Digitized by the Internet Archive
in
2011 with funding from
Boston Library Consortium
Member
Libraries
http://www.archive.org/details/macroeconomiceffOOblan
DEWEY
Massachusetts Institute of Technology
Department of Economics
Working Paper Series
Macroeconomic Effects of Regulation and
Deregulation in Goods and Labor Markets
Olivier Blanchard,
Francesco Giavazzi,
MIT
Universit'
a Bocconi
Working Paper 01 -02
January 2001
RoomE52-251
50 Memorial Drive
Cambridge, MA 02142
This paper can be downloaded without charge from the
Social Science Research Network
Paper Collection
http://papers.ssm.com/paper.taf7abstract id=257542
at
C^iP^
Massachusetts Institute of Technology
Department of Economics
Working Paper Series
Macroeconomic Effects of Regulation and
Deregulation in Goods and Labor Markets
Olivier Blanchard,
Francesco Giavazzi,
MIT
Universit' a
Bocconi
Working Paper 01 -02
January 2001
RoomE52-251
50 Memorial Drive
Cambridge, MA 02142
This paper can be downloaded without charge from the
Social Science Research Network
Paper Collection
http://papers.ssm.com/paper.taf7abstract id=257542
at
MASSACHDsETTSINSTimn
_OFTECHI\IOLOGY
Macroeconomic
effects of regulation
and
deregulation in goods and labor markets.
Olivier Blanchard
and Francesco Giavazzi
January
*MIT and NBER, and Bocconi
David Spector
for discussions,
22,
University,
*
2000
CEPR
and Ricardo Caballero,
and NBER,
Bill
respectively.
We
thank
Dickens, Francesco Franco, Paul
Joskow, Augustin Landier, Robert Solow, Justin Wolfers, Torsten Persson, and Joseph
Zeira for comments.
Macroeconomic
effects of regulation
Abstract
Product and labor market deregulation are fundamentally about
reducing and redistributing rents, leading economic players to adjust
in
turn to this
ally
new
distribution.
proves beneficial,
effects.
The
transition
it
and so
if
deregulation eventu-
comes with strong distribution and dynaniic
may imply
employment may increase
recovering,
Thus, even
the dechne of incumbent firms. Un-
for
a while. Real wages
we
build a
may
decrease before
on.
To study these
issues,
model based on two central
as-
sumptions: Monopolistic competition in the goods market, which determines the size of rents,; and bargaining in the labor market, which
determines the distribution of rents between workers and firms.
We
then think of product market regulation as determining both the entry
costs faced
by firms, and the degree of competition between
firms.
We
think of labor market regulation as determining the bargaining power
of workers.
Having characterized the
ulation,
we then use our
effects of labor
results to
and product market dereg-
study two specific
to shed Ught on macroecononomic evolutions in
issues.
First,
Europe over the
last
twenty years, in particular on the behavior of the labor share. Second,
to look at political
economy
market deregulation.
interactions
between product and labor
Macroeconomic
effects of regulation
Product and labor market regulations are often blamed
European performance of the
ulations, the
argument
last
goes,
and Europe
will soar.
~
decrease, output will increase.^
Deregulation however
Remove (many
30 years.
if
will
'
fundamentally about reducing and redistribut-
is
deregulation eventually proves beneficial,
with both strong distribution and dynamic
The
effects.
it
new
is
increase for a while. Real wages
may
distribution.
come
likely to
may
transition
im-
Unemployment
ply the disappearance or the decline of incumbent firms.
may
of) these reg-
Unemployment
ing rents, leading economic players to adjust in turn to this
Thus, even
the poor
for
decrease before recovering, and
so on.
Understanding these dynamic and distribution
effects is important, for
at least
two reasons. As many countries are embarking on a path of dereg-
ulation,
it
may
also help clarify the political
improve
design.
its
We start
of an
is
These are the
by developing,
it
may
economy constraints on deregulation, and thus
issues
in Section 1,
we examine
in this paper.
a simple general equilibrium model
economy with both product and labor market
regulation.
on two basic assumptions: Monopolistic competition
built
And
help interpret their macroeconomic evolutions.
The model
in the
goods
market, which determines the size of rents; and bargaining in the labor
market, which determines the distribution of rents between workers and
firms.
We
think of product market regulation as determining both the entry
costs faced
of labor
We
by
firms,
and the degree of competition between
We
think
market regulation as determining the bargaining power of workers.
then characterize the macroeconomic equilibrium
divide time in
two periods, a short run, where the number
and a long run, where the number of firms
'This
firms.
is for
example the theme
of a
is
in Section 2.
of firms
given,
endogenous, determined by
number of studies by the McKinsey Global
See for example McKinsey Global Institute [1997].
is
We
Institute.
Macroeconomic
effects of regulation
an entry condition. For each period, we show how the main macroeconomic
variables, in particular the real
wage and unemployment, depend on the
various dimensions of regulation.
We then
is
turn in Section 3 to the economics of deregulation. (The model
symmetric so the economics of
reversed, as those of
more
less
regulation.
regulation are the same, with sign
Focusing on deregulation
is
more
natrual in the current context). This section offers in effect a reinterpretation
of the results of Section 2, but with a focus
effects of
on distribution and dynamic
each type of deregulation.
Section 4 discusses two extensions. In our benchmark model, which as-
sumes privately
run.
the wage
efficient bargaining,
In other words, the wage
enue product of labor. Our
is
first
is
not allocative in the short
typically not equal to the marginal rev-
extension considers a standard alternative
model, known as the right to manage model, in which the wage
the marginal revenue product of labor.
The purpose
is
is
equal to
to show, a contrario,
the implications of the efficient bargaining assumption, in particular for the
dynamic
effects of labor
market deregulation. Our benchmark model also
Our second extension
assumes Unear
utility.
sumes concave
utility instead.
The reason
is
relaxes this assumption
and
as-
that, in that case, the intertem-
poral trade-offs implied by labor market deregulation are even sharper than
in our
benchmark model, higher unemployment and lower
short run, in exchange for lower
real
wages
unemployment and unchanged
real
in the
wages
in the long run.
Having developed the theory, we then turn,
conomic evidence.
We
show the
light
in Section 5, to the
our approach sheds on movements in
the labor share and in the "markup" two variables at the center of
,
cent macroeconomic research (for example
GaH and
Gertler [1999]).
We
macroe-
much
Rotemberg and Woodford
re-
[1999],
then focus on the dramatic decline in the
la-
bor share in continental Europe since the mid-1980s, in some countries by
more than 10%
of
GDP. Standard
neoclassical
mechanisms seem unable to
Macroeconomic
effects of regulation
account for this evolution, and
it
was our guess that an explanation might
be found in deregulation and the implied changes
in the
economy. Our model indeed
on deregulation
workers to firms.
The long run
in the labor
If
in equilibrium
a tentative interpretation, based
offers
market and the implied transfer of rents from
our interpretation
effects
is
correct, this evolution
is
good news:
should be a recovery of the labor share, and a decrease
unemployment.
Finally, Section 6 turns to the political
on our
in the distribution of rents
analysis,
we look
at
lation affect the welfare of
economy
of deregulation.
how product market and
labor market deregu-
employed and unemployed workers.
how governments may combine
Based
We
look at
the two so as to reduce workers' opposition
to deregulation. Based on a simple pohtical
economy model, we endogenize
the various dimensions of regulation, and, within this context, look at the
the interactions between product and labor market deregulation.
how product market
deregulation
Intuitively, reducing rents in the
may
trigger labor
show
market deregulation.
goods markets reduces the incentives of
workers to fight for a share of these rents: The benefits
worth the
We
may no
longer be
costs.
1
Monopolistic competition, bargaining, and regulation
We
think of an economy in which a
products, using labor.
We make
number
of firms produce differentiated
two main assumptions. The
first is
that of
monopolistic competition in the goods market, which determines the size of
the rents going to firms and their workers.
The second
bargaining in the labor market, which determines
to firms,
and how much to
is
the presence of
how much
of the rents go
their workers.
We divide time in two periods: The
"short run"
,
defined as the time over
which we can take the number of firms as given. The "long run" defined as
,
the time over which the
number
of firms
is
endogenous, determined by an
Macroeconomic
effects of regulation
entry condition.
We
take product market regulation as determining the degree of com-
petition
among
and the entry cost
firms,
for firms.
We
take labor market
regulation as determining the degree of bargaining power of workers.
The
specific
assumptions are as follows:
Workers
There are L workers/consumers, indexed by
has a
utility function
j.
In each period, worker j
given by:
m
VJ
=
o-/(ct-1)
[m-'/'^EC^/^~'^^''^]
(1-1)
i=l
where a
=a
products (which
g(m),
is
g'{.)
>
0,
a
a constant, and
is
m
is
the
number
of
given in the short run, and endogenously determined in
the long run).
This specification of utihty has two imphcations:
•
Under the assumption that the worker consumes
all
products in equal
proportions (a condition which, in our symmetric model, holds in equilibrium), so Czj
= Cj/m,
words, an increase in the
the utility function implies Vj
number
directly. (Technically, this result
—
Cj. In other
of products does not increase utility
comes from the presence of m~^''^
in
number
of
the term in brackets. Absent this term, an increase in the
products would increase
•
utility for a given level of
The increase in the number
consumption.)
of products however increases the elasticity
of substitution between products,
and by implication the
demand
comes from the assumption that
facing firms. (This result
elasticity of
a,
rather than being constant as in the standard Dixit-Stiglitz framework,
is
increasing in
tti.)
Macroeconomic
effects of regulation
Thus, to the extent that deregulation leads to a larger number of firms,
and by implication a
number
larger
ferent product), its effect in our
monopoly power
we want
of firms. This
model works only through
is
the effect
we think
its effect
on the
most important, and
is
to capture here.
Each period, worker
spends
of products (each firm produces a dif-
can supply either zero or one unit of labor, and
i
income on consumption (there
his
is
no saving, or capital
in
our
model, and thus no link across the two periods). His budget constraint for
each period, stated in nominal terms,
is
given by:
m
Y,P,Cii
= WjN,+Pf{u){l-N,)
i=l
where Nj^ labor supply,
to one
(if
he works),
/'(.)
is
<
equal either to zero
0,
and
P
is
(if
he does not work) or
the price index associated with
consumption:
2=1
Spending on consumption
and to non- labor income
employed
is
if
is
equal to labor income
if
the worker works,
he does not. The wage equivalent of being un-
taken to be a decreasing function of unemployment, f{u). This
can be thought of as a shortcut to capture the notion that higher unemploy-
ment implies a higher individual probability
of remaining
shortcut to a fuller dynamic specification).
Or more
hypocritically,
it
unemployed
cleanly,
(i.e.
a
but somewhat
can be interpreted as the marginal product from using an
alternative decreasing returns-to-scale
common
Note that under symmetry of consumption
"backyard technology"
(so Cij
=
Cj/m), and using
the budget constraint, the utility of worker j in each period can be rewritten
as:
Macroeconomic
effects of regulation
This expression
Products
will
be useful below.
find firms
Each product
is
produced by one firm, so
The production
the firm.
function of firm
Yi
There
number
which
no
is
of products,
is
And
capital.
there
=
is
i
is
i
indexes both the product and
simply:
Ni
no
effect, direct or indirect, of
and thus of competition, on the productivity
the
of labor
identically equal to one.
Each firm
is
run by an entrepreneur, with
utility also given
by
(1.1).
each period, the entrepreneur keeps the profit of the firm, and spends
consumption goods. Nominal
profit in firm
i
given by PiYi
is
—
it
In
on
WiNi, or
equivalently:
(Pi
- W^)Ni
Bargaining
Each period, each firm bargains with
either
work
We
in the firm or
L/m
workers.
The workers can
be unemployed during the period.
assume Nash bargaining: Together firm
i
and the workers choose
a wage and a level of employment so as to maximize the (log) geometric
average of their surpluses from employment:
/31og((VFi
where the
first
- Pf{u))
term
reflects
TV,)
+
(1
- /?) log((P, -
Wi)N,)
(1.2)
the surplus to workers from working in firm
Macroeconomic
i
effects of regulation
(under the assumption of symmetric consumption) the second reflects the
,
profit of firm
i,
and
(3
This assumption
assume
known
is
bargaining?
efficient
But
this context.
reflects the relative bargaining
not be operating on their
want to allow
higher
employment,
seenS
First, it
like
for labor.
when
In
Why
a natural assumption in
to capture the possibility that firms
demand
for the fact that,
may be
/?)
as (privately) "efficient bargaining".
we want
also,
power of workers.
may
more informal terms, we
there are rents, stronger workers (a
able to obtain a higher
wage without
suffering a decrease in
at least in the short run. Efficient bargaining naturally delivers
that implication. But any assumption which relaxed the link between the
wage and the marginal revenue product
similar results.
Section
We
of labor could yield qualitatively
return to a discussion of alternative assumptions in
4.
The
short and the long run
In the short run,
in the long run,
we take the number
we assume the number
of firms/products as given.
But,
of firms/products to be determined
endogenously, by an entry condition. Our purpose in doing so
is
to capture
the effect of the short run distribution of rents between firms and workers
on the equilibrium number of firms
We
in the long run.
assume that firms face a cost of entry equal to
of as coming from product market regulation.
about
•
which we think
c,
We make
two assumptions
c:
We
assume that c
regulation,
is
and the
a shadow
fact that
cost.
many
The motivation
is
our focus on
regulatory barriers to entry take
the form of legal and administrative restrictions on entry, rather than
direct costs.
no implication
Except
for
accounting purposes, this assumption has
for the characterization of the equilibrium.
that, in our long
run equilibrium, existing firms make pure
It
implies
profits; if
Macroeconomic
c were
effects of regulatiori
an actual
1
would be dissipated
cost, these profits
in entry costs.
In looking at the evolution of the profit share over time,
reasonable to think that, in
make
•
many
is
that c
two are equal here)
a fixed cost
is
.
is
perhaps not forever.
having a proportional rather than
for
algebraic simplicity:
profit rate (profit
if
proportional to output (or employment, as the
The reason
easier to characterize.
seems
markets, regulation allows firms to
positive pure profits for a long time,
The second
it
It trivially
It
makes the long run equilibrium
implies that, in the long run, the
per unit of output) must be equal to
livers the result that, in the limit,
c,
and de-
the equilibrium converges to the
competitive equilibrium as c goes to zero.
It
obviously eliminates the
standard issues examined by models of monopolistic competition, such
as optimality of the
number
of products
and so
on.
But they are not
the focus here, and either allowing for a non-regulatory fixed cost or
allowing the regulatory cost itself to be a fixed cost would not
make
any substantial difference to the results we want to focus on here.
Regulation
We think of regulation as being captured,
by three parameters
•
We
in the
admittedly in abstract fashion,
model:
think of a and c as reflecting two dimensions of product market
regulation.
In the context of European integration for example, de-
creases in
may
(T
reflect the elimination of tariff barriers, or standard-
ization measures
making
European Union
countries.
it
easier to sell
domestic products
Decreases in c
may come,
for
in other
example,
from the ehmination of state monopolies, or the reduction of red tape
associated with the creation of
new
firms.
Macroeconomic
11
effects of regulation
We think of p as reflecting any aspect
•
increases the bargaining
of labor
market regulation which
power of workers, ranging,
for
example, from
the existence and the nature of extension agreements, to closed shop
arrangements, to the rules on the right to
Our
goal
is
to
show how these
strike.
three parameters determine the size
and
the distribution of rents, and by implication, the macroeconomic equilibrium.
,
Short and long— run equilibrium
2
The
easiest
way
to characterize the equilibrium
is
to do so in three steps,
starting with the short run partial equiUbrium, then turning to the short
run general equilibrium, and
The short run
finally to the
long run general equilibrium.
partial equilibrium
Consider the problem faced by firm
i,
producing good
preferences of workers and entrepreneurs, the
and entrepreneurs)
is
At a
Y
is
total
demand
=
good
i
(by workers
Y P "'^
-(f)
(total output),
relative price of one, the firm faces a
demand. The
for
given by:
y,
where
demand
Given the
i.
elasticity of
demand with
(2.1)
and
y, the
demand
demand
for
good
i.
equal to one-7nth of total
respect to the relative price
is
equal
to {-a).
Taking Y, P, and the unemployment rate u as given, firm
workers associated with firm
wage Wi so
i
choose employment Ni, the price
as to maximize:
PlogiiWi
- Pf{u))
N,)
+ {1-P) log((P, -
W,)Ni)
i
Pi,
and the
and the
Macroeconomic
12
effects of regulation
where, from the production function, Ni
by
The
Yi,
and demand
Yi
relative price Pi chosen
by the firm (and the workers)
is
is
lJ,{m),
the
markup
(2.2)
of the relative price over the reservation wage,
given by:
u(m)
—
—
^—,
cTg[m)
The
•
given
given by:
^ = {l+^cim))f{u)
where
is
foUows that:
(2.1). 2 It
•
=
real
-
so /x'(m)
consumption wage (the wage
basket), VKj/P,
is
<
1
in
terms of the consumption
given by:
^=
(l-/3)/(tz)
+ /3j
So, using equation (2.2):
^=
[l
+ Mm)]/(u)
(2.3)
A graphical representation of the partial equilibrium is given in Figure
Employment
relative price,
(equivalently, output)
is
measured on the horizontal
1.
axis, the
Pj/P, and the real consumption wage, Wj/P, on the vertical
axis.
^Note the implicit assumption that, when choosing the price
iirm
i
P;,
both workers and the
ignore the effect on the change in Pi on the price of a consumption basket,
Tliis-standard-approximation
is
clearly better the larger the
number
of goods.
P
3
o
ID
w
Macroeconomic
13
effects of regulation
The demand curve and marginal revenue product
curves are drawn as
DD and MRP. The reservation wage is drawn as the horizontal line,
From
at f{u).
the point of view of workers and the firm, the efficient level of
employment
is
such that the marginal revenue product of labor
is
equal to
the reservation wage, so at point A, with associated level of employment Ni.
This in turn implies the choice of a relative price, Pi/P-, on the demand curve,
so a price equal to one plus a
fashion, the
markup
is
Given the relative
f{u)), or ^if{u).
markup
ji,
times the reservation wage. In usual
related to the elasticity of
price, rents per unit of
The workers
{1
i.i
=
l/(cr
—
1).
output are given by {Pi/P
get a proportion
wage, which plays no allocative role under
demand by
f3
—
of those rents, so the real
efficient bargaining, is
equal to
+ I3^f{u)).
Note
of both
•
that, in partial equilibrium, the real
j3
The
and
wage
is
an increasing function
/z:
higher
/?,
the higher the proportion of rents going to workers.
because the reservation wage
is
unaffected, the increase in the
And
wage
has no effect on employment.
•
The
higher
rents, of
/x,
thus the higher the real wage.
The
firm receives larger
which some proportion goes to the workers.
General equilibrium. Short run.
In partial equilibrium, each firm chooses
But, in general equilibrium, not
than one.
all
its relative price
equal in general equilibrium. Putting
=
(l
freely.
firms can have a relative price greater
Indeed, under our symmetric assumptions,
l
Pi/P
Pi/P
=
1 in
+ M(m))/(u)
all
prices
must be
equation (2.2) implies:
(2.4)
Macroeconomic
effects of regulation
In the short run, the
and by implication so
is
number
fi{m).
14
of firms
is
given, so
Given ii{m), equation
a
=
d-g{m)
(2.4)
is
given,
determines the
equilibrium unemployment rate.
Replacing f{u) by 1/(1
+ iJL{m))
in equation (2.3), the real
wage
is
given
in turn by:
W^ ^ (l+M(m)/3)
p
The
equilibrium
ing Figure
1.
is
is
is still
2.
at the point
Figure 2 starts by replicat-
where the marginal revenue
equal to the reservation wage, at point A. But now, the
implied relative price must be equal to
markup
^-^
+ ^Ji{m))
characterized in Figure
Equilibrium
product of labor
{l
1.
Given that the
over the reservation wage, and given that the
relative price
markup
is
is
a
fixed in the
short run, this condition determines the reservation wage, and in turn the
equilibrium level of unemployment.
The
real
wage
is still
set as a
weighted
average of the reservation wage and the relative price.
Return to the
effects of
P and ^ on
the real wage,
now
in the short
run
general equilibrium:
•
As was the
case in partial equilibrium, the real
ing function of
/?.
An
/3
increase in
is
is still
an increas-
increases the proportion of rents going to workers,
and thus leads to a higher
the real wage
wage
real wage.
And, because,
in the short run,
not allocative, this higher real wage has no effect on
employment, or unemployment.
• In contrast
now
This
however to the partial equilibrium
a decreasing function oi
is
because there are
partial equilibrium effect
case, the real
wage
is
fi.
now two
we saw
effects at
earlier:
A
work.
higher
The
/i
first is
the
means higher
p
-C
o
o
(U
II
JD
^„—
3
O
c
on
3 s
s +"
H
G S"
S
a
1
tiO s_-'
o S
<D
zL
J2
c
a
f-*-»
I
II
J2
cr
o
(N
zL
+
zL
+
Macroeconomic
15
effects of regulation
rents to the firm where the worker works, leading to a higher real
wage.
The second
to firms
is
the general equilibrium
effect.
come from consumers, who now pay more
The
for
rents going
the goods they
buy. So workers gain as workers, bufTose as consumers. Because, as
workers, they only get a proportion
dominates the
first.
The
real
j3
of the rents, the second effect
wage goes down.
General equilibrium. Long run
In the short run, the real wage
left
is
to firms has no effect on their
not allocative, and the size of the rents
employment
In the long run
decision.
however, rents determine entry or exit of firms. In the long run, rents must
cover entry costs. ^ Given our assumption that entry costs are proportional
to output, this condition takes the simple form:
7r^^%^+ mM)
(2.6)
(1
Profit per worker
must be equal to the shadow
cost
c.
This equation
determines the equilibrium number of products m. Recall that the number
and
of products determines the elasticity of substitution between products,
thus the elasticity of
must be such
demand
facing firms.
Thus, the number of products
as to generate a degree of competition consistent with profits
equal to entry costs.
Using the definition of
yu(r?i),
equation (2.6) can be rewritten
as:
ag{m)^^^—^
^Our two-period model cannot capture the
sumably,
if
specific
(2.7)
dynamics of entry and
exit.
Pre-
rents are less than entry costs, firms which die will not be replaced until rents
have recovered
sufficiently to justify entry. If rents are larger
enter until rents have been bid
down
to entry costs.
than entry costs, firms
will
Macroeconomic
Given that
effects of regulation
g'{.)
ing function of a:
rents,
/?:
less attractive.
A
the equihbrium
0,
More competition
making entry
function of
>
1
number
number
for a given
The number
less attractive.
of products
of firms
of firms decreases
is
also a decreasing
smaller proportion of rents going to firms also
And
the
number
of firms
is
makes entry
a decreasing function of
entry costs require higher rents, leading to a smaller
Replacing the markup from
a a decreas-
is
number
Higher
c:
of firms.
the unemployment rate
(2.6) in (2.4),
is
given by:
The
higher c or the higher
(3,
the higher the
markup required
to cover
entry costs, thus the smaller the equilibrium reservation wage, and, in turn,
the higher the unemployment rate.
Finally, replacing the
markup from
(2.6) in (2.5),
the real wage
given
is
by:
^ = l-c
The
must
intuition
is
straightforward:
—
is
role of
Because the supply of firms
in
P no
is
P and ^ on
The
real
wage must
higher unemployment. Higher
given entry costs, a lower
the real wage:
fully elastic in the long run,
longer increases the real wage.
/3
The
effect
means lower rents
number
of firms, a higher
reservation wage,
and so a higher rate
The markup,
is
ii,
Firms
equal to one.
c.
Return once again to the
•
Productivity
receive c per unit in order to cover entry costs.
therefore be equal to 1
•
(2.9)
of
an increase
now shows up
for firms,
and
/3
for
markup, a lower
unemployment.
no longer an exogenous parameter, but
determined in equilibrium by both
in
and
c,
so
we must look
is
now
at the
Macroeconomic
effects of c instead.
value of
^i,
An
increase in
c, wfiicfi
increases
tfie
leads to a decrease in the real wage. But, now,
to an increase in the
number
17
effects of regulation
unemployment
rate.
A higher
equilibrium
it
leads also
c leads to a smaller
of firms, a higher markupT^aTlower required reservation wage,
and a lower unemployment
rate.
Having characterized the equilibrium, we can now turn to the
effects
of various dimensions of deregulation. While the results have already been
implicitly given,
something
is
gained from the discussion.
We
do so
in the
next section.
Deregulation
3
Let us start with the two dimensions of product market deregulation, then
turn to labor market deregulation.
Product market deregulation.
Suppose the government increases
uct market, for a given
number
An
increase in a
a, increasing competition in the prod-
of firms.
In the short run, firms facing more elastic
demand
decrease their markup,
leading in turn to both an increase in real wages, and a decrease in unem-
ployment.
The
favorable effects however vanish in the long run.
an unchanged entry
number
the
cost, the decrease in the profit rate leads
of firms over time (Because
which die are not replaced)
its initial,
rate
.
and the
so
it is
is,
given
a decrease in
not profitable to enter, firms
In the long run, the profit rate must go back to
pre-deregulation, level.
initial level,
The reason
But
for the profit rate to return
back to
its
must the markup. By implication, so do the unemployment
real wage.
In short, this dimension of product market deregulation
is
eventually
Macroeconomic
The
self-defeating:
economy returns
These
as given,
18
effects of regulation
favorable short-run effects disappear over time and the
to its pre-deregulation equilibrium.
results are probably too strong, in that
when
measures are
looking at changes in a.
likely to affect c as well.
we take
In practice,
If for
c,
the entry cost,
many
deregulation
example, we think of c as
the shadow cost of a quantitative restriction on the number of firms
(for
example, the granting of a market to a monopoly firm), then these firms
will stay in the
will
market even
if
a
More
increases.
formally, the
shadow
cost c
go down one-for-one with the profit rate, leading to the same favorable
effects of deregulation in
results
the short-nm and the long-run. Netherveless, the
make an important
ultimately
point:
come from entry
To the
costs, then,
extent that rents in the
nothing
if
is
done to decrease entry
attempts to increase competition by other means are
costs,
economy
likely to
be partly
self-defeating.
A
Product market deregulation.
The
decrease in
c
previous argument suggests that the second dimension of product
market deregulation, a decrease in entry
even in the long run.
And
indeed,
costs,
is
more
likely to
be favorable,
it is.
Obviously, from our assumption that the
number
of firms
is
fixed in the
short run, the decrease in entry costs has no effect in the short run.
the long run,
it
an aspect which
happens to the size of incumbent
firms,
we turn
of deregulation,
total
to the pohtical
economy
of firms increases, but, as the
employment increases as
well.
number
of firms,
employment
is
in
is
will
(What
be relevant when
theoretically ambiguous:
unemployment rate decreases,
To the extent
the relative increase in total employment
in the
in
leads to entry of firms, thus to a higher elasticity of demand, a
lower markup, and thus lower unemployment and a higher real wage.
The number
But
that, as
seems plausible,
smaller than the relative increase
incumbent firms decreases.)
Macroeconomic
19
effects of regulation
In short, this dimension of product market deregulation works because
it
attacks the problem at the root, decreasing the rents the firms require to
enter
and stay
in the
market. This allows for more competition, and in turn
lower unemployment and higher real wagesT"
Note that,
for neither
dimension of product market deregulation,
an intertemporal trade-off
for real
sion leads to higher real wages
no long-run
effect,
wages or unemployment. The
and lower unemployment
the second to no short-run
and lower unemployment
in the long run.
we now
Consider a decrease
in
/?,
A
dimen-
and higher
real
and
wages
different in the
turn.
decrease in
j3
a decrease in the bargaining power of workers.
In the short run, workers give
up
rents;
Prom
decreases; equivalently, the profit rate increases.
factor
first
there
in the short-run
Things are quite
case of labor market deregulation, to which
Labor market deregulation.
effect,
is
income distribution has no
effect
(2.5),
But
their real
this
change
wage
in the
on unemployment, which remains
given by (2.4). Thus, workers clearly lose in the short run.
In the long run however, the larger rents
profit rates
equal to
have returned to their
c.
As
initial level, until
the profit rate
unemployment
rate,
The
real
wage
short run decrease in real
is
back to
is
again
markup
decreases,
and an increase
in the real
Indeed, in the long run, the unemployment rate
deregulation.
The
to firms lead to entry, until
firms enter, competition increases, the
leading to a decrease in the
wage.
left
its initial,
is
lower than pre-
pre-deregulation, level:
wage due to the decrease
in
(5 is
exactly offset
by the decrease in the markup.
In short, labor market deregulation works by changing the distribution
of rents in favor of firms, leading to
more competition
in the long run,
lower unemployment. In contrast to product market deregulation,
it
and
comes
with a sharp intertemporal trade-off, lower real wages in the short run in
Macroeconomic
exchange
we
for lower
unemployment
discuss the political
We
have shown
distribution
20
effects of regulation
economy
how
This will be relevant when
of labor market deregulation in Section
different dimensions of deregulation
and dynamic
these results in two ways,
in Europe, then to think
in the long run.
first
market deregulation. Before doing
political
so,
we
want
to use
macroeconomic evolutions
to look at recent
about the
have different
We now
on the economy.
effects
6.
economy
of labor
and product
discuss two extensions of the basic
model.
4
Extensions
Our model
is
based on a number of strong simplifying assumptions, from
linear utility, to the
form of bargaining.
of labor market deregulation later on,
first,
we modify the nature
imphcation that the wage
main purpose
is
Given our focus on the
we explore two
extensions.
effects
In the
of bargaining in the labor market, with the
is
now
allocative,
even in the short run.
The
to show, a contrario, the imphcations of efficient bargaining
for the effects of labor
market deregulation. In the second, we consider the
imphcations of concave
utility.
by workers
in
In this case, the intertemporal trade-off faced
in the event of labor
market deregulation
is
even starker than
our benchmark model: lower real wages and higher unemployment in the
short run, in exchange for lower
unemployment
in the long run.
Ex-post determination of employment
Under our assumption that bargaining
is
privately efficient, the
plays only a distributive role in the short run.
wage
Under that assumption,
labor market deregulation, in the form of a decrease in
/3,
gives rise to a
sharp intertemporal trade-off, lower wages in the short run, in exchange for
lower unemployment only in the long run.
Macroeconomic
Under the
employment
alternative assumption that
by firms so as to maximize
"right to
21
effects of regulation
is
bargained wage
profit given the
manage" assumption, things look quite
unemployment and unchanged
run and
The reason
in the long run.
is
A
now
/3
There
is
is
no
decrease in
both
real wages,
that the real wage
even in the short run, and a decrease in
—the so called
different.
intertemporal trade-off from labor marEeT^eregulation:
leads to lower
chosen ex-post
(5
in the short
now
allocative
functions as an exogenous
decrease in the reservation wage in the benchmark model.
Consider the partial equilibrium
ex-post, then workers
first.^
If
and the firm maximize
firms can choose
(1.2)
employment
by choosing a wage equal
to:
^ = (1+/?m)/M
In partial equilibrium, and for a given
(4.1)
unemployment
rate, the
wage turns
out to be the same as under efficient bargaining. But the relative price
and given
different,
by:
J=
The
price
is
now
is
a
(l
+ M)^ =
markup
(l+/x)[l
+ /?M]/(ti)
(4.2)
over the real wage, not the reservation wage.
Because firms take the bargained wage as given when choosing employment,
the wage
is
now
allocative.
Equation
ization" present in the economy.
The wage
reservation wage, and the price
equal to
is
shows the "double marginal-
(4.2)
is
(1
equal to (1
+ ji)
+ (3n)
times the real wage.
Turn to the short-run general equilibrium. The unemployment
be such that the
relative price in (4.2)
l
*What
follows
is
=
{l
+ ^i)
[I
is
equal to one,
+ (5^]
times the
rate
must
so:
f{u)
well travelled ground. See in particular Layard
(4.3)
and Nickell
[1990].
Macroeconomic
effects of regulation
This expression
from that
differs
in the
benchmark model because
of the
presence of the term in brackets. This term
is
greater than one, so, for a
equilibrium unemployment
is
higher than under efficient
given value of
/i,
bargaining.
Because the price
is
now a markup
reservation wage, the real
wage
is
over the
lower than under efficient bargaining (and
depends only on monopoly power
in the
Wi
1
implication, profit per unit
Turn
goods market):
1
P
By
wage rather than over the
is
+M
larger than
to the long run equilibrium.
under
efficient bargaining.
In the long run, the zero net profit
condition gives:
u(m)
1
So,
—
+ fi{m)
.
c,
,
_
,
/
X
cFg{mj
or equivalently:
=
,
,
\/c
an economy where firms have the right to manage has a larger number
of firms,
and thus a higher
We saw
that, given the
the number of firms
is
elasticity of
number
it
in (4.3)
to the expression for
(2.8), it follows that, if /3 is positive,
in the long
of firms,
larger, leading to
unemployment. Replacing /x(m)
Comparing
demand, and thus a lower markup.
run than under
unemployment was
higher.
But
a lower markup, and thus lower
by
its
value from above gives:
unemployment
in the
benchmark model,
the unemployment rate
is
actually lower
efficient bargaining.
Finally, the zero profit condition implies that the real
wage
is
the same
Macroeconomic
as
under
23
effects of regulation
efficient bargaining,
namely:^
Wi/P =
Now
in
/3.
1
c
consider the effects of labor market deregulation,
i.e.
of a decrease
In the short run, wages do not change, and unemployment decreases.
As the
profit rate also does not change, the long
run. Thus, under right to manage, there
this
-
is
run
is
just like the short
no intertemporal trade-off from
dimension of labor market deregulation: Workers gain
in
both the short
and the long run.
This extension makes clear the role of the assumption that wages are
distributive in our
case:
in
Wages
benchmark model.
are not fully allocative,
wages may translate
an increase
in
Concave
in
We
and
beheve that
this
is
indeed the
in the short run, a large decrease
an increase in rents going to firms rather than
in
employment.^
utility
Assuming concave rather than
linear utility for workers yields
an
inter-
esting difference in the short run effects of labor market deregulation,
a decrease in
(3.
i.e.
In the short run, deregulation leads to both lower real
wages and higher unemployment (whereas, before, unemployment remained
^Note the distinct second-best
flavor of these results.
A
shift to privately inefficient
bargaining leads to unchanged real wages, and lower unemployment in the long run.
® Allowing for
capital in the production function, together with a low short run elasticity
of substitution between capital and labor and costs of adjustment to capital, also delivers
small employment effects in the short run.
(They cannot however
a decrease in
/3
deliver the decrease in
These factors are surely important as
both wages and employment
that obtains in our next extension.)
in
well.
response to
Macroeconomic
effects of regulation
unchanged). This
we
curve, and,
happened
in
is
movements along a labor demand
in sheirp contrast to
shall argue later,
Europe
24
may be
relevant to understanding
in the 1990s.
Let us derive this result formally.^ Suppose that the
is
what
utility of
workers
given by a power transformation of our previous linear utility function:
1-77
where Vj was defined
earlier as
a
CES
vidual products, and 7
<
are risk neutral. This
not essential but
is
function of consumptions of indi-
(We keep the assumption that entrepreneurs
1.
convenient).
is
In partial equibrium, and under efficient bargaining, the real wage and
the relative price of firm
i
are
now
given by:
and
+
P 1+Pn
Wi
Pi
P
For 7
=
0,
wages
sloping.
3.
The
/?
=
0,
/3, is
the equilibrium
no longer
is
the
vertical,
same
equal to the reservation wage, and employment
revenue product
is
>
the solution
0,
is
contract curve, the set of employment and
for different values of
For
^i
the results are the same as before. For 7
characterized in Figure
real
l
is
but
is
as before:
now upward
The wage
is
such that the marginal
equal to the reservation wage. As
/?
increases however,
the contract curve slopes up, with infinite slope at point A, and decreasing
slope thereafter.
Thus,
in partial equilibrium, a decrease in
^Our derivation
is
/3
implies a
movement down
a straightforward extension of McDonald and Solow
[1983].
3
^
•4—
/
P^
/
O
o
\
\
\
/
^
o
/
^
o
a
o
CI,
^
>.
—I
W
>
a
o
o
o
x:
<
:3
a"
Oh
CO
M
S:::
£
S
S
Macroeconomic
25
effects of regulation
the contract curve, and thus a decrease both in
employment and the
real
wage.
unemployment
the real wage and
1+M
1
i
=
'^'
mi^^^r'
1+
7/3/x'
real
1,
are given by:
P
The
—
and setting Pi/P
Tiurning to general equilibrium
wage
is
1
+ Pp
given by the same expression as before.
Unemployment
can be shown to be a decreasing function of p. So, the partial equilibrium
result extends to general equilibrium:
In the short run, a decrease in the
bargaining power of workers leads to both a decrease in their real wage and
an increase
in
unemployment.
Turning to the long run general equilibrium, the condition that the
profit
rate be equal to c gives:
ag{m)
So, just as in the
in the
real
number
=
benchmark model, a decrease
of firms,
and thus a decrease
wage must be equal to
1
—
in the
of firms, the lower value of
other, the increase in the
lower unemployment.
number
Prom
/?
On
leads to an increase
markup. As before, the
of firms,
effects of
the one hand, for a given
leads to higher
unemployment.
On
the
and the lower markup, leads to
the equations above, the net effect
unambiguous: Unemployment decreases
To summarize: At the
/?
There are now two opposite
c.
labor market deregulation on unemployment.
number
in
is
however
in the long run.
center of discussions of labor market deregulation
are often trade-offs between short-run
and long-run
effects.
In our bench-
Macroeconomic
mark model, the
in
exchange
26
effects of regulation
trade-off takes the
form of lower wages
unemployment
for lower
trade-off
may
tions, labor
market deregulation has no
in
extension
in the short run, the
effect
on
real wages,
and decreases
both the short and the long run. The second extension
shows however that,
if
workers have concave utility functions, the trade-off
even more stark than in the benchmark, with a decrease in real wages
and an increase
unemployment
5
first
be more attractive to workers: Under our specific assump-
unemployment
is
The
in the long run.
shows that to the extent that wages are allocative
in the short run,
in
in the short run, in
exchange
for lower
in the long run.
Interpreting
We now
unemployment
movements
turn to the evidence.
We
in the labor share
first
show how our approach leads
different interpretation of the evolution of the labor share
recent literature.
We
then look
at,
and
from that
to a
in the
interpret, the dramatic evolution of
the labor share in continental Europe over the last 20 years.
The
A
labor share and the
markup
large recent literature has focused
on the evolution of the labor share,
with the goal of inferring the evolution of the markup of the price over
marginal cost, and by implication, learning about the pricing behavior of
firms in imperfectly competitive markets.
The
logic of this
approach
is
as
follows:
•
Assume a production
dot stands for
•
Assume
all
function of the form
F{N,
.
),
where the
factors other than labor.
that the measured
wage
is
the correct measure of the cost of
labor to the firm, so nominal marginal cost
• Define the
Y —
markup
is
given by
of price over marginal cost as n, so
MC = W/F^.
Macroeconomic
P=
•
The
labor share, call
(1
it
+ m) MC =
a,
is
a =
The
27
effects of regulation
labor share
is
(1
then equal
WN
+ m) W/Fn
to:
1
FY
l
+ i^
Fn
Y/N
equal to one over one plus the markup, times the ratio
of the marginal to the average product of labor.
Much
of the
work has gone into trying to construct accurate measures
of the ratio of marginal to average product, so as to recover ^.^ Given that
this aspect
is
not our focus here,
let
us assume, as
model, that the production function
and average product
real
wage
is
simply given by
are both equal to one,
are therefore equal.
we did
in
our benchmark
y=
A^, so
marginal
and the labor share and the
In this case, there
is
a very simple relation
between the labor share and the markup:
W
Variations in the share
come
I
entirely
from variations
in the
markup, and
thus from the pricing behavior of firms in the goods market.
This approach (obviously using a more general specification
for the pro-
duction function) has been used to construct time series for the markup,
and characterize,
for
example,
its cyclical
ford [1999] and references therein),
its
behavior (Rotemberg and
relation to inflation in
with nominal price setting (GaU and Gertler
[1999],
Sbordone
Wood-
an economy
[2000]), or the
lower frequency evolution of markups in Europe over the past few decades
(Bentolila
and Saint-Paul
[1999]).
*See the careful discussion in Rotemberg and Woodford [1999].
Macroeconomic
From
research
Under
wage
the perspective of our paper, the crucial assumption in that hne of
is
that the
efficient
is
effects of regulation
wage
is
an accurate measure of the cost of labor to
bargaining, and
more generally
not fully allocative, this assumption
bargaining, the shadow cost of labor
markup over
sets its price as a
this
is
any model
in
in
firms.
which the
not correct. Under efficient
the reservation wage, and the firm
is
shadow
seen, to a different expression for the real
cost.
And
wage
—and,
the labor share. Instead of being given by equation
we have
this leads, as
by implication,
(5.1),
the share
is
for
given
by:
W
The
labor share
the labor market.
and thus
of the
1+1.1(3
now depends on
As
in (5.1),
it
monopoly power
The
bargaining power of workers.
is
imperfections in both the goods and
a decreasing function of the markup,
of firms.
But
it
is
also
depends on the
higher their power, the higher the labor
share.
This has two practical implications.
locative, the construction of
What
is
markups
First,
if
the wage
not fully
is
in the existing literature
interpreted as an increase in the
markup may
is
al-
incorrect.
in fact reflect lower
bargaining power of workers in the labor market. Second, movements in the
labor share
may come from changes
in the structure of either the
the labor market. In the rest of the section,
we show how
be used to interpret the evolution of the labor share
this
goods or
approach can
in continental
Europe
over the last two decades.
The
evolution of
unemployment and the labor share
in
Europe
Figure 4 shows the two macroeconomic evolutions which motivated this
research in the
first
place.
It
plots the
unemployment
rate,
and the labor
share in the business sector, for four major European countries, Germany,
o
o
(0
Q>
o
o
00
00
"CD
OS
^
CD
CD
CJ5
(f>
^
o o
nm oCM
^~ o
1^
OS
0)
0)
(0
,1^
cf
cc
r^
l/J
CO
"C:
c
c:
CD
<u
c ^
0)
E
>>
o
Q.
CD
iP
CD
CM
CM
o
as
"05
00
00
00
00
CO
00
O
o
CO
00
ns
E
00
00
>»
_o
o
a
CM
"00
E
o
c
n
CM
"00
(0
o
o
c:
en
"00
"00
LL
00
00
E ^
C
0)
c
t.0
CD
CD
>-
i
CD
^
h-
CD
Cvl
Csl
0)
o
o
"h~
ID
CM
O
O
CM
O
in
o
o
o
in
o
o
o
o
o
in
in
m o
o
o in o
in
in
CM
CM
o
o in o
o
in
CO
CD
CD
00
CO
o o o o o o o o o o
Macroeconomic
29
effects of regulation
France, Italy, and Spain, since 1970.
The
•
increase in
unemployment
last
unemployment
much
for
It
points to two
in the 1980s,
main
facts:
and the persistence of high
Only
of the 1990SjJn all four countries.
in the
few years has unemployment started to decline, most dramatically
in Spain, least so in Italy.
The sharp decrease
•
in labor shares, starting in the early to mid-1980s.
For France, for example, the labor share, which had increased from
70%
in the early
74%
1970s to
to 60%, a decrease of
14%
of
in the early 1980s, has
GDP
now decreased
10%
relative to the peak,
relative
to the 1970 value.^
The
focus on only four countries
in labor shares has
This
is
common
been
in sharp contrast
is
to
The
for visual convenience:
decrease
most continental European countries.
however to the United States, where the labor share
has remained extremely stable.
One can
think of a
them having nothing
number
of explanations for these
two
facts,
many
of
One
is
to do with rents, regulation, or deregulation.
that an increase in the wage (relative to the level of technology) starting in
,
the 1970s, led initially to an increase in the labor share and not
effect
much adverse
on employment. But, as firms started substituting away from labor,
the labor share started to
reflected in Figure 4.
unemployment
fall,
and unemployment to
rise,
the evolution
In that interpretation, the low labor share and high
in the 1990s reflect too high a wage,
and a high
elasticity of
substitution between capital and labor in the long run.
The purpose
was to examine
^The business
evidence, there
is
of an earlier paper
this
by one of the authors (Blanchard
[1997])
and other explanations. The conclusion was that
sector data stop in 1998.
Based on aggregate rather
no evidence of substantial change since then.
ex-
thaji business sector
Macroeconomic
30
effects of regulation
planations based on the reaction of firms to wages and the cost of capital in
competitive goods and labor markets just could not explain the data. Readers are refered to that article (and to the follow-up in
for
a
full
Blanchard
[2000])
treatment. But the basic reason for that conclusion can be shown
simply:
Assume
that firms indeed act as profit maximizers in competitive goods
and labor markets. Assume
also that, leaving aside adjustment costs, their
production function exhibits constant returns to capital and labor. Assume
finally that technological progress is
Harrod neutral
quired to obtain a balanced growth path
be written as
F = F{AN, K),
where
A
—the
assumption
re-
—so the production function can
is
the technological
Under those two assumptions, and ignoring
level.
costs of adjustment, profit
maximization implies the following relation between the labor-capital ratio
and the
real wage:
{AN/K) = f{W/A),
Call
Then,
AN labor
in efficiency units,
and
/'(.)<0
W/A
(5.3)
the wage in efficiency units.
this relation states, the ratio of labor in efficiency units to capital
should be a decreasing function of the wage in efficiency units, with the
property of /(.) deriving from the properties of the production function.
This suggests a simple step, that of building and plotting
for
each country over the period. This
evolution of the logarithms of
data axe available
zero in 1971.
•
The
figure yields
a "wage push"
.
From
in
Figure
5,
and
W/A
which gives the
from 1971 on (not
all
the
each country. Both series are normalized to
two important conclusions:
associated with an increase in
W/A,
thus with
Associated with this increase was a sharp drop in the
labor-capital ratio, in
•
done
AN/K and of W/A,
for 1970), for
The 1970s were indeed
is
AN/K
AN/K.
the early 1980s on however,
W/A
started decreasing in
all
four
(0
o
00
(0
0)
3
Macroeconomic
countries.
31
effects of regulation
By
the late 1980s,
in all four countries.
Yet,
was actually lower than
it
and here
is
its
1971 value
the proximate cause for the
W/A
decrease in the labor share, this return of
to
its earlier level
has not been associated with a return of the labor capital ratio to
The
earlier level.
now appears
ratio
its
to have stabilized, but at a level
substantially lower than in the early 1970s. Thus, the decrease in the
wage
since the
mid-1980s has not been associated with an increase
in
employment.
Could
effects
this evolution
be explained by the long lags and the expectational
coming from a putty-clay technology, or by
costs of adjusting factor
proportions more generally? This seems unlikely. Only expectations of large
incieases in
W/A
has recovered so
in
Europe today,
in the future
little,
this
if
at
all.
could explain
why
the labor-capital ratio
Given the emphasis on wage moderation
seems implausible, and suggests the need to look
for
other explanations.
Labor market deregulation?
The model developed
namely a decrease
starting
some time
curve
in the
in
mid
shift
can
in the labor share,
is vertical)
paper suggests one such explanation,
this
in the bargaining
As we saw, such a
wage and
in
1980s.
power of workers, a decrease
or an increase
generate both a decrease in the real
either
(if
no improvement
the contract curve
unemployment. Both implications seem to
^"This
is
fit
not the only alternative explanation. Another
rather than being Harrod neutral, has
Europe (but not
in the
see Blanchard [1997]
j3,
^^
initially
and
in
is
(if
the contract
upward sloping)
the facts. Under this in-
is
that technological progress,
become biased against labor
in the last
20 years in
United States). For further discussion of this and other alternatives,
and Blanchard
[2000].
Macroeconomic
effects of regulation
terpretation, the decrease in the bargaining
power of workers, starting in
the mid 1980s, has led to a reduction in the real wage at a given level of
employment, and, by implication, a reduction of the labor share. So
employment
the
far,
have been limited. But, our model suggests, the future
effects
should be brighter, and employment should eventually more than recover.
Indeed, the steady decline in unemployment since 1998
now
these dynamics are
How
plausible
is
two elements. The
may be
a sign that
at work.
this interpretation?
first is
that, since the
Note that the argument must have
mid
1980s,
Europe has gone through
labor market deregulation, at least in the sense of a decrease in the bargain-
The second
ing power of workers.
lation have so far
dominated the
is
that the effects of labor market deregu-
effects of
product market deregulation. In
terms of our model, to explain the decrease in the labor share,
that the decline in
implies,
/?
has dominated the decline in
we would have seen an
and a stronger decline
How much
in
for
must be
Otherwise, our model
increase, not a decrease, in the labor share,
unemployment.
direct evidence
is
there for each of these two elements?
The
The main
rea-
honest answer must be: Not much, either
son
/i.
it
for, or
against.
our ignorance comes from a number of conceptual and empirical
issues, first in developing
measures of the relevant institutions, second in
constructing the evolution of these measures over time.
Most
of
what we know
at this point
comes from work by the
changes in product market and labor market regulations in
tries. Initial results
OECD
member
on
coun-
from the development and the construction of a number
of measures are reported in Boeri et
al.
[2000].
Two
of findings of the study
are relevant here:
• First, the quantitative evidence points indeed to deregulation
both
in
goods and labor markets in Europe since the mid 1980s.
•
Second
— and
this goes against the hypothesis
developed here
—product
Macroeconomic
33
effects of regulation
market deregulation, which has
initiatives,
talcen place largely as a result of E.U.
has been widespread,
if
By
slow moving.
contrast, labor
market reforms have been more piecemeal. This second piece of
dence
the
may
OECD
not be conclusive however.
and examined by Boeri
The
evi-
indicators constructed by
et al for the labor
market focus
on employment protection, unemployment benefits, and the generos-
Of
ity of pensions.
in
those, only the first
is likely
to be related to
(3
our model. And, going beyond these indicators, the larger picture
appears to be one of a weakening of unions throughout Europe. The
unionization rate has decreased, often substantially, in most European
countries over the last 15 years (see for example
The
Booth
et al.
[2000].)
general attitude of governments towards unions also appears to
have changed, and so has the attitude of unions themselves.
Indeed this weakening of unions points to potential interactions between
product market and labor market deregulation, the topic to which we now
turn.
6
Interactions between product
and labor market deregula-
tion
Another dimension
the political
which our analysis can be helpful
in
economy
of deregulation.
The questions
is
in thinking
about
are many, ranging from
simple to complex, from positive to normative. Consider the following
•
Can we
explain exactly
why workers
so strongly oppose labor market
deregulation (in the sense of a decrease in
gains?
•
What
Can we
/?)?
Who
loses
and who
are the intertemporal trade-offs?
explain
deregulation
list:
why workers
—although
also so often
not always with the
do labor market deregulation?
oppose product market
same
intensity as they
Macroeconomic
• If
effects of regulation
we think
of the degree of regulation as endogenous, say as the result
of lobbying
by workers or
firms,
market and the product market
More
"^^^^
it
likely to help
facts in the previous
be the case that the product market deregulation
coming from European integration has naturally
deregulation?
one
"
macroeconomic
pointedly, in light of the
section, could
are deregulation in the labor
likely to interact? Is
or hinder the other?
•
how
led to labor
market
Might the reaction be strong enough so as to lead to
the observed decrease in the labor share?
Fully answering these questions would require another paper.
model
gives a
number
of hints,
which we believe are
likely to
But our
be robust to
further analysis.
We
proceed in two steps. The
terize utility, for
pedestrian, step
first,
is
simply to charac-
both employed and unemployed workers, together with the
unconditional probability of being unemployed (equivalently the unemploy-
ment
rate), in the short
and the long run. The second
the questions above, and point to
Utilities,
the
some
is
then to return to
of the implications of our results.
Unemployment Rate, and Firm Employment
Denote the (one period)
utility of
an employed worker by
(one period) utility of an unemployed worker by
V^
.
V^
,
and the
Then, collecting earlier
results:
In partial equilibrium, and in the short run, the utility of being employed
is
given by:
F^ =
Employment
in firm
i is
(1
+ /?m)/(w)
given by:
Ni
= N{1 + fi)-"
Macroeconomic
35
effects of regulation
In general equilibrium, and
in the short run, the
still
two
utilities are
given by:
ylM
The unemployment
^ ^_r^
yU ^
and employment
rate,
u\f{u)^-^
Ni
in firm
i
are given by:
= -{l-u)
In general equilibrium, and in the long run, the two utilities are given
by:
V^'^l-c
The unemployment
u
I
f(u)
—
To go from
welfare,
""
l
1-/3
and employment
rate,
c
Ni
1
—L
=
(1
—
in firm
u)
i
are given by:
where ml a q(m)
—
1
—
this to a characterization of the effects of deregulation
we need three more
• First,
V^ =
elements:
and obviously, the discount rate used by workers and
trepreneurs to assess utility in the second period
• Second, a characterization of
ample before or
unemployed
on
when
after workers
in the first
period
the reform
en-
—the long run.
introduced —
is
know whether they
for ex-
are employed or
—the short run.^^
''These veil-of-ignorance issues are familiar from the research on poHtical economy of
reform. For example, see Fernandez and Rodrik [1991].
Macroeconomic
36
effects of regulation
mapping aggregate and firm em-
• Third, a characterization of the rules
ployment into individual employment
employment status
in the short run,
the
in the long
then
unemployment
all
run
rate in the long run.
If,
for
for
example,
what happens to
is
instead, workers currently
employment
what happens
to
in the long
employment
in the
example, reform leads to entry of new firms, employment
in existing firms
We
If,
know
in that firm's
run, then they will also care about
If,
unrelated to employment status
workersTtie'ed to
employed by a firm have priority
firm:
is
probabilities.^'^
may
decline, even
aggregate employment increases.
if
have nothing to add here, nor do we want to give an encyclopedic
treatment of
all
possible cases. In
what
follows,
we
focus on the case where
reforms are announced after the pre-reform equilibrium has been realized
so workers
know
those workers
same firm
their pre-reform
who
employment
We
are currently employed have priority in
in the long
run
—
so, if there is
shall concentrate
shall
assume that
employment
at the
no decrease in the employment
of an incumbent firm, their probability of
we
status.
unemployment
is
zero.
level
Finally,
on the welfare of those workers employed pre-reform.
Labor Meirket Deregulation
The first two effects of labor
deregulation (a decrease in
them unchanged
wages
in the short run, leave
who
sure to be employed in both periods, the
is
j3)
are to decrease
in the long run. For a
eflFect is
worker
thus unambiguously
negative.
There are employment
effects however. In the short run,
both aggregate
unemployment and firm employment remain unchanged. But,
run, entry of firms
is
likely to decrease
employment
^^The issues are familiar from insider-outsider models.
example
in the analysis of labor
in
incumbent firms
They play
market reform by Saint-Paul
[2000].
in the long
a central role for
Macroeconomic
31
effects of regulation
(Recall that the effect
formally ambiguous, because of the decrease in
is
aggregate unemployment.) This in turn implies that the currently employed
workers
now
decreasing their
In short,
ulation
those
is
becoming unemployed, further
face a positive probability of
utility.
why
currently employed workers oppose labor market dereg-
rather straightforward:
They
who would have been unemployed
lose
from
it.
in the future:
Those who gain are
some
employed, and those who remain unemployed have a higher
of
them end up
level of utility.
Product market deregulation
Think of a deregulation
as a decrease in
As we have
\i.
run, this decrease can be achieved through an increase in
it
must be achieved through a decrease
in
seen, in the short
ct;
in the long run,
c.
In both the short and the long run, the effects on the utility of workers,
whether employed or unemployed, are unambiguously favorable:
in
//
leads to an increase in both
c also leads to
an increase
in
V^
V^
and
and
V^
V^
A
decrease
in the short run; a decrease in
in the long run.
So why don't workers more strongly endorse product market deregulation?
The model
suggests two reasons.
First, in partial equilibrium, deregulation decreases the rents to the firm,
and thus the rents to the workers. Under symmetry,
perception
is,
as
we have
seen, misleading, as the decline in prices elsewhere
more than compensates workers
tion only affects part of the
competitive, or because
this partial equilibrium
it
for the decrease in rents. But,
economy (because the
rest of the
if
deregula-
economy was
remains regulated), then the partial equilibrium
argument may extend to general equilibrium.
If
the deregulated sector
small enough, the partial equilibrium effect will indeed dominate, and
workers in that sector worse
is
make
off.
Second, and as in the case of labor market deregulation, lower markups
come, in the long run, from entry of new firms, and higher competition.
Macroeconomic
38
effects of regulation
Thus, employment in incumbent firms
risk of
them
unemployment
for currently
is
likely to decrease, increasing the
employed workers. This again may lead
to oppose product market deregulation, despite higher wages and lower
"
unemployment.
'
Interactions
The remarks above
may
increases the wage,
initially decreases
suggest that product market deregulation, which
it.
help implement labor market deregulation, which
Because both however lead to entry, and a
decrease in employment in incumbent firms, this
may
likely
not be enough to get
the support of employed workers.
We
that,
if
want to explore a
different interaction however,
based on the notion
product market deregulation decreases total rents, the incentives
workers to appropriate a proportion of these rents
may be decreased,
for
leading
to labor market deregulation.^"^
We think of product market deregulation
as a decrease in a, which in turn leads to a decrease in
We think of
employed workers as lobbying for a higher value of
We assume that they
To do so, we proceed
as follows.
/i.
/?.
maximize the
utility of
being employed, net of lobbying costs:
+ Z^M
l + fx
l
The
first
term
cost of lobbying,
is
the short run utility
which
is
when employed. The second
taken to be quadratic in
Maximization with respect to
'
£^2
2^
/3
/?,
and a
is
is
the
a parameter.
yields:
This idea has been explored, in a partial equilibrium context, in both the labor and
industrial organization literature. See for
a survey, or more recently Neven
et al.
the Journal of Economic Perspectives.
example Joskow and Rose
[1998],
and
articles in the
[1989], Section 9, for
Summer
1999 issue of
Macroeconomic
39
effects of regulation
"
0-'
a
So a decrease
in
ji
1
+ /i
leads to a decrease in
leads to labor market deregulation.
Having
fight less hard, or nearly equivalently,
/?.
Product market deregulation
appropriate, unions
less rents to
workers are
less likely to join unions,
making them weaker.
Can product market
real
deregulation in the end lead to a lower, not a higher,
wage? In other words, can the indirect
dominate the direct
/?,
theorem, we
know
The answer
is
effect,
through the decrease in ^ (by the envelope
effect
that net utihty above must go
yes.
through the decrease in
up
if /u
straightforward to show that a condition for
It is
product market deregulation to lead to a lower real wage
I-l)
so
>
1 (or
equivalently that
P responds
strongly to
Onr formalization
be to endogenize
/j,,
has two merits. First,
it
>
/?
1/2). This will occur
we saw
a
that (2/a)/u/(l
to consider
both the short and the long run. But
unemployment
labor market outcomes
wage
As
is
well
is
it
may
help explain the macroeconomic
in
Europe since the mid 1980s.
by putting into question
known,
efficient
bargaining
is
"efficient
employment below the
level
and the
^"^
may have
affected
bargaining" in the labor
not time consistent:
typically exceeds the marginal revenue product of labor, firms
to reduce
it
shows some of the complementarities between the
'*A rather different way in which higher product market competition
market.
small enough,
is
in the previous section, the decline in the labor share
persistence of
+
more than an example. Further steps would
two types of deregulation. Second,
facts
if
is
/i.
is little
and
goes down)?
Given that the
would
agreed to in the bargain. (In Figure
1,
like,
ex-post,
given
Wr/P,
the firm would increase profit in the short run by choosing a level of employment on the
MRP
curve, thus to the left of Ni).
Macroeconomic
40
effects of regulation
Conclusions
7
Our main purpose
paper was to construct a general equilibrium model
in this
with rents and bargaining, and use
it
to think about the effects of product
and labor market deregulation. Our formalization strategy was to construct
what we think
is
the simplest model required for those purposes. Further
progress clearly requires extending the model, in at least two dimensions:
• First,
by fleshing out the relation of the parameters of the model to the
the specific institutions which determine rents in the labor and goods
A natural way of doing so is by integrating our approach with
market.
that of Caballero and
Hammour
and
Hammour
[1998], Caballero
by allowing
• Second,
by allowing
for
(for
example Caballero and
[1999]).
for a richer description of technology, in particular
both labor and capital in production, along the
of Spector [2000]
.
If
entrepreneurs,
and
between work-
rentiers, and,
by implication, the interactions
may still
decide to choose the efficient outcome;
firms have long enough horizons, they
by foregoing
profitable
profits in the short run,
outcome
in the long
competition, however,
interpretation,
they are able to achieve a more
run (see Espinosa and Rhee
may make
it
what we have seen
in the short
[1989]).
is
efficient
and more
Higher product market
more tempting to achieve short-run
profits.
Under
this
a change in the nature of bargaining, leading firms
to reduce employment at a given real wage.
both
lines
Introducing capital would not only lead to a richer
picture, but also allow to capture the fight for rents
ers,
Hammour
If so,
the qualitative effects are very similar,
and the long run, to those of a decrease
in
/3.
Using our
earlier results for
the characterization of the economy under efficient bargaining and ex-post determination
of employment,
it
follows that this change in bargaining leads to a decrease in the labor
share and an increase in iinemployment in the short run, and to lower unemployment in
the long run.
Macroeconomic
effects of regulation
41
between labor, product, and financial market reform.
Building on these two extensions, one should then revisit the issues we
discussed in the previous two sections.
In Section
employment
5,
in
we argued that the evolution
Europe
in the last 20 years
towards labor market deregulation.
done to identify the respective
But
of the labor share
was suggestive of a movement
here,
much work remains
and
In Section
size of rents
6,
to be
roles of biased technological progress, of the
adjustment of factor proportions to factor prices, and of changes
tribution
and un-
due to changes
we focused on the
in the dis-
in regulation.
political
economy
interactions between
product and labor market deregulation. But, here again, we only scratched
the surface. Progress here also requires focusing on the role of specific institutions, as well as building
on and integrating the work already done on
the political economy of labor market and of product market reform.
Macroeconomic
effects of regulation
J^2
References
and Saint-Paul,
Bcntolila, S.
Explaining movements in the labor share,
G., 1999,
Working Paper 9905, CEMFI, Madrid.
Blanchard, O., 1997, The
medium
run, Brookings Papers
on Economic Activity
2,
89-158.
Blanchard, 0., 2000,
The Economics of Unemployment. Shocks,
Interactions, Lionel
Boeri, T., Nicoletti, G.,
Robbins Lectures,
and Scarpetta,
What do
Regulation and labour market
CEPR.
Checchi, D., Naylor, R., and Visser,
J.,
2000,
unions do in Europe?, mimeo, Fondazione Rodolfo DeBenedetti.
Caballero, R. and
of Political
Caballero, R.
stitution
L.,
and
Press, forthcoming.
2000,
S.,
performance, Discussion Paper 2420,
Booth, A., Burda, M., Calmfors,
MIT
Institutions,
Hammom", M.,
Economy 106(4),
1998,
The macroeconomics
of specificity. Journal
724-767.
and Hammour, M., 1999, Jobless growth:
appropriability, factor sub-
and unemployment, Carnegie- Rochester Conference Series on Public
Policy.
Espinosa, M. and Rhee, C., 1989,
wage bargaining
Efficient
as a repeated
game.
Quarterly Journal of Economics 104-3, 565-588.
Fernandez, R. and Rodrik, D., 1991,
Resistance to reform:
Status quo bias in
the presence of individual specific uncertainty, American Economic Review 81,
1146-1155.
Gali, J.
ysis.
and
Gertler, M., 1999, Inflation dynamics:
A
structural econometric anal-
Journal of Monetary Economics 44, 195-222.
Joskow, P. and Rose, N., 1989, The effects of economic regulation,
Industrial Organization II, 1450-1506,
Handbook of
R. Schmalensee and R.D. Willig, eds.
Amsterdam: Elsevier Science Publishers.
Layaa'd, R.
and
Nickell, S., 1990,
Is
unemployment lower
if
unions bargain over
Macroeconomic
effects of regulation
43
employment?, Quarterly Journal of Economics 105-3, 773-787.
McDonald,
I.
and Solow,
Economic Review
McKinsey Global
R., 1983,
Wage
bargaining and employment, American
73, 896-908.
Institute, 1997,
France and Germany.
Neven, D., Roller, L.-H., and Zhang,
Z.,
1998,
Union power and product mar-
ket competition: Evidence from the airline industry,
CEPR
Discussion Paper
(1912).
Rotemberg,
J.
and Woodford, M., 1999, The
cyclical behavior of prices
Handbook of Macroeconomics IB, 1051-1133,
J.
Taylor and
J.
and
costs.
Rotemberg
eds,
Elsevier.
Saint-Paul, G., 2000,
The
Political
Economy
of Labor Market Reforms,
Oxford
University Press, Oxford, forthcoming.
Sbordone, A., 2000,
An
optimizing model of U.S. wage and price dynamics, mimeo,
Rutgers University.
Spector, D., 2000, Competition and the capital-labor conflict, mimeo,
MIT.
'J
6tv5
039
Date Due
MIT LIBRARIES
3 9080 02246 0742