Monetary Policy and

the Control of Inflation

John W. Crow

As presenter for this session of the symposium I see my task as

more to indicate than to prescribe. There are other reasons as well
for a cautious approach. Even if one could claim to have a good
understanding of the common features of the economies in question,
the differences among them are also very important. It is therefore
unlikely that any single prescription could satisfy all cases. And
while I can claim a background in central banking, I can only claim
a very great interest in the financial and economic transformations
taking place across Central and Eastern Europe and the conditions
under which central banks have to operate and establish a place for
monetary policy.

Some overlap with other parts of the symposium is unavoidable.

For example, we need to assume that in this morning's first session,
on the role of central banks, it was accepted that central banks will
be provided with the tools necessary to conduct monetary policy in
a way that (of course) gives central attention to monetary stability.
Monetary stability can also be described as price stability or, in the
possibly less demanding terms chosen for this session, as inflation
control.

It is also worth highlighting at the outset the particular relevance

of the fact that we are discussing monetary policy against the
backdrop of "emerging market-oriented economiesw-that is,
economies emerging from the system of "material balancing,"
10 John W. Crow

under which quantities of inputs are made available administratively

on the basis of estimated needs for fulfilling the production targets
stemming from a central plan, and moving toward one where the
price mechanism and competition in private markets play a major
role in resource allocation.

In a market economy, monetary policy is transmitted above all

through changing yields in financial markets. Until the transmission
linkages-from the instruments of monetary policy to financial
markets, and from financial markets to the rest of the economy-
have developed, monetary policy as it is generally perceived is
unlikely to be very effective. Conversely, while the general range
of market reforms or creations under way are hardly being under-
taken in order to improve the playing field for monetary policy, the
better financial markets function, and the better they become linked
to the market decisions of consumers and investors, the more
effective monetary policy will be in getting macroeconomic results.

Another reason for emphasizing the shift toward market

economies is that while inflation is bad news under any economic
system, it is particularly bad for a decentralized, that is market,
system. For this system, with its reliance on the price mechanism
and monetary exchange, monetary stability is essential if market
signals are going to be interpreted efficiently. In other words, to
function well, a market economy depends on confidence in the
money that is being used.

Broad approaches to establishing monetary confidence

This section reviews the main ways in which monetary policy

might be anchored so as to secure the credible performance neces-
sary for monetary confidence.

Bearing in mind perhaps the debates in Western Europe over the

exchange rate system, consideration needs to be given in this session
to the merits of a fixed exchange rate as the anchor for monetary
stability. Such an approach would also entail an early decision to
make the currency convertible in the foreign exchange market (at
least for current if not right away for capital account transactions).
Monetary Policy and the Control of Inflation II

Currency convertibility will also get a discussion all to itself later at

this symposium.

For monetary policy, the crucial value of relying on a fixed

exchange rate anchor is that it promises to deliver early credibility.
It does this through explicitly locking onto a currency that already
enjoys a good reputation for preserving its purchasing power.
Credibility would be enhanced if it is likely that strong trading links
will develop with the economy using the currency in question. In this
context, the European Monetary System could be an obvious pole of
attraction-either in the collective currency bundle, the ECU, or,
conceivably, through its strongest constituent currency.

Still, arguments for not betting monetary policy on the exchange

rate have considerable weight.

Reliance on the exchange rate (in effect on someone else's

monetary policy) for monetary discipline, and therefore for con-
fidence, means commitment to a fixed exchange rate that must be
seen as unwavering if it is to be credible enough to shape price and
cost decisions. Simply pegging the exchange rate may not by itself
provide the necessary demonstration of commitment. Indeed, can
anything less than some sort of monetary union be seen as "unwaver-
ing?" And practically speaking, monetary unions or EMS-type
arrangements for monetary policy convergence look a fair way off
at this point.

Fixing the exchange rate means that the discipline on domestic

costs that is imposed by an unyielding exchange rate dominates other
considerations regarding the role of the exchange rate-particularly
the role of shifts in the nominal exchange rate in facilitating shifts in
the real exchange rate (the exchange rate adjusted for relative rates
of inflation) as a way of getting the economy to adjust efficiently to
real economic changes.

As an illustration of the above point, it may be presumed that the

terms of trade will shift as CMEA trade declines and new trading
patterns emerge. More generally, relative prices will move, proba-
bly a great deal, under the impact of market transformation. It
I2 John W. Crow

follows that the real exchange rate will likely have to move as part
of the general process of relative price adjustment. Without some
flexibility in the nominal exchange rate, it may well not be possible
to get anything approximating the necessary exchange rate shifts in
real terms except at great economic cost in terms of foregone output
and increased unemployment.

What about domestic indicators, guides, and anchors?

In this area it does not seem likely that any clear-cut option would
be available. One common approach from the domestic angle is to
use monetary or credit aggregates as a way of plotting a path for
monetary policy. However, in view of the economic and institutional
transformations taking place, it seems unlikely that stable relation-
ships to either spending or interest rates will be forthcoming any time
soon.

Accordingly, the authorities might be well-advised to adopt an

eclectic approach to the use of domestic guides. An initial focus on
an IMF-styled measure of maximum domestic credit expansion,
grounded in the central bank's balance sheet, may well be called for.
Whatever other monetary management decisions are implemented,
in the prevailing economic and institutional flux the central bank will
want to keep a close eye on the pace at which its own assets are
expanding (more on this below).

The authorities will in any event want to monitor movements in

any relatively broad indicators of prices that are available, as a gauge
by which to adjust the thrust of monetary expansion. At the same
time, in this period of structural change relative prices will be
changing a great deal in all markets, and it may be difficult indeed
in the shorter run to separate out the impact on the price level of
changes in relative prices, from the impact originating in a process
of general domestic cost and/or price inflation. How well relative
cost or price changes can be distinguished from general movements
in costs or prices will of course depend in part upon the quality of
the cost and price measures. Developing satisfactory measures will
therefore be a priority.
Monetary Policy and the Control of Inflation 13

Even if monetary policy is not fully geared to the exchange rate,

the balance of payments outturn in general, and exchange rate
pressures in particular, can obviously also play a role in gauging the
degree of demand and inflation pressure, and therefore in indicating
a need to adjust the thrust of monetary policy. However, as in the
case of price movements, in this area also it will be important to
make distinctions-between on the one hand the effects of demand
on the balance of payments and the exchange rate, and on the other
the effects of influences of a structural nature.

The pace of increase of the general level of money incomes

(conventionally called the money wage rate) is clearly of great
relevance in terms of the possibilities for inflation control. However,
it may be relevant not so much because of its sensitivity to demand
pressures, but more especially because of its cost and price push
features in a particular institutional context. These features could call
for other forms of restraint as a complement to monetary restraint-
not as a way of life in a market-based economy but in the initial stages
of economic stabilization. In this way the money wage rate could act
as a temporary nominal anchor alongside any available guides that
are more directly linked to the monetary process. In this context, the
Polish experience to date will be of interest. There, wages (on the
basis of a tax-based incomes policy) apparently serve as the nominal
anchor, reinforced to some degree by a fixed exchange rate.

Building the foundations

Evidently, a great deal of institution building (beyond the questions

relating to central bank structure and powers that were presumably
covered earlier this morning) will have to occur before the
economies in question can use monetary policy in the market-based
way that is available elsewhere.

No doubt a major priority at the start is to establish a market-

oriented system of deposit gathering and commercial credit exten-
sions-the traditional field of commercial banking. But also, more
generally, the authorities will wish to encourage greater competition
in the financial markets, to channel savings to their most productive
uses. Within the OECD this objective has been met both by systems
14 John W.Crow

in which banks play a dominant role (the continental European

model), and by systems in which marketable securities are also very
important (the North American-U.K. model). Markets for short-
term funds are, in either case, of particular interest to central banks,
given the fact that their policy actions are transmitted through such
markets into the rest of the financial sector and into the non-financial
sector.

Sometimes appropriate market structures have developed only

after central banks and governments have taken the lead. For
example, the Bank of Canada realized quite early in its existence that
a well functioning money market-dealing in treasury bills, com-
mercial paper, overnight funds, and the like-would assist the
implementation of monetary policy as well as the overall efficiency
of the economy. But although the banking system as such had been
well developed for many decades, an active money market emerged
only after a series of Bank of Canada initiatives in 1953-54. (These
are documented in the appendix.)

From the viewpoint of monetary control, and therefore inflation

control, the development of the Canadian money market had two
particularly desirable features.

In the first place, the Canadian money market's development

provided an avenue for increased reliance on price-related methods
of monetary management-broadly speaking, open market opera-
tions. And in this process, reliance on jawboning and on bank
liquidity ratios to influence commercial banks' extension of credit
became less and less-to the point that these features now have no
role in Canada.

Secondly, the broadening of outlets for the placement of govern-

ment debt-to include the money market as well as the bond
market-helped to provide a first line of assurance that government
deficit financing would not impinge upon monetary control. In
general, in the absence of broad and resilient financial markets
through which to absorb financing demands, the central bank would
find it very difficult to deflect direct pressure from government
Monetary Policy and the Control of Inflation 15

deficits on its balance sheet and therefore on inflation of the monetary

base.

To deflect the pressure by, for example, imposing higher bank

reserve requirements in cash, or in government securities, is not an
adequate solution. At the very least it causes problems for the
efficiency and competitiveness of the deposit-taking part of the
financial system. A better solution would be for the government to
pay an interest rate sufficiently high that it attracts willing lenders,
and without pumping up the money supply. In general, if credit of
various kinds really has to be subsidized or channelled preferentially,
the subsidy should be out in the open and not financed through what
is in effect a tax (and therefore fiscal, not monetary, policy) on the
intermediation of savings through the banking system.

A related issue with implications for controlling inflation is the

importance of developing at an early stage a workable system of
prudential oversight for financial institutions, including determining
which institutions will have access to the lender-of-last-resortfacility
for liquidity purposes. This, too, is a separate topic of discussion in
a later session. Its importance for inflation control is to remove a
potential constraint on the conduct of monetary policy. The presence
of distressed institutions may inhibit monetary discipline, for fear of
precipitating a crisis in the financial system or of disrupting the flow
of investment finance to the non-financial sector,

The immediate challenges

Building institutions and building confidence takes time, and

meanwhile monetary policy has to do what it can with what is
available, keeping its eye fixed firmly on inflation control.

It cannot be emphasized too strongly that central to any prospect

of overall monetary control is the absence of pressure on the central
bank to inflate its balance sheet (simply put, to print money) because
of budget deficit financing needs. Depending of course on each
country's circumstances, some fiscal reining in, even fiscal reform,
is therefore probably implied. This does not necessarily mean that
deficit financing would not be available at all from the central bank.
16 John W.Crow

But such financing would have to be budgeted for in the framework

of an overall monetary financing plan that reflected the needs of
monetary control and not whatever size the deficit happened to be.

There may be a large monetary "overhangv-a form of forced

savings-as a legacy from the previous system of passive credit
creation and price control. The consequence may be an extreme burst
of spending and inflation when prices are decontrolled. This carries
the threat of derailing the process of building monetary confidence
before it even gets started.

Here, the challenge clearly is to eliminate this burden as quickly

as possible. One way of absorbing the overhang is the privatization
of existing government assets. This is a delicate problem, given the
natural concerns about selling at (possibly) fire sale prices, espe-
cially to foreigners. Nevertheless, asset sales will be a useful safety
valve. The housing stock in particular is a prime candidate for
privatization, and one that in the hands of individuals would greatly
increase their stake in the market economy. At the same time the
disposition of state assets, for which the true market value of many
is far from clear, will raise important prudential questions. If such
sales are made on credit to domestic residents or firms it is critical
that this not lead to an extension of the safety net of lender of last
resort to insolvent financial or non-financial firms. It is also possible
that an upward price level shift (for example through an increase in
the relative price of food to stimulate production) could absorb some
or all of any overhang of real balances. Currency reform, to
eliminate excess balances, clearly has a place on the list of potential
measures, but if it does not give every appearance of being definitive
and final it will surely have serious after-effects on confidence in the
monetary system. Therefore, currency reform looks more like a
constituent measure of a broader stabilization package, most likely
in the context of associated deficit-reducing fiscal measures to
remove a systemic cause of inflation, rather than a self-contained
step.

Another crucial challenge will be to ensure that "real" interest

rates on both loans and deposits are positive. This will help to ensure
a proper incentive for voluntary savings, and that there is room to
Monetary Policy and rhe Conrrol of Inflation 17

differentiate credits on the basis of risk. It will also help to diminish

any monetary overhang to the extent that real yields on these
"overhang" balances were formerly negative.

In a period when rapid structural, and therefore relative price,

changes are occurring, and barring a fixed exchange rate solution
that can be seen as definitive, the monetary authorities will have even
greater difficulties than usual in determining what rate of monetary
expansion is consistent with inflation control. Control over the
central bank's assets (the ultimate lever of monetary policy) would
likely need to be closely managed-so as tb limit the pace of central
bank credit expansion to one consistent with the authorities' best
estimates of the flow of resources accruing to the central bank
through the growth of the liability side of its balance sheet (for
example, from bank notes issued, increases in commercial banks'
reserves) in a noninflationary environment. There will be distinct
merit in conservative estimates because confidence, once lost, is not
easily regained. Such a "domestic credit expansion" approach has
the virtue of safety first, being both a potentially useful brake on
excessive monetary expansion and protective of the overall balance
of payments .l

End Notes
' s e e for example, Manuel Guitian, "Credit Versus Money as an Instrument of Control,"
International Monetary Fund, Staff Papers (November 1973).
 John W.Crow

Appendix
Establishing the Canadian Money Market

Before 1953 there was no organized short-term money market in

Canada. Business had few options other than the chartered banks for
short-term financing or money placement, at sticky interest rates set
by the banks. "Call loans" of the banks to money market dealers
were then quite illiquid and priced on the basis of the prime loan rate.
Treasury bills and commercial paper were typically held to maturity.
The banking system relied heavily on New York for short-term
liquidity adjustments.

The lack of flexibility and competition in the domestic market for

short-term funds was clearly detrimental to the efficient provision of
financial services and the allocation of savings. Moreover, from its
establishment in the mid-1930s, the Bank of Canada had been aware
that its operations were hampered by the lack of a money market. In
the absence of active markets for bills and overnight funds, the banks
would allow their excess cash reserves to fluctuate over a fairly wide
range. This meant that the central bank did not have reliable leverage
over short-term monetary conditions, even though it could set the
supply of cash to the banking system with some precision (through
open-market operations).l The banks tended to react sluggishly to
reserve tightening or easing. The Bank thought that the establishment
of an active money market would induce the banks to manage their
cash more finely and so provide a more effective channel for the
transmission of monetary policy. However, any plans that the Bank
may have wished to pursue in this respect had to be put off because
of World War I1 and its economic aftermath.

By the early 1950s conditions in Canada were in principle

favorable for the formation of a money market: official controls
imposed during the war had been dismantled, and the central bank
no longer attempted to hold down market interest rates; savings and
investment had expanded strongly; the banks, trust companies, and
insurance companies formed a core of large and sophisticated inter-
mediaries; and since bond and equity markets were thriving, there
were plenty of securities firms with the resources to be money market
Monetary Policy and the Control of InfIalion 19

dealers. Yet the overt encouragement of such a development by the

Bank of Canada met with little success until various concrete initia-
tives were taken:

1953 The Bank of Canada offered a line for short-term financing

-purchase and resale agreements (PRA)-to a group of 13 invest-
ment dealers, known as "jobbers." Under this facility, inventories
of treasury bills and short-term Government of Canada bonds could
be temporarily financed at the central bank. Each jobber was given
a specific limit on its access to PRA; the total value of the available
lines was initially about as large as the eligible inventory held by the
jobber group. Auctions of treasury bills were changed from a
fortnightly to a weekly cycle, and amounts auctioned were increased.

Since the charge for PRA was substantially less than the prime
rate, the jobbers quickly switched to central bank financing for most
of their needs, going to the limit of their PRA lines. The Bank urged
the chartered banks to provide day loans to the dealers at rates in line
with money market yields, but the banks initially showed no interest
in such an innovation.

1954 The Bank Act Revision introduced reserve averaging.

Since statutory cash requirements would now apply to average
reserve positions over a calendar month, the banks could make
sizable economies on excess cash. This led to a more stable demand
for bank reserves, and to prompter system responses to changes in
policy settings.

It was decided that credits to chartered bank reserve accounts for

Bank of Canada security purchases would be subject to a settlement
delay of at least two days. In contrast, settlement at the Bank for the
proceeds of day loans that chartered banks had called from invest-
ment dealers would be booked next day. This difference in the speed
of settlement greatly enhanced the relative position of day loans for
liquidity adjustments by the banking system.

The chartered banks now agreed to provide day loans, at rates near
those on treasury bills, up to the limit of unused PRA lines. This
20 John W. Crow

allowed the central bank to confine itself once again to last resort
lending.

The banks also agreed to reduce charges for intra-day (daylight)

overdrafts of the investment dealem2

The Bank provided wire facilities across the country for the
transfer of government securities.

These measures achieved their objectives. By 1958 an active

money market provided the channel for the transmission of monetary
policy that the Bank had sought. In 1962 the central bank allowed
bankers' acceptances to be eligible for PRA to assist their introduc-
tion in Canada, phasing out this support, later when the BA became
well established. By this point the money market was going ahead
under its own steam; and the rapid expansion of recent decades has
been propelled by profit opportunities, competition and new tech-
nology rather than by policy measures. Indeed, the contribution of
the authorities in the latter period has been more to liberate market
forces than to undertake new measures of support.

This experience would suggest that the central bank can make a
useful contribution to structural change in the financial sector. In the
right environment, such as Canada's money market in the 1950s, a
few key policy actions can propel rapid modernization. In turn, a
well-functioning money market has proved of great value in improv-
ing the competitive efficiency of the financial market and in the
implementation and transmission of monetary policy.

End Notes
'"Cash" in this context refers to the deposit claims of banks on the Bank of Canada. These
claims serve as reserves, for which there have been legal minimum requirements in Canada
as elsewhere, and as settlement balances in the payments system.

2 ~ h fee
e of a bank for a daylight overdraft is called "over-certification."Such overdrafts
are necessary to dealers in securities because of technical lags in delivery and settlement.
Monetary Policy and the Control of Inflation

References
Bureau of Economic Analysis 1990. Survey of Current Business. U.S. Department of
Commerce, April.
Connor, John M. Food Processing, An Industrial Powerhouse in Transition, Lexington, Mass.:
Lexington Books, 1988.
Fleming, Ann and Jeannine Kenney. "Will Consumers Benefit from New Dairy Tech-
nologies?" National B o d Review, JanuaryIMarch, 1989, pp. 17-21.
Hopper, Paul F. and Daryl B. Lund. Testimony before the House Subcommitteeon Department
Operations, Research and Foreign Agriculture, Committee on Agriculture, U. S. House of
Representatives, February 6, 1990.
Labuza, Theodore P. "Future Food Products: What Will Technology Be Serving Up to
Consumers in the Years Ahead?" Proceedings of the 1985 Annual Meeting of the
Association of Food and Drug Officials, Madison, Wisconsin,l985.
Putnam, Judith Jones. "Food Consumption, Prices, and Expenditures, 1967-88," U.S.
Department of Agriculture, Economic Research Service, 1990.
U.S. Department of Commerce, International Trade Administration. US.Industrial Outlook,
Washington: Government Printing Office, January, 1990.