In our earlier sessions we talked quite a lot about income and you saw the work that's
been done around modeling different interventions and minimum basic incomes featured quite heavily
in that.
We're now going to hear from Marcia Gibson from University of Glasgow
about what we might learn about the potential effects of a universal basic income from studies
of somewhat kinds of interventions; Marcia.
So thanks everybody for coming,
I'm just gonna say I wasn't sure about the audience here today,
so I've actually tried to make this as close to sort of lay terms as I could so if it's
a bit simplistic for some people I apologise.
But to begin with we will just have a look at what and some of the basic components of
a basic income are.
I'm assuming like almost all of us are aware obviously there's been so much noise around basic
income recently it's really sort of raced up the agenda.
And most people having a reasonable understanding of what a basic income is.
But just to frame the findings and explain some of the methodology, I'm going to go through
some of the kind of key components.
So there are quite a few different definitions of basic income but most of them include these
components,
what would be generally be called a full basic income,
or a true basic income,
should be universal which means it is paid to all citizens of a nation,
regardless of the income, or any other criteria.
There's no form of targeting.
It should be completely free of conditions related to employment status,
work search, activity, or any other behavioral requirement,
and payments should have a fixed value which is not affected by income from of any other
source,
and it should be paid to individuals, and in some definitions but not all,
should be sufficient to cover very basic subsistence costs.
So those who support basic income argue that it would provide economic security for more
quality and social justice and free people to engage in other productive and caring
activities.
And obviously as we've seen amply this morning and the issues with poverty and health inequalities
that we're dealing with in Scotland,
and our awareness of income, as a social determinant, an important social determinants of health,
underlying the kind of potential importance of a basic income.
Another argument that's made in its favour is it could remove the work disincentives
and the intrusive bureaucracy inherent in the existing social security system,
as well as helping people to deal with precarious and unemployment and insecurity.
So the biggest areas of controversy around a basic income are to do with how it would
be funded,
which we had some degree of insight into in Liz's presentation,
that's above my pay grade, so I'm not going to be discussing that.
And the kind of really crucial one is the extent to which people believe it would disincentives
working,
or the kind of widespread belief if people were unconditionally paid enough money to
live on they were just depart from the labour market and sit around filing their nails or whatever
So there's a couple of things that kind of need to be thought about
and when we try to think about how we might understand the effects of a full basic income.
Many studies of basic income have included small samples or samples of people spread
over quite large geographical areas,
but some of the most interesting potential effects of a basic income are only likely
to occur if it's paid to everybody
or if it's paid at least a quite high density of population.
The effects on the wider labour market would be quite different if payments were made universally
because there would be an interaction between any changes in people's labour market activity
and the demand for labour from employers.
But also in terms of community level effects, for instance for one example,
if receiving a basic income reduced stress it would have direct effects on the health
of those who receive it but I could also then feed into reductions in substance abuse,
which would in turn lead to fewer accidents and lower crime.
These would then have impacts on health service use policing and on other members of the community
who would benefit from a fewer accidents and less exposure to crime.
All of this again would have implications for the cost of services,
but it would be very difficult to observe these types of wider effects when the samples
are small or dispersed over wide areas.
In terms of permanence, studies of social interventions are generally quite short-term,
so five years is considered a long study in social research terms,
but again a lot of the effects of a basic income are likely to take a long time to become
apparent and they could change,
increase or decrease over time and also decisions that people make about any changes to their
employment activity might be different if they knew that payments were not time limited,
as opposed to knowing that they're only getting them for two or three years in a study.
So in sum it's really difficult to understand the effects of a full basic income without
implementing it first,
which is a bit of a catch-22 or alternatively is extremely challenging to design an adequate
study
which can capture these types of effects that would appear if the real basic income,
where a full basic income was implemented.
So to try to gain some better understanding of what evidence is available,
and how that kind of informs our understanding of a full basic income, we conducted a scoping
review.
A scoping review is similar to a systematic review but in the sense that it uses systematic
search methods and aims to provide an objective account of the study findings.
But it doesn't include any sort of quality appraisal of the evidence or to a degree it
doesn't include a formal quality appraisal of the evidence.
So we searched for, we didn't use a strict definition of a full basic income outlined
earlier
because if we did we wouldn't have found anything.
So we included interventions that made unconditional, regular cash payments to individuals or households
in middle or high income countries.
And the payments didn't have to be universal and we included interventions where payments
were affected by other income,
as in they were withdrawn when people earned more and we also included interventions that
made payments below subsistence level.
We included studies that reported any outcomes but in our reporting we focused on employment,
education, health, and some social outcomes.
So, in the same way that I described the sample density and the duration of the study could
have an important influence on the intervention effects.
Obviously there are other aspects should also have this kind of influence,
so for instance the value of payments, and whether or not they're withdrawn would affect
and could affect people's labour market responses.
And if it's possible to survive without paid labour people may be less likely to work and
whether
and to what extent payments are withdrawn,
could also influence people's decisions around work.
So we try to account for these factors how these factors might influence the effects
of the interventions in our analysis.
So in terms of the findings we identified 28 studies of 10 interventions,
they were extremely diverse in terms of location, population, design, timing and evaluation
approach.
Just a kind of brief note about transferability or generalisability if there are only one
or two studies on a particular topic
and they're conducted in different contexts it's difficult to know whether similar effects
would be seen in other contexts but when there are a number of studies from widely varying
settings
and the findings are similar I suggest that the findings are likely to be transferable.
So most of the studies that we found are quantitative and most used robust designs that allow us
to and attribute any effects to the intervention
rather than to other changes that occurred at the same time.
Such as randomised controlled trials or quasi-experimental study designs.
And we also found several qualitative studies some of which were well conduct others which
of which were not terribly well reported.
Quite a number of the studies had small samples and and/or a lot of subgroups included in
the analysis which makes it more difficult to interpret the findings.
So I'm going to give a brief overview of the interventions that we found bearing in mind
that none of the interventions had any employment related conditions.
So there were five large studies of negative income tax conducted in the United States
and Canada and then 1970s,
negative income tax provided a guaranteed subsistence level income to low-income families,
which was withdrawn at varying rates in relation to other income.
These studies lasted for three to five years and the samples that they use were very widely
dispersed
except for one saturation site in the Canadian study in which everybody in a small town was
eligible for payments if their income fell below a certain threshold.
The Alaska permanent dividend fund has made payments to all Alaskan residents
from the state's oil revenues since 1982
payments are not effected by other income but they are well below subsistence
however they are substantial at a household level.
The Iran cash transfer has paid all individuals a fixed monthly sum since 2010,
initially this was above subsistence level but this was very quickly eroded by inflation.
At the outset it was intended to be a permanent scheme but survey evidence shows that very
few people actually believed that that was the case,
which they were quite right not to because they're in the process of withdrawing it,
probably as we speak.
Some Native American nations have been running casinos since the mid 1990s and distributing
a portion of the revenues to all tribal members,
payments could be described as universal in the sense that everybody who is a member of the tribe
which is an independent political entity, is eligible for payments.
The value of payments varies across tribes from below subsistence level to well above
but in the studies we included the payments were quite substantial.
In these interventions young people also receive their childhood payments in a very large lump
sum when they reach 18
so that kind of has an influence on how the intervention plays out.
A pilot study of basic income was conducted in India in 2011/12, 20 villages were selected
for the study and all of the individuals resident in those villages received payments for 15
months.
The payments were below subsistence level but not affected by other income.
And finally the Indonesian government made payments to the poorest 30% of the population
to compensate for the abolition of fuel subsidies.
Households received a fixed amount regardless of household size for two periods of around
a year in the 2000s
the payments weren't affected by other income but they didn't cover subsistence
and although they were universal they were available to everybody who made the poverty
criterion,
which was quite a large population.
So although none of the interventions meet all of the criteria for a few basic income
they all meet some of the criteria
and we do have some interventions, which are universal and to all intents and purposes
permanent.
So moving on to the findings of the studies,
specifically are initially looking at the controversial employment issue.
All of the studies reported effects on labour market participation and overall there was
very little impact on labour market activity for most groups.
In the NIT studies in the 70s male breadwinners reduced their total hours by a small amount
but people spending mostly explained this longer between jobs,
finding jobs, which presumably resulted in better job fit at the end of the day.
Second earners again in the NIT studies which in the context of the time means mothers
and single parents had larger reductions in their labour market activity which they used
to spend more time in the home.
In the contemporary studies there's no effect on employment for men or for the general population,
there's an increase for women in Iran but in Alaska women are more likely to work part-time
for reasons which are not quite clear.
In several studies small business owners in fact every studying that reported small business
owners as a separate group,
the small business owners increased the hours that they worked presumably because they had
extra cash to invest in the business.
So whether the intervention was universal or permanent didn't seem to change how people
responded and similarly whether the payments provided enough to live on
or were taken away when the respondents earned money didn't seem to make people any less
likely to work.
So the frequently repeated assertion that people would simply give up work if they were
given money without conditions doesn't appear to be supported by the evidence.
Looking at education, educational outcomes and employment for young people.
Quite a number of the studies reported effects on these outcomes and they all found that
there were large reductions in teenagers working
and in the lower-income countries child labour was also reduced.
Many studies reported positive effects on attending or completing school,
in some cases effects of up to a year of extra education which is a very large effect for
an intervention which isn't actually directly aiming at education.
In the NIT studies it was reported that the time gained from leaving work was used to
stay in school
and the long-term implications of spending longer in education are really quite profound
there's good evidence to show that
the implications of this are really quite profound both an individual and a societal level
So most of the health comes are only reported by a few studies so the evidence isn't as
strong in this area
but there are some large positive effects on some outcomes.
Two studies reported very substantial reductions in low birth weights,
particularly for people who are most at risk,
and again these are reductions of a scale that any intervention that was targeting low
birth weight would be over the moon to get well above that kind of level.
And it seemed there was kind of evidence to suggest that improved diet and greater uptake
of antenatal care were the things that drove these improvements.
There are large improvements in child mental health and personality traits among eastern
Cherokee children,
whose families had received casino dividends.
And further analysis of that study suggests that the improvements are driven by better
parenting and improved mental health.
There was a large reduction in hospital admissions in the Canadians saturation site,
say mostly driven by lower admissions for accidents and mental health problems.
This effect was observed across the whole community although only 30% of the population
received any payments
and qualitative evidence from that say suggests that the NIT reduced the stigma whether associated
with receiving benefits
and allowed people to retain autonomy and dignity.
Very crucially people valued it very much because it actually allowed them to stay in work
when they're when their circumstance changed rather than allowing them to exit work.
However there were other studies that reported no effect for mental health and other health
outcomes so there's a degree of inconsistency around there.
Again there was some evidence of adverse effects two studies reported increased accidental
mortality following receipt of payments,
which were partly linked to increased substance abuse,
however the evidence is that this occurs after payments large payments of any kind.
So after salaries, after social security payments,
whatever people go to the pub basically.
And again one of these studies included young Native Americans who had received these large
lump sums.
So the studies reported effects on a wide range of other outcomes including family and
parenting offending behavior economic outcomes
such as income inequality and business investment and all of these are reported in the full
review which will be available next week.
There is some evidence from the studies of the potential for wider effects
so in Alaska we see that one explanation for the lack of effect on labour market participation
is that consumption increased
which meant that people wanted more things,
more things had to be made so there are more jobs.
Obviously no time to go into this, it's all in the report if you want to see it next week.
Just finally to conclude the evidence doesn't suggest that there are strong effects on labour
market participation
there are some strong positive effects on health and educational outcomes
and there is some evidence of potential wider effects in many areas,
and any future evaluation really needs to be thinking about the cost implications of
these kind of wider effects as they play out across society and over time.
This is the information about the review, which will be here next week
Không có nhận xét nào:
Đăng nhận xét