activity seems to
have changed less than that of M1.
However, recent declines in MIA velocity indicate that MIA is not as immune
to velocity instability as the 1982 to
1984 experience suggested. I.fwe are.to
consider MIA as a policy guidepost, It
is important that we understand what
has happened to the growth patterns of
transactions deposits-interest-beanng
and non-interest-bearing-in
this era of
deregulation and disinflation.
Deposit Rate Deregulation
The proposal to replace Ml with M~A
may be appropriate if the der~gulatIon
of deposit-rate ceilings underlies the
breakdown in Ml velocity. One of the
important, and easily distinguishable,
effects of deregulation has involved the
flow of funds into newly authorized
types of accounts. In fact, this type of
distortion was behind the FOMC's 1982
decision to de-emphasize the Ml target
temporarily.'
no. 7 (Iuly 1981), pp. 539-544. An alternative view
can be found in John A. Tatom, "Recent Financial Innovations: Have They Distorted the Meaning of Ml?" Review, Federal Reserve Bank of St.
Louis, vol. 64, no. 4 (April 1982), pp. 23·32.

The element of deposit deregulation
most relevant to Ml was the introduction of negotiable orders of wi~hdrawal
(NOW) accounts and automatic transfer services (ATS) accounts. These
interest-bearing checking accounts-:referred to as other checkable deposits
or OCDs-were introduced on an
experimental basis in Massachusetts
and New Hampshire in 1974. OCDs
spread to the rest of New England in
1976 to New York in 1978, and to New
Jerse~ in 1979. They became availab~e
nationwide in 1981. While the behavior
of Ml was measurably affected by early,
limited introduction of NOW accounts,
the effect on Ml velocity was not outside the range of uncertainty normally
associated with velocity forecasts.
The nationwide authorization of
NOW accounts at the end of 1980, however, triggered large transfers of fu,:ds
into the new accounts. Although eVIdence suggests that a complex pattern
of flows among various account types
took place, chart 2 illustrates that the
net effect was a large transfer of funds
from demand deposits to OCDS.5 This
phenomenon is reflected in th~ vel<?City
measures shown in chart 1 pnmanly
as a sharp upward shift in the level of
MIA velocity.
The introduction of Super-NOW
accounts in 1983 and the elimination of
rate ceilings and minimum balance
requirements in 1986 did not seem to .
cause the same type of initial net deposit
flows observed for the nationwide introduction of NOW accounts. One important reason may be simply that the ceilings had become nonbindin~ before
they were eliminated. That ~s, rates
were already below the maximum, so
the elimination of that constraint did
not result in deposit rate increases that
would have attracted new funds.
Regardless of their initial effects, the
new regulations have affected how
people manage their savings and transactions balances. OCD growth has
proved to be higher and more variable
than demand deposit growth, past or
present, given rates of economic growth
and inflation.
Because the major difference in the
two types of transactions accounts is
the explicit interest rates paid on
OCDs, it is often concluded that OCDs
are unlike demand deposits because
they have characteristics of savings
accounts. To the extent this is true, an
MIA aggregate might, in fact, represent a truer measure of the transactions role of money than M1.

Inflation
.
MIA may not resolve the problem WIth
the monetary targeting process, however if the drop in Ml velocity can be
traced to recent disinflation, which has
led to a prolonged and substantial drop
in interest rates. The new interestbearing transactions accounts would be
expected to show a more pronounced
response to the large changes m nO~l11nal interest rates, but the opportunity
costs of all financial assets-including
demand deposits-should
be affected.
Between 1947 and 1979, the average
level of inflation and interest rates
doubled about every decade. As interest
rates rose, both households and businesses looked for ways to reduce the
relative amount of funds held in noninterest-bearing accounts. This behavior is reflected in the steadily rising
velocity of M1.
As inflation and interest rates fell
beginning in 1982, Ml generally grew
more rapidly than expected. The .
incremental deregulation of deposit
ceilings and somewhat sluggish
response of the new floatmg-rat.e
.
accounts resulted in gradual shifts into
some of the newer accounts; but as
interest rates continued to decline, the
spread between rates paid ?n tra~sactions deposits and alternative savings
instruments narrowed.
This situation has reduced the incentive for careful economizing on transactions balances. As interest rates-and
thus the opportunity cost of holding
OCDs-have continued to fall, more
and more people have changed banking
habits; passbook accounts have been
closed and ever-larger balances have
been a'llowed to accumulate in OCDs. If
this accumulation in interest-bearing
OCDs was the only source of Ml velocity declines, however, w~ would not
expect non-interest-beanng demand
deposits to follow the same pattern:
Hence, the declines in MIA velocity
during 1985 and 1986 suggest that at
least part of the upward trend in Ml
velocity from 1947 to 1979 was related
to the upward trend in inflation and
interest rates. In the new period of declining interest rates, the impact of.
interest rate trends on demand deposltsand money demand in general-has
been more clearly revealed.
The Changing Composition of MIA
The MIA velocity declines of 1985 and
1986 would seem to negate the assumption that the "purer" MIA aggregate
can adequately fill the role that Ml

once had as a policy target. Rather, an
Chart 4 Deposit Turnover Rates
explanation that includes th~ effects of
disinflation on the opporturuty costs of
financial assets seems necessary to
Gross Turnover Rates
explain the velocity behavior of both
aggregates. This point becomes even
more apparent when one considers ~ow
deregulation has altered the composiDemand deposits
tion and characteristics of MIA.
Because OCDs can be owned by households but not by businesses, demand
deposits have become increasingly dominated by business accounts. Chart 3
illustrates the stark change in the
composition of demand deposits. After
declining gradually through the 1970s,
the ratio of business to household
demand deposits has risen sharply
since 1980.
Businesses tend to manage their
transactions accounts much more
intensively than do most households, so
the increase in the share of demand
Final Sales Turnover Rates
deposits held by businesses has been
reflected in a rise in the average turnover rate of demand deposits. This, in
turn, affects the nature of MIA's
behavior and will probably affect the
velocity trend of MIA.
The turnover rate of an account is
conceptually similar to velocity because
it defines the relative intensity with
which a particular type of account is
used. As would be expected from the
above discussion, the turnover rates
compared in the upper panel of chart 4
show that an increase in the growth
rate of regular demand-deposit turnover (as in MIA) has accompanied the
SOURCE: Board of Governors of the Federal
Reserve System.
change in ownership composition.
Interestingly, though, the average turnover rate of total transactions deposits
Similar evidence on the effects of
(as in Ml) appears to have increased at
MIA's compositional change can be seen
roughly the same trend rate of growth
in ratios of currency to deposits. The
as during the 1970s.
currency/deposit ratio is important
However, the turnover rates in the
because it reflects the relative usefulupper panel of chart 4 represent the use
ness of financial assets for financing
of deposits in ways unrelate~ to G~P,
transactions. Currency is primarily-if
including intermediate and financial
not exclusively-a transactions asset, .
transactions. The lower panel of chart
so a stable trend in the currency/deposit
4 shows turnover measures adjusted to
ratio might indicate that the relative
reflect only transactions associated
usefulness of the deposit measure was
with final sales." With this adjustunchanged. On the other hand.' if an
ment, sharp declines are evident for
increasing proportion of depoSIts. are
both demand deposits and total transheld for reasons unrelated to their useactions deposits, although demandfulness as transactions, then the curdeposit turnover remains higher than
rency ratio should show a downward
OCD turnover. After the adjustment
has been made to turnover rates,
neither the demand deposits measure
nor the total transactions deposits
measure appears very similar to the
6. The adjusted turnover rates shown in the
demand deposit component of Ml
lower panel of chart 4 are derived in Appendix C of
before 1980.
David E. Lindsey and Paul Spindt, "An Evalua-

~o------------~

tion of Monetary Indexes," Special Studies Paper
195, Board of Governors of the Federal Reserve
System.