Declining Inequality in Latin America: Are the Good Times Over?

Declining Inequality in Latin America: Are the Good Times Over?


[MUSIC PLAYING] So I’m delighted to chair
the first of our panels in this conference. And we have a very
nice group, even though two of our
panelists had to drop out. Lopez-Calva and
Facundo Alvaredo, one for family reasons and the
other one for work reasons, but the remainders
are great, so we are sure this panel
will be quite good. The first speaker is going to
be Francisco Ferreira who is now the World Bank’s Chief
Economist for the Africa region and also part of this
project I was telling you in which the World Bank is
working on inequality dynamics and also a co-editor of the
special issue of the Journal of Economic Inequalities,
a special issue dedicated to the appraisal of
databases on inequality. And Chico has many publications
in very prestigious journals. Despite being among the
multilateral institutions, he skipped up with this
publication record, and was a co-director of the
World Government Report in 2006 on Equity and Development. And now he’s going to be moving
to research very soon where he’s going to be the adviser,
the chief adviser you’re going to be, right? Senior Advisor. Senior Adviser in the Research
Department at the World Bank. He will be overseeing the
Poverty and Inequality in Agricultural Development. And among other things,
one of his stance is to help redefine the poverty
line, international poverty line, now that there is a new
set of purchasing power parity information that came
out of the 2011 Round. So we’ll have a
new property nine. The 125 will be old news
after six months or so. So Chico, without further
ado, the floor is yours. And I’m going to
do introduce you every time that you present. Thank you, Nora. And thanks to both Nora
and the other organizers for the invitation. The day started very well
with a great opening panel, and I’m looking forward to
the rest of the conference. What I want to do in the
next 15 minutes is go over quickly a brief presentation
of a joint paper that I’m writing with
Julian Messina, who’s here, and Sergio Firpo, our
co-author from the Sao Paulo School of Economics
which is called A More Level Playing Field? explaining the
decline in earnings inequality in Brazil
in ’95 to 2012. Short time, so four things I
want to do is just tell you what the question is that
we’re trying to answer. The suspects fit
like a mystery novel. The suspects, very brief on
the data and methodology, and some results and conclusion. And the question is, what
accounts for the 10 point decline in the Gini
coefficient for labor incomes in this period that I
say there, between ’95 and 2012 in Brazil. So there was a
paragraph motivating each panel for this conference. In the one for
this panel, there’s a lot of different questions. One of them is, what’s happening
to earnings and equality? And so this is what this focuses
on very, very, very narrowly. And in fact, it’s related to
two points that Rebecca made. I don’t think
Rebecca’s here now. Oh, she is. So Rebecca made points about the
importance of labor earnings, the primary income distribution
in labor markets in particular, and also issues around
horizontal inequality which weren’t intended to
be the focus of this paper but turned out to be important
in some of our finding. So the question is a very
narrow and well-defined one. This is not a particular
big picture paper. It’s really looking at this
evolution in the earnings inequality which took place
in the context of a broader decline in inequality
in household per capita income in the period. So what you see in this graph
is actually per capita income, the Gini coefficient for
household per capita income, and Brazil is, of course, just
one example of this broader trend in Latin America that
Nora described in her remarks earlier. And you see in Brazil,
this flat, very high level of income inequality at
around 60 until 2002 or so, and then a considerable
decline of about seven Gini points to 2011, and
it goes on after that. But that’s where
our data starts. And as Nora has
already indicated, and in another version, a longer
version of this presentation, I have a slide that
refers to other work that suggests that various
factors are behind this, but the dynamics in
the labor markets or the dynamics around
the earnings distribution are probably the most important. And this is a point that I
think both Nora and Rebecca made earlier. They are true in Brazil as well. So in this paper, we focus only
on labor income inequality. And here you’ve got
again the household per capita income line. These are confidence
intervals which are redundant in this case
because they’re so small. [LAUGHTER] But anyway, this
is labor income. This is what we’re looking at. And you’ll see this decline,
of 20%, from 0.5 to 0.4 in Gini points, which
is a marked decline. It is a substantial
change in the distribution of earnings in Brazil
over these 17 years or so. And the same picture
holds, in fact it is more pronounced
in proportional terms for alternative
measures of inequality, like the tile index or the 90:10
percentile ratios which I don’t show here for reasons of time. So what’s behind this? And so we, thinking of a
standard mincer framework or human capital
framework, we could think of a number of possible
candidates or suspects to account for this change. The first is the one that Nora
emphasized in her presentation, which is a human capital story. So we know that
there was an increase in the supply of education over
this period, which outpaced the increase in the
demand for education, so there’s both an increase
in supply and declines in the return to education. On the side of experience,
the labor force in Brazil was also aging, so potential
experience was increasing. But in addition to that core
story, which is in some sense the received wisdom now in
the literature about this, there are a number
of alternative things that could be going
on, some of which was also mentioned earlier. One of them is actual changes
in labor market institutions. So there was, as I’ll
show you in a moment, a very marked rise in the
real value of the minimum wage in Brazil in this period. There was also a
considerable change in the formal/informal
breakdown of the labor force with the Brazilian
labor force formalizing quite markedly over this period. So that had been less
studied to some extent. In Macelo here and
others have worked for minimum wages in
the past, but the role of this in this
particular context we felt deserved looking at as
well in this context. There are some changes in
what I call demographics. So the gender and racial
composition of the labor force, and potentially
there may have been changes in returns
to those characteristics and the degree of quote
unquote “discrimination” if we suspend this
disbelief for a moment now. But let’s not call
it discrimination, the gaps in gender and race
controlling for other things. And presumably, there could
be spatial factors as well. That’s something else
that Rebecca mentioned in terms of territorial
and spatial inequalities that could be changing. This was a period
of a commodity boom. Brazil was a big exporter of
soybeans and other things. It’s possible that
rural incomes were rising for reasons
that are not to do with education, for example. So this is what we’re
looking at a little bit. So let me show you,
what are they called, mug shots of some
of these suspects. [LAUGHTER] So here’s a mug
shot of the increase in the supply of education. So this is the cumulative
distribution function of years of schooling. So I think everyone in
this room knows a lower CDF is one which is better. So the density function
is shifting to the left. This is 1996. To the right, sorry. This is 2003. This is 2012. So you see this big increase
in years of schooling. One way to see that is
to focus, for example, on what’s happening at ten
years of schooling which is secondary incomplete. So what this tells you is that
at the beginning of the period, around 75% of the
Brazilian labor force had secondary incomplete. By the end of the period that
was lower than 50%, the rest being to the right, if
you see what I’m saying. So the distribution
of education, the quantity of
years of schooling, we’re shifting dramatically. And at the same time, as I said,
I won’t show you the picture, the population was aging though
that was, of course, a more gradual process. Moving from the human capital
to the labor market institution suspects and mug shots, here’s a
mug shots of the rising minimum wage. These are three wage
series, average earnings, median earnings, and
the minimum wage, normalized to be one at 95. So these are indices. You see in the average
and the median a decline. And average quite
marked, and median not so marked until around 2002. So actual levels were
falling in this period and then increasing afterwards. But throughout the
period, certainly since around ’97 or
’98, a very marked increase in minimum wage, which
actually doubled in real terms over this period. Now as a result of this, the
density function of earnings, I don’t know how clear
that is, but it’s actually probably clearer if you’re
looking at it that way rather than diagonally, but you
see I quite like this picture. This is the density function
shifting to the right. And at the same time
as that’s happening, this big spike at
the minimum wage. So the vertical line is
the minimum wage becoming increasingly pronounced. You see the spike of the
minimum wage becoming increasingly pronounced. Now the results we found
for the minimum wage were actually not
what we expected. Well, at least we think in order
for us to understand what’s going on, it’s important to
note these little numbers here, which tell you the mass of
the density distribution below the minimum wage and
how it changed over time. And what happened was
it went from around 22% in the early period, to
30% in the middle period, and then down to 29%. So that means that
non-compliance with the minimum wage rose
over the initial period and then slightly declined. That will turn out to
be important later on. Remember it’s a
murder mystery, so keep that information
in your mind. [LAUGHTER] The labor market was also
formalizing, as I said, so informal employment
fell from 52%, that’s Saint Carteira for
the Brazilians in the room. That’s [SPEAKING SPANISH] So self employment and informal,
the sum went from 52% to 42%, so 10 percentage points
increase in formality over this period in Brazil,
another institutional change, or change in the structure
of the labor market, that might have played a role in
the distribution earnings. So we wanted to look at that. Gender and racial composition
of the labor force. So women’s participation
in the labor force went from 38% to 42%. The whites in the labor force
went from 57% to about 47%, so also a big increase in
the non-white composition of the labor force
over this period. Some continued urbanization with
rural changing from 19% to 14%. Almost no change in the
regional composition there across the five regions. So with those suspects,
what did we do? Basically, I’ll tell
you very briefly, given the time
constraints, what we did was basically
we ran a variant of a Oaxaca-Blinder
decomposition using the [INAUDIBLE] Method
that at least the economists in the room will
probably know, which is an unconditional
quantile regression applied to these compositions. The data we used is
the [? pinages ?], the bread and butter for
people looking at income distribution in Brazil. We compared those
pairs of years. We paired the years ’95 and
’96 as the early period, 2002, 2003, 2004
2005, 2011 2012. We focused on a
wage measure which was total gross monthly
individual labor earnings. So that’s not an hourly. It’s a monthly measure,
so there’s an hourly thing that’s maybe going on there. For 18 to 65-year-olds,
men and woman everywhere, we excluded
the rural north of Brazil because [? Pennagi ?] only
started covering that in 2004. So as I was saying, what we’re
interested in is understanding this changes in the
earnings distribution which can be summarized by various
distributional statistics, say the mean wage and
the Gini co-efficient, which I’m going to
show you results for. But you could also do it for
the entire quantile distribution or for the tile
index or whatever. What the Oaxaca-Blinder
decomposition does is, as you know,
and [INAUDIBLE] may allow us to do as well for
these more complex statistics is to decompose the overall
change, say in the Gini, into two components, one
that’s the endowment effect. That’s this one here. That’s this one. That’s the endowment effect. If you think in terms
of Oaxaca-Blinder, that’s the changes in the axis,
the changes in the quantities that I’ve shown you so far. And then the other
one is the change in returns, which
is the change in the conditional distribution. So if you think in
terms of Oaxaca-Blinder, the changes in the betas. So that’s what we do. And so we end up running
a regression, which is just like an OLS regression,
except as in the [INAUDIBLE] method, it’s a
regression not of incomes but of what they call
the re-centered influence function of incomes,
which basically is the influence of each
particular bit of income on the Gini, which is
called an influence function in statistics,
re-centered to the Gini. So it’s basically
some transformation. It’s fancy stuff. Our co-author, Sergio,
knows it better than Julian and me,
publishers in Econometrica allows us to do fancy things,
but it’s basically an OLS regression but with the
RIF here on the left hand side run on the suspects. So some human capital variables,
some demographic variables, some spatial variables,
minimum wage, and informality. In this version of
the paper, and I should say this is still a work
in progress and Julian, Sergio and I are still working
on it, and there will be some changes
in the final version, although I’m told the story
doesn’t change very much. Ask Julian about that. But in this specification, we
use fully categorical variables to allow for functional
form flexibility. And I’ll show you that
regression in the next slide, but just to say when you do
use dummies, which is basically what these categorical
variables are, then we know that the
Oaxaca-Blinder decomposition is not insensitive. It’s not invariant to the
base categories, the omitted categories. So one I must make a choice. The choice we made was
the one the specification that reduced the influence
of the constant term, the unexplained constant. So just so you know that these
things actually do matter. So I don’t expect you to
look at these regressions. This is just to show you
what the dummies are. So for age, we’ve got
those to consider. Five categories, one omitted. For schooling, you’ve got
those five categories, omitted is tertiary complete. Then we got a dummy for being
below the minimum wage, a dummy for self employment in
Saint Cartera omitted as [INAUDIBLE] Carteira. Three different racial
compositions, female dummy, rural dummy, and the regions. [SIDE CONVERSATION] So then, we have one
minute, the results. So we run those
regressions, and then we decompose for each
of those groups. So what we actually run
is, in each of those groups we don’t run this
exact regression, what we run is a regression where
within each of those groups there’s a full interaction
of all the datas. So age and education
are interacted in here, race and gender are
interacted in here, region and rural are
interacted in here, and so on. So that is four
interactions within that. It’s a fully saturated
model within the group to allow for as
much flexibility, again, as possible. So run that and then
separate the effect into endowments and structure. So again, if you’re
thinking of Oaxaca-Blinder, think of Xs and betas. And we have it for the
overall period and then the two sub-periods. So given the time, let me focus
on a few key results here. So first of all,
this is the mean. So there’s an
increase, of course, in mean earnings over this
period, a lot of which comes from human capital. This is the increase
in schooling that increases the
mean quite a lot. There’s also an increase
in the mean that’s coming, and this is largely
an endowment effect. There’s also an
increase in mean that’s coming from falling gaps
in gender and race, gaps, and from the minimum wage. But those are coming from
the betas rather than the Xs. Those are coming
from the structure. That is to say, it’s not
to do with the increase in labor force participation
of women necessarily. It’s to do with the fact
that the gap between men’s and women’s earnings or
between whites and blacks is decreasing in Brazil,
controlling for education. This is not an
educational effect. Then there’s the
sub-period breakdown which I don’t have time to go into. Now this is means, so
this is not inequality. So for inequality, let’s look
at the Gini a little bit. And here, the basic
story is there is a big decline in the whole
period, only a small part of which in our specification,
comes from the human capital story, which as
Nora pointed out is a big part of the story
in general and one that we’ve often focused
on as an important part. So what we see our paper
contributing to is, it conforms that it
does that, and it does that in a way that
we always expected, which is a countervailing
endowment effect and a strong structure effect. Let me say a word about that. So what this is, is
this is the returns falling as a result of their
being many more skilled people and fewer unskilled people. This is the returns falling. This is what Francois
Bourguignon, Nora, and I in some
earlier paper called, The Paradox of
Progress, which is that the mechanical effect
of having more schooling, when the returns to
schooling are convex, is actually to
increase inequality because there’s more mass in
the convex part of the solution. So together those lead to this
kind of decline over here. Now interesting things. There’s a big part of
the decline in the Gini which is due to these three
structural effects here. Those are decreasing gaps
between men and women, between the races. This one is urban and
rural, so the segmentation, the urban rural segmentation
and the regional segmentation falling, and a
greater formality. The last part of the story,
which was the surprising one, is the minimum
wage which I fully expected in the
beginning, minimum wages would be a contribution to this. It turns out, looking
at the overall period, that minimum wages
were unequalized. And the reason they
were unequalizing was an endowment effect. What’s an endowment effect? It’s composition, not returns. So that has to do
and that comes all from this first period
which is exactly the period when noncompliance
was rising from 22% to 30%. I gave you a hint earlier
that that was important. That is the fact
that minimum wage is rising during a time when the
labor market couldn’t support it, led to increases
in noncompliance that were unequalized. In the latter period when
the labor market was booming and could support the larger
increase in minimum wage, these minimum wages
were equalizing, both through an endowment and
through a structure effect. So that’s the story. And given the time, I’m
going to skip all of this and just go to my
concluding slide. So the basic message from
this little paper of ours is in contrast to
earlier periods, the story of these 17 years
in Brazil was a happy one. Unemployment, which I
didn’t show you, fell. Earnings rose, both
average and median. Not only did the
average earnings rise, but they rose most for
those groups of workers who used to earn the least. Not only workers
that were poorer, but also the racial
minorities, women and people in rural areas. There was a compression in
the schooling premia, which is very important, but
perhaps even more impressive was the reduction in
wage gaps amongst workers that are equivalent in
terms of human capital. Controlling for that,
there’s this big increase in people that differ along
race, gender, location, and type of job. For this conference, given the
title, the question to add here is I have no idea about this. [LAUGHTER] But I guess the
interesting question is which parts of
this happy story will survive into the new
regime, when growth is again lower, and we have a number
of complex challenges facing the Brazilian economy. That’s a question for
us all to discuss. So thanks very much. [APPLAUSE] Thank you very much, Chico. Now we turn to
Tasha Fairfield who is an Assistant Professor at
the London School of Economics and also an affiliate to
the Department of Government at LSE. And Tasha’s going to bring
in the top 1%, at least in one country. You’re going to look at Chili. And I’d like to
say, and I’m going to buy a book that she
has published or is about to come out, called,
Private Wealth and Public Revenue in Latin
America: Business, Power, and Tax Politics. It sounds very interesting. So Tasha, the floor is yours. Thank you so much. I’m really excited to be here. So my research
does, in fact, focus on unequality in taxation. And I’m going to start by saying
a few things about this paper with Michel Jorratt on
Chilean Top Income Shares, which is most directly
related to the panel themes. And then I’m going to say
a little bit about my core research on the
politics of inequality in this project on taxation
that I’ve been working on. So I just want to try out
some findings from this book that I think speak to the
role of economic elites and governance and prospects
for redistribution. So my co-author and I
after about 10 years were able to get access to the
full universe of income tax returns for two
years, 2005 and 2009. And using that data, we were
able to reconstruct top income shares without having to do
any extrapolation because of the full universe
that we got. We supplement that data with
the tabulated information from 2003 to 2013. Got a little bit of Macintosh
to PC conversion issues. So this is the only
data that’s publicly available, is this tabulated
data from 2003 to 2013. So our estimates for those
years are much more approximate. So if you focus on what
should be the blue line here, you can see that the
top 1% share in Chile is among the
highest in the world up there with Columbia and the
United States and South Africa. The red line above it is
showing top income shares that have been adjusted for evasion. Now the issue
that’s going on here is that Chile’s
tax system created particularly strong
incentives for business owners to under declare
their distributed profits and their dividends. So what happens is that if
you reinvested your profits in the firm, you were
paying only a very low 15% corporate tax rate. If you took out dividends
and distributed profits, those paid much higher personal
income tax rates of up to 40%. So to give you an idea of the
scale of the avoidance evasion problem that was created
by this very large gap between the personal income
tax and the corporate tax, national account
figures suggests that the actual amount
of distributed profits is three times as
much as what gets declared to the tax agency. So we think that is
important to look at, shares that have been
adjusted for evasion as well, even though the shares from the
rest of the Top Income Database don’t do that kind
of adjustment. You’re talking about
the [INAUDIBLE]. I’m sorry. You’re talking about the
database of the size. Right. So the other top
1% shares here have been taken from the
Top Incomes Database, so they do not take
evasion into account. But because of this
particular peculiar aspect of the Chilean Tax
System, we think that this is probably
a better way to compare or that these are certainly
underestimated for Chile. Now when we include
accrued profits in the definition
of income instead of distributive profits,
our top income shares go up quite a bit,
to at least 23%. So what you’re
seeing here is that including unrealized capital
income increases inequality. So this is a novel
aspect of our paper. We were able to get information
from business tax returns that allowed us to iteratively impute
accrued profits to the owners. And this is
something that hasn’t been possible to do for
other countries yet. So the best that we can
do in terms of comparisons is to look at top income
shares for other countries that include capital gains. So this is clearly not
entirely compatible in terms of comparisons, but
it does give you an idea of how extremely
concentrated capital income is in Chile. Now despite that very
tremendous income concentration, Chileans in the top
1% until recently were paying an effective
tax rate of only 15%. So you can compare that
to the rate in the US was actually 24%, which
was substantially higher. And of course, the US is
hardly an example of equity. So coming back to
the panel themes, have the rich been
getting richer? Well, from our data,
it’s not really clear whether we have a
decreasing trend for Chile, because remember, we just have
the micro data for these two years, 2005 in 2009,
and we don’t entirely trust what the tabulated
data is showing. But what is clear is that the
rich in Chile and elsewhere in Latin America, for
which we have studies, are very, very rich. And they could certainly
be paying higher taxes for the sake of contributing
to more inclusive societies. So that takes me to the
politics of inequality and redistribution. And when I started this project
on business power tax politics, scholars were paying a lot of
attention to social policy, but there was very
little work being done on the tax side
of redistribution, even though in a lot of cases
weak tax capacity really constrained social spending. And of course, we know
that low tax capacity creates all sorts of other
problems for the State’s ability to try to promote
inclusive development. But there is growing
recognition now that’s progressive taxes
on income and wealth could actually raise quite a
lot of revenue in Latin America, precisely because they’re
so concentrated at the top. And then in addition,
we have economists like Piketty and
Emmanuel Saez, who have shown that progressive
income taxation can really be an important redistributed
tool in its own rights. The main question
that I’m asking here is, essentially, how and when
can democracies tax economical elites. And my framework for
addressing this question draws on business power. So instrumental which
involves political engagement, and structural which has to
do with market reactions. So business, as
you would expect, is a very important
player in tax politics, whether we’re talking about
individual firms and investors or whether we’re talking
about business associations as organized political actors. My starting point
is that business and economic elites
more generally tend to resist higher taxes. They want to protect their
core material interests, but they may or
may not have power to be able to shape
policy decisions. So I’m basically trying to
get at these questions of who gets what, when, and how. And this business
power framework gives us a lot of leverage for
being able to explain variation in how and when elites are able
to get the policies that they want. So I just want to
take you very briefly through my
explanatory framework. Instrumental power
is the capacity to engage in deliberate
political actions, like lobbying. I’m using a power
resources approach, so I’m identifying specific
sources of instrumental power that help economic elites
shape policy outcomes. So these can be relationships
with policymakers that enhance access
and create bias in favor of economic elites. And they could be
other resources that help business get what
it wants more effectively. So for example, relationships
with policymakers would include ties
to political parties that treat business and economic
elites as a core constituency. Another example would be
recruitment into government, where business people
are given appointments in a cabinet or a ministry. Other resources here. Cohesion, I find to be
especially important. So by cohesion, I
just mean the capacity to engage in collective
action and maintain unity. So organization plays
an important role there. Structural power comes from
the profit-maximizing behavior of firms and investors. So if policymakers think
that a reform might lead to reduced
investments, they may rule it out for fear of
harming growth and employment. So the first key point
about structural power is that it does not
require any organization or political action. Instead market
signals coordinate how firms and investors
react in the economic arena. The second important point
is that structural power works through
policymakers’ expectations about how investors are going to
respond to a particular reform. So that means that
policymakers have to believe that the reform
that they’re considering will actually create negative
investment incentives that might mess up the economy. If they don’t, if they
don’t anticipate that, than structural
power is either weak, or it’s simply not
relevant for policy making. So what this means is that
policymakers’ perceptions matter for structural
power and whether it’s going to have an effect. And I just want
to point out there to remember that even
highly-trained economists may disagree about
what kind of effect a reform is going to have,
given all the complexities that we get in real
world economies. So I essentially argue
that business interests will shape policy decisions when
either their instrumental power or their structural
power is strong. And when you have
more than one type of power, more than
one source of power, you’re going to see
business getting what it wants more consistently
and more extensively. Business power explains
most of the cases that I look at in the book,
but in a handful of cases popular mobilization
mattered as well. The popular mobilization
can play an indirect role by generating revenue
needs, for example to finance social spending that
popular sectors are demanding. It could also play
a more direct role if popular sectors
reject broad-based taxes, so that can force governments
to try to raise revenue from economic elites instead. And you could even
have popular sectors mobilizing to demand
that economic elites pay higher taxes. Now what’s I think
quite remarkable here is how few cases there are
where this dynamic is relevant. So I see this in only two out
of 30 some reform episodes that I studied. And in cases where a popular
mobilization was absent, taxing economic elites rarely
became electrically salient. And this really
underscores the importance of looking at business
power, because in most cases taxing economic elites is
a realm of elite politics. So this is how my
argument plays out at the cross-national level
for these three countries that I look at. So in Chile,
businesses very strong instrumental power kept key
tax reforms off of the agenda until very recently. In Argentina, business
was generally much weaker, so governments had
a lot more leeway to be able to increase taxes on
income and wealth and profits. However some sectors,
like finance, did have substantial power
during specific time periods and that allowed those sectors
to block tax increases that had sector-specific impact. And then in Bolivia, we
have some very interesting interactions between
business power and popular
mobilization that led to variation across
different tax policy areas. In the case of Chile, we also
see some really interesting variation over time. So in previous years,
strong business power had kept important tax
reforms off of the agenda. After Bachelet was
reelected in 2013, we see this very radical
reform initiative that was designed to
dramatically increase taxation of economic elites. So what changed there
was, first of all, the student mobilization
counterbalanced business power. And secondly, businesses
instrumental power declined after the 2013
elections, in large part due to the right’s lots
of seats in Congress and their poor showing
in that election. Since the conference
agenda themes also include this issue of
politically feasible reforms for reducing inequality,
I also just wanted to flag that I identified this
menu of reform strategies that can be used to facilitate
at least incremental or marginal tax increases
when you strong business power and no popular
mobilization to speak of. So these reform
strategies essentially involve either tempering
elite antagonism, not bothering them too
much, and/or mobilizing public support. And there are ways
that you can do that through how you design
the policy on the tax side with this whole
equation, or there are ways to do that by linking
the tax increase to benefits for different sectors. Now I’ve been emphasizing
that public opinion matters much less for tax politics
than business power, but public opinion can
matter at the margins, and it can make a difference
for passing incremental reforms. So politicians who
might otherwise favor business and
economic elites can feel electoral
pressure to do otherwise. I just wanted to flag one
especially effective strategy for mobilizing
public support would be linking to popular
benefits, like social spending. And this is something
that governments in Chile have done ever since
1990 very effectively. So coming back to the
question of economic elites and their contribution to social
inclusion, a lot of authors have argued that strong business
organization and cohesion facilitate socially
desirable outcomes. So that would include
higher direct taxes and good economic governance
and equitable outcomes more broadly. But in highly unequal
societies, these factors may be much less beneficial. So in quite a number of
Latin American cases, we’ve seen strong
business organizations really helping to block
redistributive reforms. So the point that
I want to make here is that organization and
cohesion are sources of power. And economic elites can use
them to promote their interests in ways that may or may not
contribute to the broader public good. And then regarding
democracy and inequality, on the one hand this is not
a [? median ?] [? better ?] story. The poor are not setting
the tax rate on the rich. And by and large,
taxing economic elites has been a debate
between economic elites and political elites with this
very marginal role for voters. But this is not a story of
monolithic elite influence either. So money as a resource is
clearly very important, but there are other sources
of power that matter, too. And those sources of power,
including cohesion and linkages to particular parties
vary quite a bit, even across Latin America’s
highly unequal democracies. So this business power
framework, I think, really allows us to count
for the multiple means and multiple mechanisms
of influence. And just looking forward to the
panel tomorrow on governance, I think that this
kind of approach could be quite fruitful
for analyzing prospects for reducing inequality
in Latin America. Thank you. Thank you very much. [APPLAUSE] Thank you very much, Tasha. Now it’s Florencia Torche. She is an Associate Professor
in the Department of Sociology at NYU and also has an
affiliate of the Steinhardt School of Education
and the Global Institute of Public Health. And I think we’re
going to hear more about mobility,
intergenerational mobility, and how it’s
related to education and then inequality broadly
in an international comparison framework. I don’t know. Not quite. Thank you, Nora, very much. [LAUGHTER] [INAUDIBLE] Yeah, yeah, somewhat. [LAUGHTER] Well, anyway, it’s
going to be good. Does anybody have a laser
pointer by any chance? Don’t leave home without it. [LAUGHTER] Thank you very much always. I don’t know. So thanks so much
for the invitation and for organizing this
important conference. So I suppose I was
invited to talk about intergenerational
mobility which I’ve done some research on,
but to be honest with you I decided to take
advantage of the fact that I’m the only
sociologist in this panel and that I have some
dissatisfaction with the work on intergenerational
mobility, starting with my own to just move a
little beyond that. So I want to pose the
question about how does inequality matter? And talk a little bit
about potential cause why we care about inequality
and the potential consequences of inequality for the
intergenerational transmission of advantage. So I really enjoyed
and appreciated the first panel that gave
us a macro level overview. I want to talk about more micro
level mechanisms if you will. So one main reason why
high inequality matters is because it’s correlated with
low intergenerational mobility, low equality of opportunity. And we do know
that Latin America is the most unequal region
of the world as was shown. And it’s also based on the
very limited information that we have the least mobile. At least when we consider
economic mobility, the story’s different
for a class mobility as sociologists would claim. But as important as
these findings that has been systematized in
the Gatsby curve, that occur popularized, there are at
least two dissatisfactions with the focus on
intergenerational mobility as a negative corollary
and a source of concern with high inequality. The first is that
intergenerational mobility, that is the association between
well-being of the parents and well-being of
the adult children, refers to how things
were decades ago. The level of inequality,
the institutional framework, the policies in place when those
children, that now we observe are adults, are growing up. And therefore, that may
have little relevance for current policy making. As Nora was indicating
this morning, one important driver
of the decline in inequality in Latin America
is educational expansion that happened two decades ago. And that, of course, matters. But usually decision makers want
to know what to do right now. The second dissatisfaction
with the work on mobility, it’s bivariate focus. The study of
mobility is furiously bivariate in the
sense that it really focuses on parental income,
say adult children income. Anything that happens in
between, and what really matters in terms
of how inequality gets reproduced
across generation, is left a little bit aside. And that’s not a fault
of the study of mobility. Getting the intergenerational
association right is not minor feat,
but it just opens the question of what
is it that drives the transmission of advantage. And here we start, there’s
room for speculation. Some people claim it’s mostly by
the constraint, social capital, investment in education. Some people bring
genetics up, and it’s just in a transmission, family
socialization, and so on. So what I want to invite to
add to the study of mobility and the relevance of
inequality for these is the relationship
variously called between social origins,
circumstances, family background, whatever
happens in the origins, and children’s outcome
studies outcomes early in the life cycle,
before we move to when people become adults. And I want to focus on
the specific mechanisms for the transmission
of advantage across generations,
starting as early in the life course as possible. And this comes from
recent findings in the biological
sciences that suggest that early impacts,
early shocks, what happens to you early
on in the life course, can really set trajectories
of disadvantage that is hard to undo. I’m going to give you a very
brief example based on my work, because I know it
better, of course. So I’m going to use
the example of stress. Stress is one possible
mechanism whereby both absolute deprivation,
that is poverty, and relative deprivation,
that is inequality, where you are placed in the
hierarchy affects life chances. And the effect of
stress exposure, I’m going to claim here
trying to emphasize the early stages
of the life course, may start even before birth. And this matters because if
poor children are exposed to stress early on
in life, then that can set a set of developments
that are detrimental. The issue, of course,
is that stress is correlated with many terrible
things that happen to people. So if we want to
address this question, we need to isolate stress
from its unfortunately common correlates. How do we do that? One way to do it is
natural experiments. That’s what I do. I use a major earthquake
in 1982 in my native Chile as a natural
experiment for stress. Why? Because an earthquake is perhaps
the closest possible thing in the real world
that to allocating stress at random in the lab. Who gets the earthquake, who
doesn’t within Chile, which is prone to earthquakes
is really random. And because this earthquake
that happened in Chile in 2005 caused very high levels
of stress and anxiety, but very limited damage, so we
can claim with some certainty that stress and not
everything else. So using that and starting
as early as possible in the life cycle, I analyzed
the effect on birth outcomes. That’s the earliest I
could see, after birth. And exposure, this is
the time of exposure. This all based on a bunch, of
course, of inference approaches that I’m not going to
get into at all now. But exposure in the first
trimester of the pregnancy. If you were unlucky
enough to be in utero during the first
trimester of gestation, that population on
average gets a decline in birth weight of 50 grams. That’s small, but is an average
across the entire population. Then if we move to the
probability of low birth weight, weight below five
pounds, this is really bad. We know that low
birth weight is bad. We see that if you were
unlucky enough to be born, to be in utero in
the first trimester when the earthquake
occurred, that population experienced an increase
in low birth weight by two percentage points. The average in Chile of
low birth weight is 5%. These means an
increase from 5% to 7%. Depending on the
size of population, we’re talking about
hundreds or maybe thousands of babies who have not
been born low weight had the earthquake not occurred. So this provides
evidence that I think is somewhat persuasive that
exposure to stress, even before birth, matters. But I’m interested
in inequality. That’s what, that’s
invitation here, to talk about how
inequality matters. And birth weight is a marker
for later disadvantage, but they’re not deterministic
pathways of disadvantage. Or are they? That is the question. And as we know now
from neurobiology and from the social scientists,
the effect of this early shock, or any early shock
for that matter, may persist over time such
that, as Heckman claims, a child who falls behind
may never catch up. So inequality
becomes persistent. To address this question,
we need a panel survey, and we are analyzing the
effect of being exposed to this major stressor in
utero on cognitive ability during childhood. Cognitive ability is a
major, major outcome. It’s a really, really
important determinant of educational attainment
and socioeconomic well being, so it matters in its outcomes. We use for, I suppose there
is no psychologist here, but if we use the Wechsler
Intelligence Test, very, very common. If you have kids,
you know it, too. Say that again. If you have kids,
you know it, too. There you go. There you go. You get this WISC reports. So a variable dimension of
the WISC performance total. This is just to show
that the effects are consistent across the
dimensions of cognitive ability. Let me [INAUDIBLE]. Whatever that measures. [CHUCKLES] Whatever that means. It matters in its outcome,
so it’s good enough for me. The effect is a decline
in the WISC by 0.2. These are standard deviations. 0.2 standard deviations, not
statistically significant. These are the
confidence interval. They touch zero. 0.2 is not altogether
negligible. I mean those of you who
do policy interventions, we find a 0.2 standard
deviation increase, and there is something
to be considered. But it’s not significant. This is, however,
the overall effect. And it’s very
likely, based on all of the evidence
that we have based on the contributions of people
here, that this effect should be sensitive to
inequality, should depend on economic advantage. So the next step
is to assess that. I’m going to divide the
population into three groups, three classes, if you will. Families in which the mom
has less than high school, high school graduates, and
moms with college education. These are my three classes. And when I stratify
the effect by taking into account inequality,
these are the findings. And if you don’t
forget everything I said, and most likely we’re
going to forget part of it, this is what matters. The effect of stress
exposure in utero is very large and negative
on cognitive you abilities, very large and negative
among the poor. This is between half and
2/3 of a standard deviation. This is huge. There’s few interventions
that compare with this. The effect among the so-called
middle class and the wealthy, the other two bars,
is non-existent. This is just one
empirical exercise. We can think of many. What does this tell
us about the question that we’re trying
to address today? I want to offer some
speculation, reflections, somewhat speculative. Well, why is that
this effect very early life course is so stratified? We can rule out many things. I did. I’m not going to show that. It’s not because of
unequal exposure. It’s not that the
poor are more exposed. It’s not because of
unequal reactivity. It’s not because there is
an unequal initial affect on birth weight. The most likely explanation is
the unequal parental responses to initial disadvantage. This is one of the many
reasons why inequality matters in the
reproduction of inequality over the life course. What the qualitative
work, in-depth interviews, is strongly suggesting is
that affluent, or in this case I should say middle class
and affluent families, are successful at compensating. They mobilize large
amount of resources to fully and successfully
compensate for the child’s early disadvantage. Poor families, at the moment
in the context I am analyzing, are not able to do so,
and the disadvantage persists over their
early life course. So let me just try
to use this example to draw conclusions about
the topic we are discussing. First of all, it is not based
on this evidence, the case that a child who falls
behind never catches up, but the ability to
catch up appears to be extremely unequal. One reason why poverty
is so toxic for children is because of the
stresses it involves. I found that disadvantaged
children are not only more likely to be exposed
to stress, this is not this research, we
knew that from before, but also less likely to
catch up through in this case a critical mechanism which
is family compensation. This matters because given
that cognitive ability is so relevant for later outcomes,
educational and beyond, then the implication
of this finding is that intergenerational
reproduction of advantage and disadvantage may be quite
advanced even before the child moves to secondary school. And this suggests that looking
at these processes early in life, as opposed to just
the bivariate correlation between the parents’
standings as adults, children’s standings
as adults is really a new challenge for
those of us interested in the intergenerational
reproduction of inequality. Interestingly, even
we haven’t talked much about the pattern
of inequity, but I think it’s very well know
that the pattern of inequality in Chile is characterized
by concentration at the top. The excess inequalities
in large part driven by the fact that
they rich are too rich. And of course,
inequality is always related to concentration,
but this is rather extreme. But here the difference
that makes the difference is the difference between the
poor, people at the lower end, and the rest. And this is rather
pure speculation. The alteration of this
mechanism for the reproduction of inequality that
this research suggests, which involves family
dynamics, most likely will require rather sustained
and persistent, long-term reduction of inequality. A change of a few
years will likely not alter these deeply
embedded dynamics that result from socioeconomic
constraints and scarcity. So with this I end. Thank you. Thank you very much. [APPLAUSE] So where’s Chico? [SIDE CONVERSATION] So where’s Richard? Richard, do we need to
finish at 12:00 sharp. We shouldn’t go much over. Five minutes. Ten minutes at most. So then let’s be efficient,
and let’s use this time, if you don’t mind, for questions
from the audience who has not been able to speak so far. But I think it was a very
nice panel that gave us very different perspectives. Sociologists, political
science, and economics. And I think it’s a
great opening to what we are going to be discussing
throughout this conference. So anybody wants to jump in? Yes? Julian? I have a very
particular question to the fascinating paper
that Florencia presented. And I wonder
whether you actually have looked at whether
what you are finding has to do with income or not. Whether this has to do
with income [INAUDIBLE] constraints by the poor. Because you were talking about
inequality, but what you really were looking at was
mother’s education. So I wonder what is
the mechanism you think that is driving these results. Is it income? Is it mother not
knowing what to do? Is it relatives that matters? Is it inequality that matters? Or is it [? absent ?]
[? that matters, ?] that people were poor, and they
were not able to cope. Yeah, that’s a really
important question. I’m going to say a couple
of very brief things, and then we can talk some more. The dimensions of disadvantage,
the way I can measure them here in this survey are so highly
correlated that I cannot disentangle very much the
different dimensions that could possibly produce this effect. Based on the interviews,
however, an explanation that really involves all of
them seems to be emerging is that given different
dimensions of disadvantage, constrains, marginality,
lack of access. These things from
surveys and so on, rather than pure human capital
or [INAUDIBLE] constraint, seem to contribute
to a particular type of parental socialization
and parenting style. So I went more beyond the
net effects, if you will. But let me stop there for now. Anybody else? [INAUDIBLE] Well, your paper is
very much related to the front page of the Harvard
Review in the last issue, is that the work by
Sendhil Mullainathan was doing at least, essentially
reaching similar conclusion that scarcity, which is very
much related to poverty, create a lot of stress and makes
it very difficult to adjust. I assume you are
familiar with this work. I can lend you the paper. Thank you. I had the chance to read it. This is extremely
important work, not only because of
the topic, but also because of the pains the authors
take in finding identification strategies that would capture
something close to a cause and effect. However, my speculation
it’s a little different. Their excellent at capturing
the short-term effect of stress. [INAUDIBLE]
I mean, after the crop, before the crop. This changes over the year. And one of the papers,
farmers recover, if you will, after the crop when
there is less scarcity. These findings may suggest
that the effect of inequality and disadvantage may get
not only under your skin, but also under your psyche in
a way that is more long term. And therefore, altering that
may require more time than just, a transfer and then
you see anything. In that sense, I think
they’re complementary. Rebecca? I have two questions. One for Florencia, in the sense
of public policy implications and how with these results,
you go for a policy implication that doesn’t tax parents more. How do you go beyond
individual responsibility, in terms of parenting and
also social responsibility. That’s one question
because that’s a debate. I would like to hear
what there is proposed. And also to you in
terms of the tax reform because I would
like to know if you have this tax, the whole
issue of specific taxes for specific tax. If you rise taxes and you have,
how do you call it in English? [SPEAKING SPANISH] Targeted. Excise. Excise tax. Excise. Exactly. Excise. So that’s it. So you said that that
will help, the tax reform because it will help
also for a public mobilization. But I would like to
hear because there is a huge resistance from the
economic community for that although I believe
you are right. But also that you didn’t
mention that you will also be able to mobilize public
support for tax reform if there is more transparency
and accountability in the expenditure itself. And that was not
part of what I heard, so I would like you
to refer to that, too. [SIDE CONVERSATION] Let me be very brief. I think that’s really
an important question. And really just say
that I understand that a risk is to
say, well, parents are not doing a good job. And let’s push it. It’s a culture. There’s something
wrong in the culture. Anyone who says
that, I would like to invite them to go
to these households and see the lives of people and
see which type of expectation and which type of priority and
strategies can you engage in, given massive levels
of disadvantage. So I think that is a quite
myopic interpretation. A better interpretation
after understanding the lives of people is,
like I said at the end, in order to provide some
minimal resources for families to be able to cope with an
adverse and scarce world, is you have to provide sources
of safety, security, stability and resources,
economic resources, for a rather extended
period of time as opposed to just little things
here and there. But I know that is a little
short, but let me stop here. So I think what you
were asking about was earmarking the
idea that you increase a tax with a specific spending
destination or a goal in mind. So a couple of
things to say, I know a goal is tantamount
like that because it creates budget rigidities. The Chileans have
generally done that in a way that isn’t explicitly
earmarking, because earmarking is not constitutional
in Chile, but they do it in a way that makes
clear that revenue is going to be devoted to that
purpose in a way that’s not necessarily rigidly earmarked. So that’s one way to
perhaps get around that. All of the structures
that I’ve put up certainly have trade offs. So one thing will
get you something. It will give you something
very more negative in a different realm,
whether it’s political or it’s budget rigidity. So it’s all about managing
sets of trade-offs. In terms of accountability
on the expenditure side and the importance
of that, certainly I think that that
is very important. You’re right. To temper that
side of the debate, I think that you will find
in any Latin American country that the economic elites
who are facing taxation are going to say that
that’s a problem, no matter what country you’re in. So if you talk to business
and economic elites in Chile, they’re going to talk about
how inefficient the State is. And they’re going to
say there is corruption and so on and so forth, and
that was 10 years ago as well. Just as if you go Argentina,
the State is ineffective. It’s corrupt. And it’s not transparent. So why should we pay more taxes? So to me, that’s a constant. And the perception
[INAUDIBLE] among the people, especially among
economic elites, that can be an excuse,
indeed perception, but I don’t know
how directly it’s linked to the
actual, what we think of as trying to get
some more objective measure of transparency. But then also in regard
to that, so my focus is on trying to tax economic
elites, not middle class. And so when you do
that, I think you curtail some of
that reaction or are you limit it to a
particular sector that’s likely to be
resisting it, no matter what. But I mean, clearly,
transparency is important. I’m just not sure that there is
a way to measure it effectively or to relate it to
perceptions or to correct it in the short term per se. Does that address? Sergio, [INAUDIBLE]. So this is a [? particular ?]
[? action. ?] So I was surprised as you with the
result of minimum wage. So [? any ?] scenario since
both risk factors are slowdown in economies, in
Brazil in particular, how do you see that it is a
minimum wage playing a role? How important that will
be and what will be. I Know this is very
hard to anticipate, but I was really worried
to see a [? decrease ?] in minimum wage has happened
over in Colombia and so on. It has a negative effect or
a zero effect on equality finding, so I suppose
now the economy was to start slowing down. How you would see the expect
of that particular variable in the future? Yeah, so we were very
surprised as well. I had anticipated
this paper when I started writing it as being
a paper that would bring labor market
institutions to account for a much larger share of
the decline than at the end it turns out of being
[? tremendous. ?] So the way I think
we interpret it, but Julian can answer
this is, I mean, this is where these
Oaxaca-Blinder type things can sometimes be informative. We’ve all done them, and we
all hesitate a little bit. And there are
numbers of problems. But the negative effect,
the effect countervailing inequality, came from
this endowment effect, really pointed to that
increase in the non-compliance. So I think what was happening is
you’re raising, you’re pushing, this policy farther than
the market will take it. In an economy, perhaps like the
United States or some other one where if minimum wages
were ever to rise here. [LAUGHTER] That might lead to unemployment. In an economy like ours,
it leads to informality. So there’s an easy
mechanism to deal with that. But then actually, because
you are pushing the protection beyond where the
market can take, that turns out to
have lead, controlling for everything else, to
an unequalizing effect. When the economy was growing
very fast and it could support, so you had the miracle
years of the 2000 to 2008, unemployment was falling,
formalization was rising, and the economy could
support the minimum wage was rising, than it was
unequalizing like we expect. So I mean I guess I’m just
repeating what I said earlier. But what that leads me to expect
in the future is that if we go through a period when the labor
market cannot support that increase, i.e. low growth and so
on and so forth, then to push, to keep pushing that policy, may
be a mistake because the market may adjust as it
did in the past, which is to increase
informality. So it points in some
sense to something that I guess
economists have always known which is that there is a
limit to what minimum wages can do. It’s not particularly novel. But for a while, because we
were observing all these things all together, we
thought it was hard being a new liberal economist
talking to President Bachelet about raising minimum wages
and reducing inequality. Well, we can. [LAUGHTER] We’re doing it. Well, but there
may be some limits. And I don’t know. I mean, I think that also
desegregating the two time periods is very important. I don’t know your name. Sorry. It’s Miguel [? Pas. ?] Miguel [? Pas. ?] Yes. He was next. I’m going to follow the
strict order of hand raising. Thank you very much. If you didn’t mind. A question for Tasha. That connects also to
what Florencia was saying. In the framework of power
that you were talking about, and the power structures, which
are the counter powers that persist beyond the
cohesion of economic elites when you were in a country
where the economic elites have patronage over the
political elites in a non-transparent way? [INAUDIBLE] or just
trying to make it simpler, how can movements, like the
student movement in Chile, be a force beyond the
momentum that they started and generate a debate that
goes further along when economic elite is so
intertwined with political elite and it sponsors it? So how can it be enough
to break that dynamic? Yeah. Yeah. That’s an excellent question. I hope you got a
good answer for that. I mean, certainly
in Chile, I think we’re seeing a huge impact
of that movement in terms of everything that’s
happening now. So how rare it it that we would
get that kind of dynamic that really opens up the
scope of what’s possible and [INAUDIBLE] finance reform
and everything that we’re seeing now coming on the agenda. I’m not sure. But if you want to think about
Bolivia, perhaps [INAUDIBLE] that I’m most focusing on. Well, yes and no. I mean, you’re always going
to have a lot of this dynamic because there’s so much
wealth in the business sector that’s going to
influence the politics, then it’s going to feed
into the political elites. And there’s going to be
money changing hands there, and so on and so forth. But if you get people
on the bottom who are outside of that elite
[INAUDIBLE] linkages and so forth, really
mobilized to expect that, the political
and the economic elite are threatened in their
ability to stay in power, then you are going to
start seeing changes that wouldn’t otherwise happen. So that was Bolivia leading
up to Morale coming in. I don’t know if
that addresses it, but that’s what comes to mind. It’s a good question. I have three more
hands that I am going to give the opportunity. President Livros first. Then [? Marcelo Medi ?],
and then [? Steven. ?] Could we collect
all the questions, and then give everyone
a last chance, just in the interest
of efficiency. Yes. I think that the
three [INAUDIBLE] makes it in the
sense that they were able to deal with some different
things at the same time, like government, wages,
[INAUDIBLE] of power, and then the issue
about high inequality and closely related with low
intergenerational mobility. And I would like to
address the last one. In what sense, in a sense
that in Chile’s 1990, you have about 200,000
students going to University. In Chile 2011, you have 1M
plus 100,000 in just 31 years. The interest point is this. In 1990, of that 200,000 up
to 20% of incomes in Chile were able to send
their kids 40%. So 40% of the university age
belonging to the upper 20% were able to be
at the University. And the lowest quantile, only
4% were at the University. And that’s perfect
[INAUDIBLE] what you say because if you are the
children of fathers with both of them university
degree, then very often they are going to be in the
upper 20% of the [INAUDIBLE]. What’s happening in 2011? The [INAUDIBLE] are
off [INAUDIBLE]. In 2011, what you
have is 1.1 million. And the 40% of the
[? hike ?] is 30% became 60% of that increases. And when we got to the
[INAUDIBLE] what for, now it was 16%/ It’s still a
tremendous difference. 60% versus 16%. Can you see that? [INAUDIBLE] Would you say
that one increased only 50% but the other [INAUDIBLE]
time, but that’s a joke. It’s simply still behaving
the way that you mentioned. What this important is to
think that in 2011, of those 10 university students,
seven was first generation in their families. So what about the
intergenerational mobility that is taking place at that moment? And those students
that were in the street were not the 200,000 of 1990,
but were the one million. And to what extent
what’s happened was because they were
the children of democracy because, by definition, all
of them were born in 1990. What you had was a
tremendous change because given the
tax system in Chile, they were very happy to
be at the University, extremely happy because
you have to pay for that. And seeing that the solution
was to use some credit, and you pay the credit, then
you have to go to the street. [INAUDIBLE] But what I think what
may be in the future, we forgot the question
of social mobility, when you have the
going from seven until ten, first generation,
and that took place in 20 years. What [INAUDIBLE] to
be about the future? Are you going to get that
intergenerational mobility in such a way, or things are
going to be back to normal? Thank you, Marcelo. [INAUDIBLE], you
raised your hand. I just wanted to know,
in your calculations does the effect of the minimum
wage on the basic pension, is that not counted,
or is that counted? Because of course, that doesn’t
matter what the labor market conditions are because they’re
going to get paid more. And that was an important
part of the Brazil analysis in your book. Do you want to [INAUDIBLE]? I thought there was another. He declined. Oh. Oh. Well, the question that was
posed is extremely relevant. We know that education provides
both the most important avenue for reproduction to the extent
that wealthy parents can afford more and better
education for their children. And at the same time, the
main avenue for mobility and the most amenable to public
intervention for that matter. So it’s crucial. The case of Chile is
interesting because it really emphasizes the
important distinction that analysts need to
make between the expansion and allocation. Chile as many countries in
Latin America, different levels had experienced a massive
expansion in Chile, as you were highlighting,
at the post-secondary level. And that is important
cause that increases the chances for everyone. The question about mobility
is more about, as you said, with your empirical examples of
trends, it’s about allocation. Give that there are more places
at the top, who gets where? And as you indicated,
there has been massive expansion for everyone. The allocation based
on social origins has changed very
little, very little. In addition to that, we have the
dimension that [INAUDIBLE] will be [INAUDIBLE] which is
when quantitative inequality declines, meaning more
people move to the top, two things may happen. First there is a new
top, which it’s not going to happen very soon, but still. Second, there’s
horizontal inequality, and type of institution starts
matter increasingly more. So it’s a question
about quality, where you get within that system. Having said all that,
this is all to say, from certain perspectives,
expansion is very good news. In terms of mobility,
the jury’s still out, and it crucially
depends, of course, on institutional arrangements
and policies that facilitate those transitions for
the most disadvantaged children in society. Before you answer, Steven
[? Speck, ?] can I another one? Mm-hmm. Because rising, an
increasing minimum wage, in addition to setting wages
for pensioners and transfer recipients, et cetera, may
also pull the wages of others because there’s a multiplier
effect at the bottom which, I think, you’d have
to really comment on. Can I add something, Chico? [LAUGHTER] In particular, given the
increase in the minimum wage, why is that informality
decline so substantially? What happened that
made that possible? So three great questions. And Steven’s question,
in this analysis that affect of the minimum
wages on benefits, pensions, and other things is not included
because this is only labor income, so that’s excluded. But what your point
is important is it cautions us to say that
you shouldn’t extrapolate from this to the effect
of minimum wages, or you should be very
careful in extrapolating from this the affect
of minimum wages on the distribution of
household income per capita. Because there, that other
effect would matter. And I think actually it’s an
interesting thing to look at. What happens when you
bring that into effect because there’s
all these non labor incomes that are increasing. And there, there is no
non-compliance issue. Typically, those of us who
have worked on this a lot, because they’ll say there
are much better ways of transferring
money to poor people than through the minimum wages
because of various reasons. But nevertheless, the
empirical question is, does it have
a positive effect? We’re not really taking
that into account. It’s not addressed here. Your question was on? When minimum wages actually
become a puller of wage, even [? the ?] [? advance ?]
[? in the ?] increases. So I keep looking at Marcel
because he was, I think, the original author of
the lighthouse effect. No, it wasn’t [? my job. ?] Unfortunately not. [LAUGHTER] At least in Brazil. In the applications of Brazil. So who was the
lighthouse inventor? Who did the lighthouse effect? He will tell us later,
but his earlier work pointed to the existence of
that in Brazil very strongly. And that of course,
again, is not captured here
directly because we have simply a dummy for
whether you are below or above. Now of course, those
effects are in the data, but they’re not
necessarily captured Ascribed to. by that dummy. So that another
interesting point that one would like to look at.
[INAUDIBLE] start taking notes. There’s a [INAUDIBLE]. And Florencia [INAUDIBLE]. Informality. When informality declines. So in the paper we
do have a little bit of a discussion of
that, [? and though ?] and again, Marcelo would
know better than me. But there was a change in
the technology of enforcement that was quite substantial. So the [INAUDIBLE] government,
the same government that was implementing
all of these things also put a lot more effort behind
the enforcement of contracts within [INAUDIBLE]. So [? these ?] [? guys ?]
would come more often, and so On. And in addition, this
was done not only in the Ministry of Labor, but
there was the strengthening of a Public Prosecutor’s
Office ability to look into those realms. And so there’s a little
bit of a discussion on those institutions,
institutional change and other things. So we don’t have a causal
[? analogy ?] with that, to see whether that
it the sole cause. There may have been
issues on the demand side as well, efficiency,
wages, whatever, during a period of [? growth. ?]
That’s another paper. We [? merely ?]
point to the fact that there are some
institutional changes that are likely to contribute. So lunch time, and let’s
give a hand to our panel. [APPLAUSE]

Author: Kennedi Daugherty

Leave a Reply

Your email address will not be published. Required fields are marked *