NBER WORKING PAPER SERIES

THE THEORY AND MEASUREMENT OFMACROECONOMIC DISEQUILIBRIUM IN

CENTRALLY PLANNED ECONOMIES

Richard Portes

Working Paper No. 1875

NATIONAL BUREAU OF ECONOMIC RESEARCH1050 Massachusetts Avenue

Cambridge, MA 02138March 1986

The author is Director of the Centre for Economic Policy Research,

London; Professor of Economics, University of London; and DirecteurdEtudes Associe. Ecole des Hautes Etudes en Sciences Sociales,Paris. My research on this topic has been supported by theEconomic and Social Research Council, most recently under grantB00230048. I have benefited greatly from collaborative work on theproblems discussed here with Richard Quandt, David Winter,and Stephen Yeo, none of whom bears any responsibility for thispaper. Nor do those who have commented on the paper, including JoeBrada, Wojtek Charemza, Paul Hare, Ed Hewett, Alec Nove, and TomWolf. The paper was written for a Festschrift in honor of FranklynHolzman. The present version is a revision of Birkbeck DiscussionPaper in Economics, No. 161, October 1984. The research reportedhere is part of the NBER?s research program in InternationalStudies. Any opinions expressed are those of the author and notthose of the National Bureau of Economic Research.

Working Paper #1875March 1986

The Theory and Measurement of MacroeconomicDisequilibrium in Centrally Planned Economies

ABSTRACT

The paper considersequilibrium j centrallyexplicit aggregatjve, macinstitutional relatjonshifresh, coherent frameworkopened up a range ofand generated severabehaviour, insightsearly 1980s, whichthreats to it, and sometheory and econometrics.disequilibriumnuanced view ofconventional wIeconomies'•.

on macroeconomicI defend thein theory,It has offered a

1' many CPE phenomena,ical investigation,rk on planners'of the 1970s andequilibrium and

conorny macromacro model and

Richard Portestpartnent of EconomicsBirkbeck College7—15 Gresse Streetlondon W1P 1PAENGtD

issues in recent researchplanned economies.roeconomic approachPS and measurementfor the analysis

possibilit1 importantinto CPE P0

0ies for empirspinoffs: wolicy problems

on macroeconomicopments

centreddevel in market eThe quantity—rationing

econometrics give a more useful as well as a moremacroeconomic reality in CPEs than thesdom characterizing them as perpetual "shortage

The Theory and Measurement of Macroeconomic Disequilibrium in

Centrally Planned economies

Richard Portes

Birkbeck College, University of London

and

Centre for Economic Policy Research

January 1986

1. Introduction

The line of research I began several years ago into the

macroeconomics of the centrally planned economies (CPEs) relies

heavily on several important predecessors. Among these the work

of Franklyn Holzman (especially 1956, 1960, 1968) played a major

role. This derived not only from the conceptual framework, but

also from the open and undogmatic spirit of inquiry underlying

it. Though not an econometrician, Holzman knew that theory had

to admit confrontation with the data; though not a mathematician,

he knew that his understanding of Soviet economic institutions

hadtobe representable in theoretical terms. His own work

This paper was written for the conference on the Soviet Union and

Eastern Europe in the World Economy held at the Kennan Institute

in Washington, D.C. on 18—19 October 1984 and will appear in a

Festschrift for Franklyn Holzman. The present version has been

significantly revised.

1

quickly became part of the conventional wisdom, but more wise

than conventional, and certainly not inhibiting to younger

scholars seeking challenging problems. If he thought he had been

there before you, he would not hesitate to refer you to the

appropriate passage; but this was always good sense rather than

exaggerated sensitivity, professional rather than personal. His

arguments were vigorous and provocative but not intolerant.

This essay is offered in the same spirit.

The motiv

elucidate

atio

the

n fo

gene

on CPE macro

macroeconomic

economics wa

equilibrium

the evidence

of Eastern5

in a specific app

inflation in the

te

omi

i te

u

and

ity

ono

of

empirical

macroe con

prerequi 5

"equilibrieconomies,

with quant

but the ec

magnitude

uat e

omie

mor

a va

the

nd

e

1

r this research

ral problem of

lication, to eval

Soviet—type econ

sting required a

cs than was then

was to clarify

(see Muellbauer a

Dlouhy, 1984, in th

rationing is a state

my is at rest, and we

"di s equilibrium" in

allocation or price vec

a self—reproducing eq

s to

and,

sed

his

ra 1

A

of

ke t

ium

ar,

the

ma

for repres

Europe. T

and gene

he CPEs.

concept

for mar

equilibr

o not cle

measure

ance fro

orate

for t

pr i ate

1978,

ext). An

markets d

rinciple

the dist

e elab

ilable

app ro

Portes

CPE cont

n which

can in p

terms of

tor.

uilimarket—c

(1980) d

economy

subsequent

lear ing

iscusses

char act e

ly seeks

rized by

to measure

Similarly,

brium state

Kornai

of an

ye shortages which heper va si

(1982).

2

that of recent non—

which has stressed

t ons. To call this

1974) school" is

ts which appear in

Grandmont—Laroque

(1978) — Neary-

and expectations

Any application to socialist CIPEs of contemporary neoclassical

A corresponding theoretical framework is

Walrasian general equilibrium macroeconomics

the need for consistent microeconomic founda i

the "Clower (1965) — Barro—Grossman (1971,

perhaps too selective and omits some key elemen

the sequence Hansen (1951) — Benassy (1975) —

(1976) — Malinvaud (1977) — Muellbauer—portes

Stiglitz (1983), such as the aggregation problem

of quantity constraints.

theory,

inf ormed

deve

by

loped in

a thorou

and for

gh know

capitali

ledge of

of us wtheir institutions. Tho

served our apprenticeshi

s (and their language

or reason why this fie

ng scholars. Without

sibly choose the simpl

ful theorizing off the

of assumptions whi

listic". But reasona

er (indeed, agree to d

lifications, These wi

he particular model.

CPE

maj

you

sen

use

set

r e a

diff

simp

of t

economies, must

pean economies

Franklyn Holzm

detaile s of particu

well und that this i

conomics relatively

ackgroun er, one can

assumpti red to get

Even one may tak

not un y regarded

d well—i economists c

on the i and justifiab

ny case the objectiv

deller' of assumptio

that seem prima facie inappropriate to another may not ref le

se

ps with

s) can

ld of e

that b

ifying

ground

ch are

ble an

iffer)

11 in a

One mo

st market

East Euro

ho, like

d studie

ers tand

attracts

d, howev

ons requi

with it,

iversal 1

nformed

mport ant

depend on

s choice

be

and

an,

lar

5 a

few

not

any

e a

as

an

le

es

ns

ct

Seriousignorance or disregard of the relevant institutions.

3

criticism requires demonstrating rigorously either the

assumptions or (with evidence) that they

conclus ions.

by testing well—defined hypotheses.

All this does not deny but rather supports the validity of alter-

native approaches. One is a variety of institutionalism, seeking

distinctive general lessons in the specificities of socialist

planning, finding those generalities in a formalism which is not

primarily directed towards proving theorems or generating

testable hypotheses. The symbols used in this formalism are

often difficult to translate into measurable variables, and the

research strategy comes close to rejecting the standard model—

4

inconsistency of

lead to inappropr

The

uns

fac

macroeconomic

upported asse

ts" not based

the

iat e

s of CPEs has a

rtions purely

on the ,and c

on the casual

in arg or to

one below

erati support

s is ger acc

y lo fairly

CPEs. Empiri

eholds, planne

eriously unles

ttracted more

a priori rea

onfident,

empiricis

dismiss em

of fering

an essent

eptable.

dat a

most

ument

level

ons to

no lon

ng and

than its share of

soning, "stylized

elaborate verbal

m. For a theorist

pirical results or

various general

ially quantitative

We have for some

arguments founded

to use "obviously"

"absurd" is only

qualitative consid

proposition. Thi

time had reasonabl

series for s

behaviour o

should not

ever a 1

f CPE hous

be taken s

data. Disagreements

reliable macroeconomic time

cal assertions about the

rs, enterprises or markets

s they are supported by the

over empirical issues can be resolved only

The following

programme on

with alternat

roeconomic ap

measurement.

analysis of

possibilities

important spi

CPE policy pr

on macroeconomic

2. Theory

I defend the

eory, instit

red a fresh,

phenomena,

1 investiga

on planner

e 1970s and

mac-

and

the

of

era 1

into

t red

some

hypotheses—tests procedure (Ko

yields insights nevertheless,

empirical work within the more

below)

rnai 1980, less

as do a range

standard neoclass

so Iof

ical

n 1982).models

canons

Itand

(see

sectionsmacroecon

cons

omic

ider issues

equilibrium

raisin

ed by

CPEs

my own research

and comparisonsive views, explicit aggregative,

proach in th utional relationships

It has offe coherent framework for

many CPE opened up a range

for empirica tion, and generated sev

noffs: work s' behaviour insights

oblems of th early 1980s, which cen

equilibrium and threats to it and

developments in market economy macro

I take as commonbehavioural as we

economic behaviour

part of our maintai

the consumer goods

relations between

theory and econometrics.

ground for this survey that CPEs exhi

11 as technological regularities; sta

of CPE households, planners and enterprises

ned hypothesis. I limit the discussion here

and labour markets, while recognising t

enterprises and planners and disequilib

5

bitbleisto

hatri a

within the state productive sector have been th

interesting theory and empirical work (e.g

"cycles"). I accept the consensus from thi

normally exhibit generalized excess real demand

non—market character of transactions) in the

sector, for intermediate and investment goods.

here, however, to assess the relative importa

initial plans (including systematic "optimism"

and exogenous shocks) and systemic features 1

incentive system.

e subj

., on

s work

(quali

state

It is

nce of

about

ike an

ect of much

investment

that CPEs

fied by the

product ive

not relevant

overly taut

unknown data

output—based

Both demand for and

and supply functions

supply of

6

the

consumption goods and labour will

depend on events in the state productive sector. Total demand

for consumption goods will be affected indirectly by events in

the labour market and directly by the demand from state enter-

prises for some consumption goods. The latter is probably of

secondary importance (Kornai, 1980, p. 497), but the interplay

between planners, enterprises and banks which determines wage

plan (over)fulfilment is a key element of the story (Holzman,

1956, 1960; Portes, 1983). We discuss it and the supply side

further below. Whether they do jointly yield excess demand is an

empirical question. To answer it does require specifying demand

and hence the transmission of behaviour in

those functions, but not

at behaviour.

the state productive

internal processes wh

sector through

ich generate th

Although the labour market is equally important, both theory and

2.1. Definitions

data

diseq

are w

uilibri

and

is

eaker, so assessing the nature

um there is much less advanced. This

since the spillovers from the

labour supply could in principle

about the former. although eval

00 (Portes, 1974; Wiles and

of the chief reasons why the pla

nal observers should be interes

isequilibrium is this spillover

1951;

regrettable

market onto

information

difficult t

Moreover, one

well as exter

goods market d

work (Hansen,

typically

the distribution

extent of

particularly

ption goods

significant

it might be

ski, 1979).

hemselves as

consumpt ion

incentive to

consum

reveal

ua t i ng

Ro stow

nners t

ted in

on the

Galbrai

result in an ineffi

th, 1952).

cient allo

Excess

cation of

demand

goods

will also

and affect

What is macroeconomic

of real income; both will concern the planners.

disequilibrium and how do we recognize it?

rices

movem

eve

level

n

t

When p

expect

demand

elimina

hyperin

prices

either

positiv

(Winiec

or

te

fla

are

hid

e e

ki,

are free to vary

ents of the price

supply, even if

the dis equil ibrium

tions and as Karna

endogenously stic

den or repressed

xcess demand).

1985), but this

and practical

with some

o provide e

ice flexi

does not i

expects

ontrolled

on (I con

elessly c

sarily gi

such pr

(as it

1, 1980

ky or c

i nfl at i

Some car

unne c e 5

distinction.

stickiness

vidence of ex

bility may

n market eco

for CPEs).

we may ob

sider here

onflate the

yes up a us

A price index may be

we

ces S

not

nomy

When

serve

only

two

eful

conceptual

7

notund e

appr

lullexis

outp

supp

rat ion

demand

pricesmislea

rise in an

y— corrected

is simply "a

the aggregateexceed the cor

of goods and

an, 1974, p. 8

onary gap"

the "gap

unhelpful

suppre

opriate0

h

may "hide"

accurately calculated,

price index. Repressed

situation in which, at

demands for current

responding aggregate

labour services are

7)fl This excess real

whose consequences for

terminology can be

to say that rationing

sses it), since this

ly defined as above, is

an important point,

lation.

a global moneta

concept). This

called "monetary

actual stocks a

equilibrium on

balance of flow

household liquid

determinants.

falsified, may cover an unrepresentative basket of goods, and may

quali

inf lat

y weig

on the

any of these phenomenaty change

ion, i.e., a

hted, qualit

other hand,

ref lect

rlying

opriatel

at ion,ting wages

Ut and labo

lies... [so]

prices,

services

r chases

d Grossm

and

ur

pu

an

is th

presit i

ed (Barro

(a flow)

are sup

ding, and

e

5

5

"i

ed.

at

nflati

But

best

"hides" inflation

suggests that the pr

actually increasing.

that repressed infla

(rather than

ice level, appr

That usage als

tion may cause

obscures

idden inf

Rat ioning

aggregate

is one

level,

repressedis forced

inf latsaving

ion.

(note

At thethat as

consequence of

ts mirror image

ry magnitud

is a flow of

overhang").

nd by many ot

the consumpt

demand and

assets and se

y an aggreg

stock (the

by desired

we investig

we look at

the stocks

re only part

e, this is inherenti

saving rather than a

Flows are determined

her factors. When

ion goods market,

supply, of which

llers' inventories a

8

ateso—

and

ate

the

of

ial

Simply put, forced saving is that part of their flow of disposa—

ble income which household

Indeed, putting it so simply

it should ever occur — forcilike pushing on a string

holds voluntarily (maxim

goods) substitute (expec

current consumption; or

than they would like) i

their labour supply and

rationing) marginal uti

Note, therefore, that

sustained repressed inf

planning and households

"voluntary" component of

its only cause must be

c

y

spend but cannot.

ts the question why

sounds odd, rather

o reasons: house—

supplies of

unavailable

more labour

om reducing

reduced (by

induce.

hroni c,

central

ue, the

le, and

labour supply.

How important is this effect? Wiles and Rostowski (1979) query

it on mainly a priori grounds, except insofar as they claim that

9

s would like to

immediately promp

ng people to saveThere are only tw

izing subject toted) future consu

they involuntari

.e., they are cons

hen e income as m

lit

if there actual

lation since the

rationally expect

forced saving wil

inhibitions on cutting

of income

available

mption for

ly supply

trained fruch as the

would otherwise

ly has been c

beginning of

it to contin

1 be negligib

Households can of course reduce their labour supply qualitatively

("effort") as well as quantitatively, or quantitatively in ways

that will not be recorded in the data. This in no way affects

the essential argument., nor theoretical modelling of it: Excess

demand for consumption goods should have a spillover effect in

reducing effective (quality—corrected) labour supply, and hence

output, this is the basis of the "supply multiplier".

wage—elasticities of labour supply are observed to be low or

In theory,

disequilibr

above sugge

demand and

"notional" or

loning macro

macroeconomic

The definitions

between aggregate

y, forced saving

red holdings of

is one determi—

mediately arise

length. First,

negative. A simple calculation from equation (4) of Portes

(1981) shows, however, that the response of labour supply to

(rationed) consumption will depend on the difference between the

wage—elasticities of unconstrained and constrained labour supply

functions. These are likely to vary together, so the Wiles—

Rostowski claim shows nothing. Kornai recognizes that "the issue

is basically an empirical one," but then without examining

evidence asserts "that the intensity of shortage is not one of

the variables explaining labour supply either positively or

negatively (1980, p. 479)." My prejudice goes the opposite way,

along with Hansen and Gaibraith; but we should remain agnostic

until the evidence is in. Howard's (1976) results suggest a

significant effect, but I cannot accept them (nor his general

argument) for the reasons suggested in Katz (1979), Nissanke

(1979), and Portes and Winter (1980).

then, how should we measureium in the consumption goods market?

st trying to measure the difference

aggregate supply, or alternativel

(again,

f inanci

nant of

of whic

the

al as

exc e

h we

difference

sets is notss demand).

mention two

between actual and desi

what we want, although itThree major questions imand examine the third at

should we use

quantity—rat

"effective" demand and supply? The

literature suggests the latter

10

(Muellbauer and Portes 1978) but there are counterarguments.

"di scouraged consumer effect").mpirically

the correction (but

with the converse, the

2.2. The Aggregation Problem

e aggregation

"Only micro

rtifact which

unts." This

in principle

fessor might

1 artifacts —

nomic policy

at least in

I clearly do

believe there are interesting theoretical propositions and policy

11

Second, the formal or informal rationing schemes used under

excess demand may be "manipulable", in the sense that a household

may influence the amount it receives, in particular by presenting

a demand exceeding what it actually wants, knowing that the

probability of success or the fraction of revealed demand which

it will achieve is less than unity. This would exaggerate

measured excess demand. Weinrich (1984) considers the

theoretically appropriate measure of excess demand in these

circumstances for the stochastic rationing case, but we have as

yet no practical procedure for making see

Burkett, 1985, who deals e

Much the most fundamental and important issue is th

problem. Tartarin (1982) goes so far as to say,

disequilibria are real, macro disequilibrium is an a

exists only in the domain of the national acco

suggests not only that our measurement problem is

insoluble — even the most ambitious assistant pro

decline a research grant for measuring stat istica

but also that neither monetary theory nor macroeco

are of any interest in considering disequilibria,

CPEs. Having written on both (Portes, 1980, 1983),

interpretations in the aggregative economics of CPEs which

involve macroeconomic disequilibrium in essential ways.

policy cannot

structure of

Then how can

macroeconomi

disequilibri

clarify the

and in po

respons ibi li

we distinguish between distorted relative prices and

c—monetary causes of the phenomena we identify with

urn in CPEs? We have suggested thought experiments to

theoretical distinction (Portes and Winter, 1980),

licy terms we might consider the separate

ties of the Price Office and the Ministry of Finance

(monoba

suffice

relative

nk) and Ministry of Labour. Macroeconomic

to deal with problems created by the

prices, the system of incentives, etc.

we must come to terms with the problem of

e exce

ye and

ns is ts

ss demand

negative

Operationally, however,

identifying aggregat

there are both positi

and slacks. Kornai i

as vectors, and that

shortage and slack

"aggregate excess de

whose "normal state"

when

exces

it woul

...(1980,

mand" mi

was not

that they must rernai

d be a "grave mista

p. 45)." He wou

ght be well—define

pervasive shortage

ers have adjusted

at the micro level,

s demands, shortages

n disaggregated,

ke...to 'net out'ld accept thatd in an economy

In a "chronictheir attitudes

sive fo

found..

..( 1980,

shortage economy," however, "Buyto chronic shortages...where mas

spending and queueing are to be

is not an operational category.

12

rced substitution,

.aggregate excess

p. 477)."

forced

demand

The key point in this argument is the change in buyers'

behaviour,

(constraint)

arguments

underlying

Consider t

adjust, in the

expectations.

man i f e s tat ion,,

will substitu

unavailable at

consumption b

suppose, howe

altered. Shou

not in these c

do we charact

Moreover, woul

which Kornai c

expectations?

be defined md

the understan

expectations,

so

te

existing rela

ehaviour are

ver, that mdi

ld we say that

ircumstances a

erize macroe

d not the "ye

ha rac t e ri z e s

We would agr

ependently of

ding that th

if any) held by agents in

preferences

onic shortage.

Workers will

(cons traint)

ffect is one

our market. Others

hose preferred but

dity preference and

We do not normally

supply functions have

ss supply of labour is

category"? If so, how

ps in the 193Os?

indicators" with

cted by quantity

ss demand cannot

the economy, on

tity constraintu

constraint in

which w

expect at

of their

those func

he case of

that

These

their

expecta

quantity

tions are

e interp

ions are

demand

tions ar

long—run

sense of

The " isco

me p e wi

(accept)

ret to mean

dif ferent.

functions,

e not alte

mass unem

changing

uraged w

11 leave t

d

eopl

but the

red by chr

ployment.

quant ity

orker" e

he lab

jobs for t

qui

inferior

tive wages. Li

also affected

vidual labour

aggregate exce

n "operational

conomic relationshi

ctor of disaggregated

shortage also be affe

ee that aggregate exce

the "normal state" of

is includes the q an

the economy.

The existence of forced substitution with excess demand is not a

major theoretical problem. After all, at the aggregate level,

both forced saving and the response of labour supply to a

the goods market are examples of forced substitu—

13

tion. Spillover effects of this kind have been modelled exten—

sively, and

taking plac

forced subst

substitution

Although it

effect will

case stop us

in demand are low

is always welfare—

depend on the elast

from doing applied

and it is nevertheless

reducing, the importan

icities, and it should

macroeconomics.

extensive.

ce of that

not in any

We all need to use macroeconomic aggregates, even Janos Kornai

discussing CPEs (1980, Secs. 16.2—16.5, 17.6, 18.5, 21.1—21.2,

14

all quantity—rationing macro deals with transactions

e at "false" (disequilibrium) prices. In practice,

itution is a cause for worry only if elasticities of

In seeking to measure aggregate excess demand, we cannot avoid

"netting out" shortages and slacks. This does indeed present a

problem in theory, as suggested by the Barro—Grossman definition

of repressed inflation quoted above. In a two—market model, they

require that both be in excess demand. This stipulation that the

vector of excess demands be semipositive (so no "netting out" at

false prices is necessary) goes back at least as far as Hansen

(1951), and it is reflected in the consistent usage of the

quantity—rationing macro literature: excess demand for goods

with excess supply of labour is "classical unemployment", a quite

different regime (the converse is "underconsumption"). In

practice, however, the early period of central planning, when

most goods were in excess demand and labour in excess supply, isbetter characterised as "repressed inflation on the goods market"(Kornai, 1980, Sec. 11.7; 1982, pp. 102—108).

and Chap. 19 and the explicit macro modelling of 1982, passim).

cond It

minimi

excess

econom

imposs

the ag

Sec.

caref

ion" (

mum of

suppl

ies not

ible to

gregati

12.1).

ully. a

I n problem as

uantity—ratio

agents on th

transact io

e quantity

and "discret

nd regimes.

shortages,

oeconomics with

Deaton and Mue

must simply

open to empiri

results are con

o disequilibria.

ng Kornai's conc

sufficient reason to

ning macroeconomics:

e short side of the

ns) , the "minimum

transacted is thee switching" between

Even for markethowever, it is quitea purist approach tollbauer, Chap. 6 and

do macroeconomicscal evidence (not atradicted when one

Yet he

reject

the "5

ma rke

would take th

the central c

hort—side rul

t achieve th

e aggregat o

oncepts of q

e" (that all

eir desired

he aggregatand demand),

excess demachronic

that t

supply

y and

subject todo applied

on problem

Instead,

lways remai

macr

(see

one

ning

itsicrusi

priori

allows

Burkett

assertion)

explicitly

's (1985) s

that

for m

tudy

perfectly

In fact,

epts gives

consistent with our own; see Sec. 3.4 below.

however,results

One internally consistent way of approaching

context is "smoothing by aggre

sketched by Muellbauer (1978).

labour markets, for example, s

excess demand, we find that tota

and vacancies, total supply

unemployment, so total excess

unemployment; as the real wage

falls in a continuous manner,

15

gation", whose 1If we aggregat

ome in excess s1 demand is the

is the sum of

demand equals

rises, aggrega

the problem in our

engthy pedigree ise over many microupply and some insum of employment

employm ent and

vacancies minus

te excess demand

with a smooth unemployment—

vacancies tradeoff. Aggregate employment will always lie to the

left of the wedge formed by the aggregate demand and supply

curves

2.3. Macroeconomic Models

I do no

models

of mac

in Por

has on

only h

market

ud

lete

ment

own

It

ur;

ce)

and

A diagra

aggregate

m in Portes

consumption

and

goods

Winter

market

(1980)

in CPEs.

applies this

Note that the

to the

larger

the elastic

forced subst

approximate

which would

cons equently

condition e

approach has

itution, the closer will observed aggre

the underlying aggregate demand and

be observed in full Walrasian equi

the more suitable is the discrete—sw

mpirical model. The "smoothing by

been developed further by Malinva (

stronger is

gate behaviour

supply curves

librium, and

itching, mm—

aggreg ion"

1981), Kooiman

(19

app

dat

and

84), Gouri

ears espec

a indicatin

in excess

eroux and

ially prom

g the propo

supply.

Laroq

is i ng

rtions

ue (

forof

1985) and Mar

applications

micro markets

tin (19

where t

in exces

86). It

here are

s demand

t have space to discuss in detail the alternative comp

within which one might consider the theory and measure

roeconomic disequilibrium in CPEs. I have set out my

tes (1979, 1981); Wolf (1985) discusses it and others.

ly two aggregate markets, for consumption goods and labo

ouseholds hold money; there is no "parallel" (flex—pri

there are only two categories of agents, households

16

and foreign

households'

there is no

based on t

a coherent

ibrium and a

goods market

plan construc

1983, 1984a),

between this mo

ian general equi

(1980,

of plan

trade are

quantity

incorporated, but to

expectations are not

planners, inventories

only a limited extent;

treated explicitly; and

macroeconometric model

nevertheless provides

macroeconomic disequil

it in the consumption

specification of the

in Fortes et al. (

major similarities

both are non—Walras

the "real" and the

mechanisms for adj

that Kornai

his theoreti

theoretical

n adequate ba

as we have

tion process

Dlouhy (1984

del and that

librium model

and both

equilibri

Secs. 19.4 19.6) gi

construction, especially

ex ante

attempt to specify a complete

cal framework. It

representation of

sis for measuring

done. Adding the

developed further

) suggests three

of Kornai (1982)

s; both deal with

have endogenous

um. I would add

yes a similar

in assuming that

to avoid excess

con t r o

ustment

1" spheres

towards an

representation

the planners are rational and do try

demand, ceteris paribus. Stupnicki (1985)

model with Kornai.

Several alternative models have been prop

assumptions. Fickersgill(1980a), Duchene

and Nuti (1985) explore the implications o

Hare (1982) incorporates some of Korna

framework of Malinvaud (1977), which is

Brada (1982) assumes that enterprises also

(1983) specifies a larger macroeconom

consumption block resembles our treatment.

17

carefully compares my

osed with different(1984), Henin (1984)f a parallel market.

i's ideas into the

similar to our own.

hold money. Hulyak

etric model whose

There remains much

work to be do

inventories, and

little advant

planners and

them. This

behaviour, b

Campbell (1

establishment

firm under t

transactions

between hous

point of view of

258) ."

There are n

elements of

tarism are often

economies.

ne with quantity

foreign trade. For

for the essential

ng assumptions are

nt set sufficient for

some propositions and

esis testing. I would

just about rich enough

Some have queried the

lassical macro theory,

11 as disequilibrium

that they and a simple mone—

CPEs than to comparable market

constraint expectations,

our purposes, however, I see

age in breaking up the state productive sector into

enterprises and modelling the interaction between

is clearly essential in modelling investment

ut not for the consumption goods market. I follow

970) in believing that "the state production

[SPE] can best be considered as a single gigantic

he unified direction of the planners...intra—SPE

...lacking the market characteristics of those

eholds and the SPE, are of less interest from the

pricing and macroeconomic equilibrium (pp. 255.

a priori guidelines

model. All simplifyi

o general,

a CPE macro

"wrong", so one chooses a minimal cohere

one's objectives. Nine were to formalize

to provide a basis for econometric hypoth

claim that the institutional framework is

for the purpose and not seriously flawed.

basic strategy of applying to CPEs the neoc

in its quantity—rationing version, as we

econometrics. I believe, however,

better suited to

18

2.4. Behaviour of Households and Planners

and Asselain

STEs.

On the other ha

between repressed

the "suction syste

different insti

regularities are

the differences he

of firms' behaviour

Rostowski

that

and

PEs

so

ral

ing

old

nd, Kornai specifically rejects the analogy

inflation in a fix—price capitalist economy and

m" of a socialist economy, "for the latter has

tutional conditions and its behavioral

therefore different, too (1980, p. 558)." Of

cites, however, all but one relate to aspects

not relevant to the consumption goods market,

19

Others take a similar line in this context. Wiles and

(1979) de

"led him

macroecon

(p. 62)."

far incr

relation

Western

scribe the institutional characteristics of CPEs

[Portes] correctly to suppose that Western micro—

omics...would be fully applicable to households in C

Pickersgill (1983) states, "The work I have done

eases my confidence in the similarity of behaviou

ships and therefore of the appropriateness of us

methodologies in the study of Soviet househ

be ha vi our

behaviour in

(1981) takes the same view of saving

Modelling the planners is more difficult

vious analogies in neoclassical theory.

plan—construction process with an explici

tion (Portes et al., 1983), which incorpo

objective like that in Kornai (1980, p.

appears to be successful empirically

and

We

t pla

rates

490,

there are no ob—

extend our basic

nners' loss func—

a steady growth

and 1982). This

substantially above the natural rate continuously.

Moreover, just as the rate of unemployment is useful though crude

as an index of excess supply in the labour markets of capitalist

countries and is usable in cross—country comparisons despite data

comparability problems, so we can construct a similar index for

excess demand on the consumption goods markets of CPEs. We take

the demand and supply functions obtained by disequilibrium

estimation (see below), calculate "fitted" demand and supply for

each year, and take their difference as a percentage of observed

20

nai does

and repre

acc

ssed

ep t

i nfthe symme

lation in atrycap

bet

ital

while the last concerns the adjustment of buyers' behaviour to

shortage, discussed above. I therefore maintain that this

analogy is indeed instructive, and our intuition is not

misleading.

Although Kor ween Keynesian

unemployment ist economy, he

rejects any symmetry between the former in a capitalist economy

and the latter in a socialist economy (p. 478). He maintains

that suction in a socialist economy is sui generis because its

"mode of operation, its management system, and the behavior of

the agents within it have adjusted o chronic shortage which is

continuously maintained by the ins itutional relationships (p.

478)." Again, we suggest above that similar adaptation occurs in

long—run mass unemployment, and one might think that Marxists

would go even further to say that the institutional relationships

of a capitalist economy are such as to maintain unemployment

tt

consumption (Portes and Winter. 1980 — an alternative version,

Finally, the relations between

macroeconomi

demand is ge

The planners

generate taut

to overfulfil

and exogenous

of consumer g

accept much

consumption g

admits the po

gives some ins

The view that consumption plans

balanced, optimistic) and hence

is decisively rejected by aggreg

and planned consumption in three

supply (Cs) will always fall sh

(CD) will always exceed it, so

Cs) would suggest that actual C

eory of

excess

1983).

system

d pressure

by errors

and supply

Although 1

ction for

xplicitly

an. This

work, nor

are chronically unrealistic (un—

underfulfilled (Winiecki, 1985)

ate time—series data for actual

CPE5. Some argue a priori that

ort of the plan (C*) and demand

CS < C* < CD. Then C = mm (CD,

< C* always. In Poland, however,

based on stochastic simulation of the fitted functions, is

calculated in Portes et a1. 1983). This measure can also be

used in mntertemporal and international comparisons.

CPE institutions and the th

c equilibrium appear in my analysis of how

nerated, reproduced and controlled (1977a.

objectives and the enterprise incentive

plans in the state productive sector an

1 them, and these forces are exacerbated

shocks. This may affect the demand for

oods, but less than is commonly thought.

of Kornai's analysis of plan constru

oods (1980, Secs. 16.1—16.4), my model e

ssibility of an ex ante "unbalanced" p1

ights but has no effect on the empirical

does

diseq

the precise source of consumption goods market

uilibrium.

21

19 57—7 5

second

e 24 yea

957—7 5,

ly, Kor

osely (

Poland.

by which the plan was exceed

Poland is

control.

until the Gierek

half of the 1970s.

rs 1957—80, and in

then once more in

nai's general view

1980, Sec. 16.1) is

He shows that the

ed in 1957—70 was

systemic features and institutional regularities in a shortage

22

we find

policies

For Czec

Hungary,

the follo

that wage

borne out

C > C' for 10 of the 19 years

ensured C < C throughout the

hoslovakia. C > C in 15 of th

C > C in 11 of the 19 years 1

wing quinquennium. Similar

plans are fulfilled very cl

by Farrell's (1975) study of

average

2.6%, al

lax in m

amount

though

onetary

Kornai's views, however, on the roles of relativeI cannot accept

prices and of price rigidity in creating and maintaining

shortages. He cites for example. groups of products for which

the relationship between price and average cost goes in the

opposite sense of the intensity of shortage as "a new empirical

argument against the view that 'shortage exists when relative

prices are too low' (1980, p. 500)." That argument ignores the

demand side and is hence a glaring non—sequitur. Equally, as

Gomulka (1985) points out, Kornaits assertion that "there is no

behavioral regularity ensuring that a rise in the price level of

consumption goods eliminates the shortage (p. 497)" simply doesnot follow from the a priori arguments given (no evidence iscited); and it is inconsistent with his view that the household

faces a "hard" budget constraint. Yet these are key elements in

his downgrading of the role of prices and because of

economy. I regard them as unsupported assertions.

3. Measurement

Only empirical testing can ultimately resolve the basic issues to

which this analysis is directed. Assertion and casual empiricism

will not do. To arrive (ideally) at an agreed interpretation of

the results of such tests, however, the theoretical framework

within which they are carried out must be broadly agreed as well.

The discussion above is intended to extend the area of agreement.

3.1. Econometric Results

Our own examination of the data, tests, results and conclusions

are set out in a series of papers (Portes, 1974, 1977a; with

Winter, 1977, 1978, 1980; et al.1 1983, 1984a, 1984b, 1985).

shall not attempt to summarize them, although I shall reply below

to the relatively few specific criticisms which they have

attracted. There seems to be an irresistible temptation,

however, for those who reject our methods or our conclusions to

misrepresent both by setting up straw men which can easily be

tumbled without taking the effort to address specific issues of

theory, econometrics or data. It may therefore be useful simply

to quote for the record a few summary statements from these

papers:

23

1. "...theresometimes open,Moreover, therecountries during

no evidence of

Soviet—type CPEs

was sign

in the

may have

particula

sustainedsince the in

if icant inflation.) both

CPEs until that time

been brief episodes

r periods since then..

repressed inflation

id—1950s (Portes, 1974

repressed and

[mid—1950 5]

in particular

[but] there is

in any of the

)U

our

old sare

ifyttle

pen

24

avings

the pla

pothes

n be an

sectorof th

11 degr

nd] we

equilibrium

functions in

usibility of

is that thealysed using

in Westerne data...The

ee of marketcan draw no

we treat both

2. "We have specified

hous

The

ehold

5uc

and estimated [in an

framework]

four CPEs.

the estimat

behaviour of

the same me

countries...

residuals ca

equilibrium

further concl

sides o

1 7 ) ."

demand

cessful

port

us eh

that

test

us li

the

about

es sup

the ho

thods

[and]

n tell

during

us ions

for money and s

estimation...and

underlying hyector in CPEs ca

used for thisto the qualityabout the overaod, however...[a

repressed inflation untilf the market in a disequilibrium framework (P—W 1978, p.

3. "...Begin with the estimated

observation was generated by an excess

allocating each year...on the basis of

or falling short of] 0.5...[indicates]

dominant regime in three out of the fo

very crude description of the results,

probabilities [that an

demand regime] ...Simplythe probability exceeding

excess supply was theur countries. This is a

however. More precise

efforts at hypothesis testing are made [below, from which...we

of serious empirical

25

quoted) dismissal of

as such.

conclusions...[but]cannot

be ii ev

hypoth

consum

1970s I

we hay

1970s

ext end

other

draw strong, unambiguo

e that the evidence..

esis of sustained repr

ption goods and service

in our four CPEs (P—W

e not stated any such v

especially in respect

our econometric analy

work suggests

goods market in recent years

us empirical

.clearly ju

essed infla

s since the

1980, pp. 1

jew for the

of the USS

sis because

rising pressure of

St if

ti on

mid-

49, 1

pen

R (t

of d

dem

we

ies rejecting the

in the market for

1950s [to the mid—

55—6)." Note that

od since the mid—

o which we cannot

ata problems); and

and on the Soviet

consumption

Contrast our actual conclusions with Wile

mtinfpla

con

Kor

fin

in

erpretat ion

lationary

nning...(p.

dude [that

nai (1982):

dings...'On th

three out of

(see below).

s and Rostowski's

of them 'Yin Portes's view there is no demand

pressure in STEs because of successful

61)." Tartarin (1982): P—W "emphatically

there was] general excess supply (p. 25)."

"I shall quote one of their important

is basis excess supply was the dominant regime

the four countries.' In my opinion [emphasis

conclusion is absurd. All four countries should be

chronic—shortage economies...(p.35)" Such distortion

work, with the (widely

added], the

considered

it as "absurd", is not worthy of reply

3.2. Validity of Nacroeconometrics for CPEs

How do we and others obtain well—behaved functions, significant

and plausible parameter estimates, reasonable out—of—sample

performance, etc.? Are all these results merely statistical

artifacts? If one does admit they might have some validity,

however, even some of the most cherished "stylized facts" must be

queried. Podkaminer (1982), for example, using a model totally

26

Let us now turn to substance. Kornai

macroeconometrics for CPE markets

to be chronic and widespread, f

econometric technique can work:

trade} to d [initial demand] becos

rejects from the outset all

believes

circums

shortages

tances no

in which he

or in these

"the Inference

es Inadmissible

effort" (1982, p. 98). Consider the

we and others do

ns drawn from the

then shortages

Yet this does

from y

(1980,

contrapobtain

ory as

must be

eem to

[actual

p. 98)'"

ositive

results

well as

either

be the

however: If

on restrictio

etric criteria

not chronic.

CPE macroeconomet

Woifling and others

it is a "vain

of this proposition

which satisfy a pri

conventional econom

not widespread or

case. The large

Szakolczai, Sujan,

easily dismissible.

ing on a smaller sc

consumption goods

Charemza and Cronic

(1982), Podkaminer

Hungary, Poland and

ncare

5

models

well-kn

For care

ale which

markets,

ki (1983),

(1982), an

ful estimation and hypothe

seems to yield coherent r

we cite (only by way of

Hulyak (1982), Klaus and

d Timmer (1982); this wo

of Welfe,

own and not

sis test—

esults on

example)Rudlovcek

rk coversCzechoslovakia.

different from our own.. gets results which suggest aggregate

labour]...(Dlouhy, 1984)."

3.3. lation and Forced Savings

excess supply of consumption goods in Poland up to the early

1970s (revised estimates in Podkaminer 1984, still allow

equality of aggregate demand and supply in the late 1960s). The

conclusion may look rather less "absurd" if reached by various

routes with different data Sets. Klaus and Rudlovcek (1982)

summarize their results unambiguously: "We strongly believe that

global equilibrium at the consumption goods market was the

prevalent characteristic of the Czechoslovak economy in the last

quarter of the century." Others are simply not willing to accept

what Kornai regards as obvious: "...we do not find useful (and

even realistic) to accept the assumption of permanent (or

'normal') shortage on these markets [consumption goods and

a few specific topics before passing to the contribu—We now

t Ions

indicaPort es

for th

(1981)

fasterp.a. f

take

of disequil

tors of sho

(1977a) cite USSR over

shows thei

than the "o

or Bulgaria

ibrium macroeconometrics on the one h

rtage on the other. First, hidden inf

es estimates by others of slightly over

two decades. More recent work by Altonr "alternative" consumer price index gfflcial implicit consumer price index"

1.6% p.a. for Czechoslovakia, 1.3%

and and

lat ion.

1% p.a.

et al.

rowing

by 2.4%

p.a. forthe GDR, 1.1% p.a. for Hungary, and 1.6% p.a. for Poland.

27

These

are unlikely to be underesti

with unsupported remarks like

inflation in Poland in 1975 at

of nominal incomes and the CPI

actually fell in that year!).

mates.

that in

8% (whic

would

They should be contrasted

Nove (1979) putting hidden

h, with official estimates

suggest that real incomes

unjustifia

ange in i

low), long

learly dem

Asselain shows

a long period

e ratio, if in

economies with

income but dif

ating faster w

view of Wiles

Soviet—type

it were the

secure grou

authors (e.g.,arising from

atlo of savingsretail sales

a) and Asselain

ye meaningless

onstant savings

ncrease in thelow; and that

rate and rate oftes, the economyr wealth—income

[Portes] rightly

SavingsextensivWi ni e ck

beh

ely

1

avi our

with c

1985)

of CPE households has been investigated

onventional me

continue the

thods.

confus

ble me

ncome

afterons t ra

asu

0

Port

ted

res

r

es

th

Some

ionsthe r

h to

1977

y gi

tacle i

are

ng 5

like

ealt

1974,

t the

y tha

derab

o cks

savi

w

(

a

impeccabl

of consiitlal st

the same

theoret ically

(flow) to ch

(consumption f

(1981) have c

answers. Thus

rate implies

wealth— incom

comparing twogrowth of realwhich is inflratio. In the

ridicules theincome) as ifmay even be on

Soviet—type bankers,

nt inflation ra

exhibit a lowe

Rostowski, "He

fere

ill

and

habit

Keynes

nd in 5

of

ian

et t

quotingmarginal

ing up hi

savings! (ch

propensity.

s judgment

ange in

..Port es

againstby the ratio

trade (1979,

who are alarmed

liquid savings to the flow of retail

28

of the stock ofpp. 64—65).'

Serious work on household assets and savings in CPEs seems

in not showing much e

on this

dicate t

HPS in

rced sa

r

in

he

fo

no

te

d

could show

Ofer (1980)

olds save in

consumer go

lar views in her

study evidence

and empirical analysis

hat "The full set of

of forced saving.

hould not be surprise

the hypothesis of a

in Western market e

) get good results

lovakia, 1965—81 the

aving ratio was not

d not seem to depend

ho constructs estima

SR, finds an overall

t ratio (as elsewhere

dity is not excessive

There is no agreement

we should expect savings rates

vidence of forced

for the USSR, but

hat there may have

the late 1960s and

ving. Her cross—

t the proposition

amounts of their

to purchase," and

vey of both time

Asselain's (1981)

ds him to conclude

racteristics of a

thus never found

t econometric work

gs—income relation

es (p. 43)." Klaus

standard savings

asting perf ormance

by international

pply—side events.

f household cash

of decline in the

tern Europe), and

5)" by comparison

authors, however,

in CPEs to be lower,

ill

ies

up

(1976)

results

ward shi

is clea

(1980)

ft in t

remarkably unanimous

Pickersg

er time—ser

substantial

1970s which

n work with

oviet househ

for lack of

presses simi

some

"does

ordina

ods an

t supporly large

services

later sur

(1983).

saving.

her lat

been a

early

sectio

that S

income

she ex

sen

care

for

typ

tog

see

of

and

fun

was

com

Pee

hol

cur

tha

wit

es

ful

sev

i cal

ether

ks fi

the 5

Rudl

ct ion

good

paris

bles

ding

renc

t "S

h ot

and budget

theoretical

eral CPEs t

situationand one s

rst to test

ame type as

ovcek (1982

for Czechos

and the s

on and di

(1981), w

s in the US

y to deposi

oviet liqui

her

lea

cha

..are

d thasavin

conomi

for af o rec

highon sutes 0trendin Eas

(p. 7

amongCPE 5.

on whether

29

higher, or about the same as in their market—economycounterparts. It should be stressed that none of these studies

pretends to test for excess demand, nor should they be so

interpreted. They and the work cited in Sec. 3.2 are simply

consistent with taking equilibrium as a working hypothesis.

Kornai (1980) asserts that the "Clower—Barro—crossman train of

thought" provides "theoretical background for the argument" that

the "time path of savings in the household sector [can indicate]

whether the consumer goods market on the macro—level is in a

state of aggregate excess demand or aggregate excess supply (pp.

477—8)." This is puzzling. He is presumably referring to our

work, yet the precise object of our analysis in Portes and Winter

(1978, 1980) is to use the quantity—rationing macro model to show

why and how in disequilibrium, this sort of argument may give

quite the wrong answer. Even more inexplicable, therefore, is

his own use of departures from trend of household savings as one

of his indicators of shortage (1982, p. 139) — an aggregate

indicator, at that!

3.4. Disequilibrturn Econometrics

Now we turn to disequilibrium econometrics (see Quandt, 1982, for

a survey). Part of the maintained hypothesis is that one

observes only the quantity transacted, and that supply and demand

are latent variables to be estimated, along with the functions

which generate them. Kornai, too, believes that it is impossible

30

to separate consumption supply and demand in the observed data,

and seeks himself to measure it.

Proper hypothesis testing in this area requires a clear statement

equation, however, we have

implicitly assumes that

ur market. This was not an

outhakke r—Taylor function

h was easy to estimate, did

that they are latent variables,

ption. Yet he

tion", with b

right—hand si

ed form if one

We believe it

irectly — they

and

co nc

oth d

de.

real

is

are

that they

ludes that

emand—side

It is diff

ly admits t

pref erable

in principle

combin

there

and

icul

he p

t

mine actual consum

"consumption func

variables on the

such a quasi—reduc

disequilibrium.

latent variables d

Kornai must agree, since he regards shortage as a latent variable

e to det

is only

supply—s

to just

sibility

timate

ureable,

e r—

one

ide

ify

of

the

as

tOs

o es

meas

of the maintained hypothesis, the null, and the alternative. The

maintained hypothesis must explicitly admit an alternative to

excess demand — otherwise, it is logically impossible to test for

its presence. This fundamental point leads us to reject Howard's

(1976) results and to find unhelpful the model of Brada (1982),

which in this respect begs the basic question. Our own main-

tained hypothesis does allow excess demand, equilibrium (on a set

of measure zero), and excess supply in the consumption goods

market.

In specify

chosen a

households

empirical

because we

ing

K e

are

jud

ne e

our consumption demand

ynesian" form which

constrained in the labo

gment. We chose the Hded a specification whic

31

not exhaust too many degrees of freedom or require great atten—

estimates

"full emp

on Western economies

1 oyment".

ima

gni

uld

ore

we

in

he

re

To

ta

ur

Ours are not the

consumption good

suggest several

implement a "qua

models we use,

tests, and both

quite robust.

hypothesis" of

only versions of disequilibrium mod

s markets of CPEs. Charemza and Qu

alternatives; and Charemza and Gron

si—disequilibrium" model for Poland.

however, we have performed a wide

our theoretical framework and our r

For example, it is possible to test

continuous excess demand against our

32

els for the

andt (1982)

icki (1983)

Within the

variety of

esults seem

the "Kornai

more general

tion to the error structure, and gave a few clear prior restric-

tions on the coefficients, violation of which would tell us that

something was seriously wrong. This standard function permitted

ready international comparisons and avoided difficult data prob-

lems. If one were to reject it on the grounds that the estimates

will be biased unless there actually was excess supply of labour,

one would have also to reject almost all consumption function

which might ever have experienced

As remarked above, our est

detail. We ourselves reco

empirical model which sho

research; to incorporate f

dispel any misconceptions,

(which raise great problems

econometric work.

tes have not been challenged

ze important extensions to t

be a high priority for futu

ign trade and inventories.

have never used inventory da

e.g., what is "saleable"?) in o

alternative. Our results clearly reject the restriction

ions, if they in principle

John Burkett (1985) has made Kornai operational by ingeniously

specification

the discouraged

slack, all sepa

subs titution.

countries, are e

the category of

slack exceeds sh

and for 1956—80

slack for 13 of

direct estimates

well with our

aggregate excess

ortage throughout

(except 1972) in P

the 24 years 195

of shortage, net

estimates of agg

(Portes, Quandt and Yeo, 1985). Kornai would doubtless reject

the test, because he believes that the aggregation problem

invalidates any discrete—switching, mm—condition model. As

suggested above, taken too seriously, the aggregation problem

invalidates all empirical macroeconomics. On his hypothesis,

however, we could never obtain information about the latent

consumption demand variable. But it is then inexplicable how we

(and others) continue to obtain well—identified, well—behaved

demand funct do not exist or are

unobservable.

developing methods fo

effect" and with the

r dealing with

coexistence

directly estimates po

consumer effect)

rately, allowing

His results, co

ntirely consistent

the "discouraged consumerof shortage and slack. Histential demand (allowing for

total shortage, and totalfor spillovers like forced

vering five East European

with those we obtained using

demand. He finds that total

1957—78 in the GDR and Hungary

oland, while shortage exceeds

5—78 in Czechoslovakia. His

of slack, correlate remarkably

regate excess demand. Even

demand does not changeallowing for potential (but discouraged)

33

this picture.

Thus Burkett c

hypothesis.

Czechos lovaki

consumer dema

subs tant ially

supply. Thus

theory .... c

evidence of p

claims that "t

almost tautolo

cavalier ap

oncludes "One cannot reject the di

Shortages exist but except i

a are usually overshadowed by sl

nd, purged of the discouraged cons

greater than actual demand but sel

application of techniques sugges

an yield conclusions consistent

revalent aggregate excess supply

he spurious precision of advanced

gically defines imbalance out of

proach to rigorous empiric

switching

case of

Potential

effect, is

as great as

by Kornai's

th Portes's

". Nuti (1985)

econometrics

existence." This

al testing is

egate discrete

by Burkett's

the micro as

's indicators of shortage.

saggregated level, in orde

acterize shortage. Yet he

be highly correlated (19

cro" (synthetic) index of

series of estimated exc

The simple correlation

oints are quite similar.

says we

roperly

ects the

p. 15).

tage for

demands

in fact

regards

a discrete—

screte

n the

ack.

um er

don

ted

Wi

uns

swi

res

wel

ubs tantiated

tching mode

ults from a

1 as the mac

by

1,

no

ro

any dire

and it

del which

level.

ct

is

in

criticism of

a fortiori

cludes mba

the aggr

rejected

lances at

Finally, we turn to

need many of them,

and comprehensively

individual indicat

Moreover, he finds

Hungary close to o

(Portes and Winter,

r = .604, and the tu

Ko

at

to

ors

his

ur

198

mi

rnai

a di

char

to

"ma

own

0).

ng p

He

rp

exp

82,

shor

ess

is

He

ours as "invalid", because they are derived from

34

switching macro model. We had argued previously however, on

"smoothing by aggregation" grounds, that one could expect a well—

defined correspondence of just this kind (P—W 1980), so we were

not surprised. Burkett's (1985) results are further, perhaps

definitive confirmation that our approach using aggregate excess

demand produces results which, so far from being "invalid" or

"absurd", are in fact consistent with and complementary to moredisaggregated methods.

We finish on thi

and empirical prin CPEs may now

disagreement bet

policy implicatiof excess aggreother problems 1— see Pryor, 197

getting closer.

4. Conclusion

The line of research surveyed here was originally motivated by

scepticism towards the conventional wisdom that the household

sector of CPEs has been subject to sustained repressed inflation

since central planning began. This conventional wisdom has since

been elaborated and defended by Kornai indeed elevated with

considerable flourish into a general characterisation of CPEs as

35

s reassuring point of agreement. The theoretical

oblems of measuring macroeconomic disequilibrium

be somewhat clearer, the areas of agreement andter defined. The main practical question, withons, is still to distinguish between the effects

gate demand and distorted relative prices (or

ike the inadequacies of the distributive network

7, Skurski, 1972, and Turcan, 1977). We may be

is much which is stimulating in this

ned currency from the worseningia in some CPEs (especially Poland)

it distorts the facts over the whole

Os and thereby reinforces confused.

rate forecasts and unjustified policy

conventional wisdom, the underlying

ation and become immune to challenge

buttressed by the evident political

These are not just academic

macroeconomi

developments h

microeconomic

external balan

borrowing was

external debt

Marer, 1985)

consumption

macroeconomic

that investm

disequilibria

misleading in

Stupnicki (1985

Portes presents clues how

e countries

itioned the

Proper unde

acroeconomic

in foresee

the Polish

ho took th

er" would

of the earl

are the primary

old sector would

right. Macroeconomic

ho r t a g e

analysis,

macroecon

since the

period si

incorrect

conclusio

truths are

from the

appeal in

economies". Therewhich has gai

omic disequilibr

early 1970s. But

nce the early 195theorizing, maccu

ns. As with most

belied by exagger

data. This is

such a critique of central planning.

disputes. It is important to get the

cs of thes

eavily cond

reforms.ce and the m

essential

problems and

Those w

as a "buff

adjustment

ent cyclesin the househ

interpreting

) concludes in

evolution of the Hungarian

rstanding of internal and

bases of trade and foreign

ing and interpreting CPE

crisis (Portes, 1977b and

e conventional view of

have misunderstood CPE

y l980s. The proposition

cause of macro

have been equally

the

his

past several years. And as

Kornai—portes comparison, "only

to improve the situation".

36

I have set out the theoretical innovations in Fortes (1981.

1983). There are, however, equally significant practical lessons

to be learned from eschewing the simple certainties of the

"shortage economy" in favour of a more complex, more nuanced and

better balanced view of the realities of CPE macroeconomics.

37

REFERENCES

Asselain, J. C., 1981, Mythe ou realit de l'pargne force dansles pays socialistes, in: M. Lavigne, ed., Travail etmonnale en systeme socialiste (Paris, Economica).

Alton, T. et al., 1981, Official and alternative consumer priceIndices in Eastern Europe, Research project on NationalIncome in East Central Europe, OP—68.

Barro, R. and H. Grossman, 1971, A general disequilibrium modelof income and employment, American Economic Review 61.62—83.

Barro, R. and H. Grossman, 1974, Suppressed inflation and thesupply multiplier, Review of Economic Studies 41, 87—104.

Benassy, J. P., 1975, Neo—Keynesian disequilibrium theory in amonetary economy, Review of Economic Studies 42, 503—24.

Brada, J., 1982, Real and monetary approaches to foreign tradeadjustment mechanisms in centrally planned economiesEuropean Economic Review 19, 229—244.

Burkett, J., 1985, Slack and shortage in socialist economies,mimeo, Brookings Institution.

Campbell, R., 1970, Macroeconomic models and central price—setting in the Soviet economy, in: Gehrels, F., et al.,Essays in economic analysis and policy (Bloomington, IndianaUniversity Press), 253—270.

Charemza, W., and R. Quandt, 1982, Models and estimation ofdisequilibrium for centrally planned economies, Review ofEconomic Studies 49, 109—116.

Charemza, W. and M. Gronicki, 1983, Rational expectations wageillusion and consumption excess demand: an empiricalinvestigation for Poland, Birkbeck College Discussion Paperin Economics no. 143.

Clower, R., 1965, The Keynesian counter—revolution a theoreticalappraisal, in: Hahn, F.and F. Brechling eds., The theoryof interest rates (London, Macmillan).

Deaton, A., and J. Muellbauer, 1980, Economics and consumerbehaviour (Cambridge, C.U.P.).

Dlouhy, V., 1984, On the problem of macroeconomic equilibrium incentrally planned economies, paper presented to EconometricSociety European Meetings, Madrid.

Duchene, G., 1984, Marchs parallles et d6squilibre dans leseconomies soclalistes, in: X. Richet, ed., Crises l'Est

1

(Lyon, Presses Universitaires de Lyon).

Farrell, J., 1975, Bank control of the wage fund in Poland: 1950—70, Soviet Studies 27, 265—287.

Galbraith, J. K., 1952, A theory of price control (Cambridge,Mass. Harvard University Press).

Gardner, R., and J. Strauss, 1981, Repressed inflation in theSoviet Union, European Economic Review 16, 387—404.

Goniulka, S., 1985, Kornai's soft budget constraint and theshortage phenomenon: a criticism and restatement, Economicsof Planning 19, 1—11.

Gourieroux, C., and G. Laroque, 1985, The aggregation ofcommodities in quantity rationing models, InternationalEconomic Review 26, 681—700.

Grandmont, J. N., and C. Laroque, 1976, On temporary Keynesianequilibria, Review of Economic Studies 43, 53—67.

Hansen, B., 1951, A study in the theory of inflation (London,Allen and Unwin).

Hare, P., 1982, Review article: Economics of shortage and non—price control, Journal of Comparative Economics 6, 406—425.

VHenin, P.—Y., 1984, Equilibres avec rationnement d'une conomie a

planification centralise et secteur parallle: une analysernacroconomique. Document no. 90, UER Analyse et PolitiqueEconomique, Universit de Paris I.

Holzman, F., 1956, Financing Soviet development, in: Capitalformation and economic growth (Princeton: PrincetonUniversity Press for the National Bureau of EconomicResearch).

Holzman, F., 1960, Soviet inflationary pressures, 1928—1957,Quarterly Journal of Economics 74, 167—188.

Holzman, F., 1968, Soviet central planning and its impact onforeign trade adjustment mechanisms, in: Brown, A. andE. Neuberger, eds., International trade and centralplanning (Berkeley: University of California Press).

Howard, D., 1976, The disequilibrium model in a controlledeconomy: an empirical test of the Barro—Grossman model,American Economic Review 66, 871—879.

Hulyak, K., 1982, Disequilibrium models for consumption goods inHungary, paper presented to International Conference ofApplied Econometrics, Budapest.

2

Hulyak, K., 1983, An experimental disequilibrium macromodel forHungary, mimeo.

Katz, B., 1979, The disequilibrium model in a controlled economy:comment, American Economic Review 69, 721—25.

Klaus, V.. and V. Rudlovcek, 1982, Savings function as an inverseproblem of disequilibrium consumption modelling: a case studyof Czechoslovakia, paper presented to InternationalConference of Applied Econometrics, Budapest.

Kooiman, P., 1984, Smoothing the aggregate fix—price model andthe use of business survey data, Economic Journal 94, 899—913.

Kornai, J., 1980, Economics of shortage (Amsterdam, North—Holland).

Kornai, J., 1982, Growth, shortage and efficiency (Oxford,Blackwell).

Malinvaud, E., 1977, The theory of unemployment reconsidered(Oxford, Blackwell).

Malinvaud, E., 1981, Econometric implications of disequilibriumtheory, INSEE Working Paper no. 8114.

Marer, P.. 1985, Les relations commerciales Est—Ouest, Revued'etudes comparatives Est—Ouest 16, no. 3, 5—16.

Martin, C., 1986, Disequilibrium models with dispersed trading,Ph.D thesis, Birkbeck College, University of London.

Muellbauer, J., 1978, Macro theory vs. macroeconometrics thetreatment of disequilibrium in macroeconomic models,Birkbeck College Discussion Paper in Economics no. 59.

Muellbauer, J., and R. Portes, 1978, Macroeconomic models withquantity rationing, Economic Journal 88, 788—821.

Neary, J. P., and J. Stiglitz, 1983, Towards a reconstruction ofKeynesian economics: expectations and constrainedequilibria, Quarterly Journal of Economics 98, 199—228.

Nove, A., 1979, Inflation in communist countries, in: Politicaleconomy and Soviet socialism (London, Allen and Unwin), 178—191.

Nissanke, M. K., 1979, The disequilibrium model in a controlledeconomy: comment, American Economic Review 69, 726—32.

Nuti, D.M., 1985, Hidden and repressed inflation in Soviet—typeeconomies: definitions, measurement and stabilization, mimeo.

Ofer, G., and J. Pickersgill, 1980, Soviet household saving: a

3

cross—section study of Soviet emigrant families QuarterlyJournal of Economics 94, 121—44.

Peebles. 0. 1981, Money incomes and expenditures of thepopulation of the Soviet Union: a East European comparisonSHong Kong Economic Papers 14, 53—78.

Pickersgill, J., 1976, Soviet household saving behaviour, Reviewof Economics and Statistics 18, 139—47.

Pickersgill, J., 1980a, Repressed inflation and price controls inthe Soviet household sector, Kennan Institute conferencepaper.

Pickersgill, J., 1980b, Recent evidence on Soviet household savingbehaviour, Review of Economics and Statistics 62, 628—33.

Pickersgill, J., 1983, Household saving in the USSR, in:Modigliani, F. and R. Hemming, eds., The determinants ofnational saving and wealth (London, Macmillan).

Podkaminer, L., 1982, Estimates of the disequilibria in Poland'sconsumer markets 1965—1978, Review of Economics andStatistics 64, 423—431.

Podkaminer, L. 1984, The political economy of permanentdisequilibrium in Poland's consumer markets, mimeo.

Portes, R., 1974, Macroeconomic equilibrium under centralplanning, Institute for International Economic Studies,University of Stockholm, Seminar Paper no. 40.

Portes, R., 1977a, The control of inflation: lessons from EastEuropean experience, Economica 44, 109—130.

Portes, R., 1977b, East Europe's debt to the West, ForeignAffairs 55, 751—782.

Portes, R., 1979, Internal and external balance in a centrallyplanned economy, Journal of Comparative Economics 3, 325—345.

Portes, R., 1980, Effects of the world economic crisis on theEast European economies, The World Economy 3, 13—52.

Portes, R., 1981, Macroeconomic equilibrium and disequilibrium incentrally planned economies, Economic Inquiry 19, 559—78.

Portes, R., 1983, Central planning and monetarism: fellowtravellers? in Desai, P. ed., Marxism, central planning andthe Soviet economy (Cambridge, Mass., MIT Press), 149—65.

Portes, R., and D. Winter, 1977, The supply of consumption goodsin centrally planned economies Journal of ComparativeEconomics 1, 351—365.

4

Portes, R., and D. Winter, 1978, The demand for money and forconsumption goods in centrally planned economies, Review ofEconomics and Statistics 60, 8—18.

Fortes, R. and P. Winter, 1980, Disequilibrium estimates forconsumption goods markets in centrally planned economies,Review of Economic Studies 47, 137—159.

Fortes, R., R. Quandt, D. Winter, S. Yeo, 1983, Macroeconomicplanning and disequilibrium: estimates for Poland, 1955—1980, NBER Working Paper no. 1182.

Fortes, R., R. Quandt, D. Winter, S. Yeo, 1984a, Planning theconsumption goods market: preliminary disequilibriumestimates for Poland 1955—80, in: Nalgrange, P., and P.—A,Muet, eds., Contemporary macroeconomic modelling (Oxford,Blackwell), 254—271.

Fortes, R., R. Quandt, D. Winter, S. Yeo, 1984b, Estimating thesize of plan errors, Princeton University Financial ResearchCenter, Research memo no. 48, appeared (in French) inAnnales de l'INSEE 55/56, 245—255.

Portes, R., R. Quandt, and S. Yeo, 1985, Testing the "all—excess—demand" hypothesis, mimeo.

Pryor, F.) 1977, Some costs and benefits of markets: an empiricalstudy, Quarterly Journal of Economics 91, 81—102.

Quandt, R., 1978, Tests of the equilibrium vs. disequilibriumhypotheses, International Economic Review 19, 435—452.

Quandt, R., 1982, Econometric disequilibrium models, EconometricReviews 1, 1—63.

Skurski, R., 1972, The buyers' market and Soviet consumer goodsdistribution, Slavic Review 31, 817—830.

Stupnicki, K., 1985, The disequilibrium theory: R. Fortes v. J.Kornai, a critical comparison, Working Paper 2, Inst itutCybernetyki i Zarzadania, Szkola Glowna PlanowaniaStatystyki, Warsaw.

Tartarin, R., 1982, La thorie du ds6quilibre applique auxgconomies du type socialiste, mimeo.

Timmer, J., 1982, Disequilibrium and buffer stocks, ErasmusUniversity.

Turcan, J., 1977, Some observations on retail distribution inPoland, Soviet Studies 29, 128—136.

Weinrich, 0., 1984, On the size of disequilibrium in anequilibrium with quantity rationing, CORE Discussion Paper

5

No. 8418.

Wiles, P., 1982, Inflation in USSR 1982, paper presented toNASEES Conference.

Wiles, P., and J. Rostowski, 1979, Soviet type inflation: an opendebate, Est—Ouest 10:2/3, 59—75.

Winiecki, J., 1985, Portes ante portas. a critique of therevisionist interpretation of inflation under centralplanning, Comparative Economic Studies 27, 25—52.

Wolf, T., 1985, Economic stabilization in planned economies,International Monetary Fund Staff Papers, 32, 78—131.

6