Popular in Social Sciences

Books

City Time: On Being Sentenced to Rikers Island

Independent Scholar David A. Campbell

Culture and Society

The City as a Growth Machine: Toward a Political Economy of Place

New York University Harvey Moltoch

Economics

Gender and Economics

University of California Santa Barbara Shelly Lundberg

American Politics

Keeping the Republic

Boston College Marc Landy

Banks and Policy Organisations

Greening the financial sector

European Investment Bank Matteo Ferrazzi

British Politics

Entrepreneurial thinking: the politics and practice of policy impact

University of Sheffield Matthew Flinders

Culture and Society

Carceral Entrapments: Views from the Prison/Street Interface in India

Indian Institute of Technology Delhi Mahuya Bandyopadhya

The unintended consequences of infrastructure development

View more insights in Social Sciences - Economics - Microeconomics

View more figures from Institute for Fiscal Studies

View the page of Antonella Bancalari

Read the Study

Antonella Bancalari discusses the effect of unfinished sewerage infrastructure on early-life mortality in Peru

Image courtesy of interviewee. November 27, 2023

Generating Workspace

Export as PDF

This video video discusses a research paper focusing on the implications of the implementation phase of public goods delivery, particularly in the context of sewage infrastructure projects in Peru from 2005 to 2015. The study highlights that while substantial attention has been given to the planning and completion phases of public infrastructure, the implementation phase—where projects are neither completed nor abandoned—has received less scrutiny.

The author notes that ineffective implementation of sewage projects led to significant health hazards, including increased infant and child mortality rates due to waterborne diseases and construction-related accidents. The analysis shows a rise in infant mortality rates by approximately 0.74 deaths per thousand and under-five mortality rates by 0.16 per thousand during the project's implementation. The author stresses that these findings reveal a dire need for better compliance with health and safety regulations, as well as increased efficiency in completing infrastructure projects to avoid prolonged exposure to hazards.

Despite government investment exceeding $3 billion for these projects, over 40% of initiatives remained halted, posing risks to community health. The paper also emphasizes the necessity for policy makers to enforce stricter health regulations during construction phases to mitigate risks of accidents and diseases, stressing the urgency of completing infrastructure works effectively to improve public health outcomes.

Language:

Export as PDF

X-TIMESTAMP-MAP:LOCAL=00:00:00.000,MPEGTS=900000

00:00.155 --> 00:03.765
This paper relates to the broad literature that,

00:04.625 --> 00:08.125
it has been longstanding understood that effective provision

00:08.145 --> 00:11.925
of public goods plays a crucial role in fostering economic

00:11.925 --> 00:13.845
growth, development

00:13.865 --> 00:17.485
and poverty reduction, mostly underscore by the state

00:18.005 --> 00:19.045
capacity literature.

00:19.785 --> 00:23.005
Um, now, economic analysis of public service

00:23.105 --> 00:26.445
and public good deliveries have largely focused on the

00:26.725 --> 00:30.405
planning and the completion phases, delving, for example,

00:30.755 --> 00:33.445
into resource misallocation issues

00:33.905 --> 00:36.525
or procurement during the planning phase.

00:37.065 --> 00:39.925
Um, and during the post-completion phase,

00:40.245 --> 00:43.445
research has mostly focused on, the effectiveness

00:43.545 --> 00:46.885
of public infrastructure, which is at the core of,

00:46.885 --> 00:50.765
public goods and services in enhancing living standards in

00:50.765 --> 00:52.085
low and middle income countries.

00:52.705 --> 00:56.325
Um, however, there has been comparatively less focus

00:56.865 --> 01:00.085
on the welfare implications of the implementation phase,

01:00.155 --> 01:04.325
this in-between phase when a after projects are planned and

01:04.325 --> 01:05.725
before they are completed.

01:05.945 --> 01:10.925
So, essentially this is what, this paper is, is,

01:10.945 --> 01:12.565
is shifting the focus towards that.

01:13.345 --> 01:16.005
Now, this also links to,

01:16.195 --> 01:19.925
this longstanding literature on trying to understand,

01:20.585 --> 01:23.205
how effective it is for living standards.

01:23.205 --> 01:25.285
The provision of seaward infrastructure,

01:25.695 --> 01:27.525
which has been widely regarded

01:27.585 --> 01:30.205
as the greatest technological advance in public health

01:30.825 --> 01:32.085
and once completed

01:32.085 --> 01:36.205
and in use, this type of infrastructure, has been proven

01:36.385 --> 01:37.485
by previous research

01:37.785 --> 01:41.005
to reduce early life mortality in both advanced

01:41.105 --> 01:42.805
and lower middle income countries.

01:43.545 --> 01:46.965
Um, and now with trillions of dollars estimated

01:46.965 --> 01:50.645
by the World Bank to be destined, during the next years

01:51.185 --> 01:56.165
to basic infrastructure such as sewage, it becomes essential

01:56.665 --> 02:00.925
to maximize the social benefits derived from these projects.

02:01.865 --> 02:04.205
Um, however, the full realization

02:04.305 --> 02:07.405
of these benefits hinges on effective implementation

02:07.985 --> 02:11.805
and a recurrent issue that hasn't been rigorously studied

02:12.705 --> 02:15.205
is the fact that compliance with health

02:15.205 --> 02:17.485
and safety hazards is really low

02:17.505 --> 02:21.685
and really poor, in, in public infrastructure projects.

02:22.185 --> 02:26.805
Um, and the media is plagued by reports of, health

02:26.805 --> 02:29.245
and safety hazards are created accidents

02:29.345 --> 02:33.245
or even indirect effects in public health through,

02:33.435 --> 02:37.405
open ditches that expose the local population to,

02:37.665 --> 02:41.405
ground, breeding sources of, of infection.

02:41.985 --> 02:45.125
And, and that ultimately led to,

02:46.245 --> 02:49.005
increase in, in, in the, in or,

02:49.025 --> 02:51.485
or jeopardizing the disease burden and,

02:51.505 --> 02:54.045
and increase the mortality rates of children.

02:54.265 --> 02:56.405
But this hasn't been, rigorously assessed.

02:56.465 --> 02:58.565
So that's what, this study is,

02:58.745 --> 02:59.845
Is starting to do.

03:00.195 --> 03:02.325
This study is situated in the context

03:02.505 --> 03:04.565
of a nationwide implementation

03:04.625 --> 03:06.645
of sea rich infrastructure in Peru,

03:07.105 --> 03:10.765
and Peru is an emerging apple middle income economy

03:11.435 --> 03:15.605
that has been displaying above worldwide average wellbeing

03:15.665 --> 03:18.765
and very strong macro indicators during the last decades,

03:19.265 --> 03:21.325
but has been performing really weekly in terms

03:21.325 --> 03:24.845
of institutional capacities as, shown

03:24.865 --> 03:26.405
by a report of the OECD.

03:26.985 --> 03:31.085
So the government of Peru invested over 3 billion US dollars

03:31.305 --> 03:35.405
to develop civil rich infrastructure in urban areas as part

03:35.405 --> 03:39.445
of the National Sanitation Plan, which span over the period

03:39.465 --> 03:41.485
of 2005 and 2015.

03:42.385 --> 03:44.085
Now, the effective implementation

03:44.465 --> 03:47.085
of sea rich infrastructure projects as part

03:47.085 --> 03:50.925
of this plan was compromised by significant deficiencies

03:51.185 --> 03:55.845
as it has been shown by reports from former ministers of,

03:55.945 --> 03:57.205
sanitation, housing

03:57.265 --> 04:00.205
and construction, as well as reports from the local,

04:00.305 --> 04:02.405
agents, from from local agents and,

04:02.405 --> 04:05.205
and for example, the local office of the World Bank that,

04:05.345 --> 04:07.485
there was, really poor adherence

04:07.585 --> 04:08.925
to health and safety guidelines.

04:09.015 --> 04:11.125
There was ineffective supervision

04:11.465 --> 04:15.685
and, substandard technical planning, that led to,

04:15.985 --> 04:20.325
tangible hazards such as these unsafe open ditches

04:20.705 --> 04:23.085
and increased health risks,

04:23.745 --> 04:26.245
to the extent in which there were cases

04:26.385 --> 04:29.125
of children drowning in these ditches that got full

04:29.225 --> 04:31.485
of stagnant water,

04:31.945 --> 04:34.685
or also some grounds that, the open,

04:34.795 --> 04:37.645
they just became landfills, and,

04:37.665 --> 04:41.765
and became, breeding ground for vectorial diseases.

04:42.305 --> 04:44.645
In this paper, I assess the impact

04:44.665 --> 04:45.965
of the poor implementation

04:45.985 --> 04:48.205
of sea rich infrastructure on infant

04:48.205 --> 04:49.845
and under five mortality rates.

04:50.465 --> 04:53.565
And these are exactly the outcomes which were expected

04:53.625 --> 04:56.565
to improve, following the, the completion

04:56.565 --> 04:57.645
of such projects.

04:58.265 --> 05:02.285
Um, so to this end, I create a novel link of several sources

05:02.345 --> 05:06.005
of administrative data, vital registries, census

05:06.305 --> 05:07.485
and spatial data,

05:07.985 --> 05:11.925
and that generated a panel data set for every year

05:12.225 --> 05:13.765
and every district municipality,

05:13.775 --> 05:16.285
which is the lowest jurisdictional level in Peru,

05:16.965 --> 05:20.765
encompassing around, 1,400 district municipalities

05:21.265 --> 05:23.645
for every year between 2005 and 2015.

05:24.585 --> 05:27.445
Now, through budgetary reports of these projects,

05:27.645 --> 05:30.965
I identified the timing of the initiation

05:31.065 --> 05:34.565
of the implementation phase across districts,

05:34.865 --> 05:37.365
and essentially this implementation phase is set

05:37.385 --> 05:41.045
as the period when at least one re project remains

05:41.045 --> 05:42.805
incomplete within a district.

05:43.785 --> 05:47.285
Um, now, the estimation strategy here,

05:47.285 --> 05:49.205
the methodology exploits the fact

05:49.205 --> 05:52.565
that the infrastructure rollout across district

05:52.565 --> 05:55.885
municipalities throughout the national territory,

05:55.915 --> 05:57.045
that was pushed by the,

05:57.045 --> 06:01.205
the National Sanitation Plan was mostly based on engineering

06:01.525 --> 06:04.115
considerations, and that I find that

06:04.215 --> 06:06.395
before this, the initiation of this project,

06:06.805 --> 06:09.555
there was no break in the trends in mortality.

06:09.585 --> 06:11.155
There's only a break right

06:11.155 --> 06:13.955
after the start of the construction phase.

06:14.495 --> 06:18.715
So I leveraged this quasi random variation in the timing

06:18.855 --> 06:22.275
of the initiation of the implementation phase,

06:22.775 --> 06:25.835
of seaward projects across different district municipalities

06:25.835 --> 06:27.715
throughout the whole national territory,

06:28.175 --> 06:30.515
and the fact that around one fourth, one quarter

06:30.615 --> 06:32.955
of municipalities were never intervened.

06:32.955 --> 06:36.555
So they can serve, let's say, as a pure control group,

06:36.705 --> 06:39.475
that, that I, that I use in this methodology,

06:40.215 --> 06:43.315
my findings revealed an increase in mortality rates

06:43.415 --> 06:44.755
during the implementation

06:45.015 --> 06:47.715
of seal rich infrastructure in Peruvian district

06:47.715 --> 06:50.755
municipalities between 2005 and 2015.

06:50.775 --> 06:51.915
So that's the study period.

06:52.715 --> 06:56.475
Specifically, I find that infant mortality rates rose

06:57.255 --> 07:01.115
by an average of 0.74 deaths per thousand infants,

07:01.285 --> 07:03.955
while under five mortality rates increased

07:03.955 --> 07:07.115
by zero point 16 deaths per thousand children under five.

07:07.895 --> 07:11.635
Now, these effects translate to 5%

07:11.695 --> 07:14.195
and 4% increase in the,

07:14.195 --> 07:17.115
respectively in the mortality rates when compared

07:17.115 --> 07:19.355
to the pre-construction average, rates.

07:19.815 --> 07:21.275
But these are average effects,

07:21.335 --> 07:23.835
and they seem to be kind of a lower bound,

07:23.985 --> 07:27.515
because when we look at effects of the length of exposure,

07:28.415 --> 07:30.475
the the coefficients intensify.

07:30.655 --> 07:35.035
So, so essentially for the infant mortality rate,

07:35.425 --> 07:37.715
they range between one additional death

07:37.815 --> 07:40.435
to three additional deaths per thousand infant.

07:41.055 --> 07:43.595
And, and for the under five mortality rate, they range

07:43.595 --> 07:44.835
between 0.26

07:44.855 --> 07:49.395
and 0.77 additional rates per thousand under five,

07:49.875 --> 07:52.195
children, which essentially they ident, they,

07:52.195 --> 07:54.075
they intensify over time.

07:54.215 --> 07:56.835
So between the fourth to the seventh year

07:56.835 --> 08:00.195
after the project initiation, the increase is equivalent

08:00.195 --> 08:03.275
to about 13% compared to the mortality rates.

08:03.655 --> 08:06.355
Pre-construction, the average mortality rates across

08:06.835 --> 08:10.195
districts pre-construction, and in the eight to 10 years

08:10.195 --> 08:14.795
after initiation, it surges close to 20% compared

08:14.795 --> 08:17.035
to the mortality rates pre-construction.

08:18.135 --> 08:21.235
Um, now the magnitude of this estimated increase,

08:21.235 --> 08:23.955
it is not worthy, especially considering

08:23.955 --> 08:26.635
that at the beginning of the study,

08:26.815 --> 08:29.875
the average mortality rates stood, for instance, at

08:30.415 --> 08:33.835
1.5 per thousand infants due to malformations.

08:34.495 --> 08:38.315
Um, and for, under five, children,

08:38.495 --> 08:42.315
it stood at 0.06 per thousand children due to cancer,

08:42.445 --> 08:44.995
which is something that we have a lot of research on.

08:45.415 --> 08:49.915
And yet the mortality effects that I find due due to,

08:50.185 --> 08:52.395
this implementation, this bad implementation

08:52.395 --> 08:54.995
of infrastructure project is four times larger

08:55.195 --> 08:58.125
Than what we find, in terms of cancer

08:58.305 --> 09:02.445
for under five, children, at baseline.

09:03.545 --> 09:07.765
Um, now you might be wondering, I gave you a little bit,

09:07.985 --> 09:09.725
of, of flavor of what was happening

09:09.745 --> 09:12.685
and why these projects, post such hazards.

09:12.905 --> 09:16.405
Um, so, so I find that the,

09:17.005 --> 09:19.445
increase in mortality rates are, are mostly driven

09:19.505 --> 09:21.205
by these construction related health

09:21.205 --> 09:22.205
and safety hazards

09:22.205 --> 09:23.765
that I was talking about at the beginning.

09:24.305 --> 09:26.365
So rough, roughly 80%

09:26.465 --> 09:29.645
of the estimated impact can be ascribed

09:29.745 --> 09:34.365
to waterborne mortality, which encompasses deaths due

09:34.365 --> 09:36.965
to infections, parasitic diseases,

09:36.985 --> 09:41.405
and gastrointestinal complications, as well as deaths,

09:41.475 --> 09:44.765
that take place during the perinatal period, which most

09:44.765 --> 09:47.605
of them tend to be due to infectious diseases.

09:48.065 --> 09:50.605
Um, or problems or, or,

09:50.825 --> 09:55.005
or further complications due to, wash related,

09:55.065 --> 09:59.685
so water hygiene and sanitation related, hazards.

10:00.465 --> 10:03.885
Um, now the remaining mortality effect is ascribed

10:03.905 --> 10:07.215
to accidents and external causes,

10:07.355 --> 10:10.975
and these deaths include traffic related incidents,

10:10.975 --> 10:14.895
which is consistent with, local reports of,

10:15.155 --> 10:18.095
really bad signaling in these open ditches

10:18.095 --> 10:21.935
that divert traffic chaotically into previously quiet areas

10:22.065 --> 10:23.855
where children are used to run freely.

10:24.245 --> 10:27.335
There's also, within this category,

10:27.705 --> 10:30.455
there are also deaths caused by false exposure

10:30.555 --> 10:33.655
to mechanical forces, and drowning

10:33.655 --> 10:37.295
and submersion, which is consistent with this,

10:37.585 --> 10:38.735
media,

10:38.735 --> 10:41.935
this newspaper article talking about this really striking

10:42.205 --> 10:45.415
case of children drowning in this open ditches that get full

10:45.415 --> 10:46.855
of, filled with water.

10:47.635 --> 10:51.495
Um, now notably I observe no discernible impact,

10:51.515 --> 10:54.655
so there's no no breaking in the trend of mortality rates

10:54.655 --> 10:58.295
after the initiation of the implementation phase in,

10:58.475 --> 11:01.495
the deaths that are unrelated to construction works,

11:01.725 --> 11:06.675
such as mortality caused by congenital mar malformations

11:06.815 --> 11:11.635
or non-communicable diseases, such as, complications

11:11.655 --> 11:15.115
of other organs, and, and,

11:15.335 --> 11:16.955
and neoplasms, for example.

11:17.655 --> 11:20.555
Um, now this, let's say this placebo test

11:20.555 --> 11:24.355
of looking at other types of mortality actually,

11:24.445 --> 11:25.875
helps in,

11:26.155 --> 11:28.835
bolstering our confidence in the causal interpretation

11:28.895 --> 11:32.235
of the results that, that I give to the main findings.

11:33.015 --> 11:37.435
Um, and finally, I examine heterogeneous effects

11:37.575 --> 11:39.915
by the intensity of construction activity.

11:40.535 --> 11:43.835
And for this, I exploit the fact that districts

11:43.835 --> 11:47.795
with greater geographic suitability for,

11:48.085 --> 11:51.595
local seaward systems started more projects,

11:51.895 --> 11:54.685
but also, completed fewer

11:55.275 --> 11:58.285
resulting in a more intense construction activity

11:58.465 --> 11:59.725
during the study period.

12:00.305 --> 12:03.165
So I leveraged this, leveraged this variation,

12:03.625 --> 12:06.965
and I find that the effect on under five mortality is eight

12:06.965 --> 12:09.125
times greater in districts

12:09.125 --> 12:11.845
where infrastructure activity was more intense.

12:12.545 --> 12:16.205
So my findings serve as a cautionary message,

12:16.465 --> 12:20.045
to the government effectiveness agenda of policy makers

12:20.105 --> 12:23.245
and also within the field of research on state capacity.

12:24.065 --> 12:28.165
Um, they stress the need for, stricter enforcement

12:28.265 --> 12:31.405
of health and safety regulations in public projects

12:31.865 --> 12:33.285
to prevent accidents

12:33.285 --> 12:35.325
and the spreading of diseases,

12:35.325 --> 12:36.685
during the construction phase.

12:37.225 --> 12:40.845
And this can include low hanging fruits such as containment,

12:40.875 --> 12:41.925
ears and fences.

12:41.925 --> 12:44.205
These are things that are fully preventable.

12:44.905 --> 12:48.085
Um, now in line with previous literature,

12:48.365 --> 12:52.005
I also find a significant decline in mortality once all

12:52.005 --> 12:56.245
infrastructure, works were completed in a district,

12:56.795 --> 13:00.165
meaning that we will have to get there quickly, we'll have

13:00.165 --> 13:03.325
to get to the point in which all open digits are closed

13:03.505 --> 13:05.165
as quickly as possible.

13:06.105 --> 13:09.165
Um, so these results also underscored the importance

13:09.165 --> 13:11.565
of avoiding unnecessary halting

13:11.665 --> 13:15.365
and delays that can expose the local populations to hazards

13:15.425 --> 13:16.725
for extended periods.

13:17.465 --> 13:20.445
Now, it's worth noting that in, in the study,

13:20.445 --> 13:24.125
in this study context, only 13% of municipalities

13:24.275 --> 13:27.845
that started projects successfully completed all of them

13:28.025 --> 13:31.165
by the end of the National Sanitation Plan in 2015.

13:32.465 --> 13:37.005
And furthermore, nearly 85% of district municipalities

13:37.115 --> 13:40.125
that initiated sewage projects stole them

13:40.585 --> 13:41.845
for at least one year,

13:42.225 --> 13:45.245
and with an average cost per project lasting approximately

13:45.345 --> 13:46.405
two and a half years.

13:47.225 --> 13:48.845
Um, now districts

13:48.845 --> 13:51.925
that experience this project holds endured the

13:51.925 --> 13:54.685
implementation phase for over double the duration.

13:55.345 --> 13:57.005
Um, and on average,

13:57.175 --> 14:00.285
these halted projects had dispersed 40%

14:00.285 --> 14:02.965
of the contractual sum, meaning that it's not

14:03.165 --> 14:06.085
that they were halted just when the paperwork was initiated

14:06.225 --> 14:08.245
or on the verge of completion, let's say no,

14:08.245 --> 14:10.285
when these ditches were, were, were closed, no,

14:10.285 --> 14:12.925
they were actually halted mid-construction.

14:12.985 --> 14:17.725
So when these ditches are, are open, and, and

14:17.725 --> 14:21.405
moreover, by the last year of the study in 2015,

14:21.795 --> 14:26.205
over 40% of all initiated projects remain halted,

14:26.785 --> 14:28.405
and thus at risk

14:28.425 --> 14:31.245
of abandonment in mid-construction if they are

14:31.265 --> 14:32.405
not restarted.

14:33.345 --> 14:36.845
It is of course, worrying when considering the opportunity

14:37.035 --> 14:39.645
cost, cost of these dispersed funds,

14:40.225 --> 14:42.805
but it is even more distressing the fact

14:42.805 --> 14:47.445
that mortality rates increase with lens of exposure to, to,

14:47.505 --> 14:51.205
to the implementation phase, which was, delayed

14:51.225 --> 14:53.765
and was was expanded more than, than

14:53.765 --> 14:55.325
what it was supposed to.

14:55.325 --> 14:58.165
Given this inefficiencies, such as stalling projects

14:58.545 --> 14:59.805
or just abandoning them.

00:00 - This paper relates to the broad literature that, it has been longstanding understood that effective provision of public goods plays a crucial role in fostering economic growth, development and poverty reduction, mostly underscore by the state capacity literature.

00:19 - now, economic analysis of public service and public good deliveries have largely focused on the planning and the completion phases, delving, for example, into resource misallocation issues or procurement during the planning phase. and during the post-completion phase, research has mostly focused on, the effectiveness of public infrastructure, which is at the core of, public goods and services in enhancing living standards in low and middle income countries.

00:52 - however, there has been comparatively less focus on the welfare implications of the implementation phase, this in-between phase when a after projects are planned and before they are completed. So, essentially this is what, this paper is, is, is shifting the focus towards that. Now, this also links to, this longstanding literature on trying to understand, how effective it is for living standards. The provision of seaward infrastructure,

01:25 - which has been widely regarded as the greatest technological advance in public health and once completed and in use, this type of infrastructure, has been proven by previous research to reduce early life mortality in both advanced and lower middle income countries. and now with trillions of dollars estimated by the World Bank to be destined, during the next years to basic infrastructure such as sewage, it becomes essential

01:56 - to maximize the social benefits derived from these projects. however, the full realization of these benefits hinges on effective implementation and a recurrent issue that hasn't been rigorously studied is the fact that compliance with health and safety hazards is really low and really poor, in, in public infrastructure projects. and the media is plagued by reports of, health and safety hazards are created accidents or even indirect effects in public health through,

02:33 - open ditches that expose the local population to, ground, breeding sources of, of infection. And, and that ultimately led to, increase in, in, in the, in or, or jeopardizing the disease burden and, and increase the mortality rates of children. But this hasn't been, rigorously assessed. So that's what, this study is, Is starting to do. This study is situated in the context

03:02 - of a nationwide implementation of sea rich infrastructure in Peru, and Peru is an emerging apple middle income economy that has been displaying above worldwide average wellbeing and very strong macro indicators during the last decades, but has been performing really weekly in terms of institutional capacities as, shown by a report of the OECD. So the government of Peru invested over 3 billion US dollars to develop civil rich infrastructure in urban areas as part

03:35 - of the National Sanitation Plan, which span over the period of 2005 and 2015. Now, the effective implementation of sea rich infrastructure projects as part of this plan was compromised by significant deficiencies as it has been shown by reports from former ministers of, sanitation, housing and construction, as well as reports from the local, agents, from from local agents and, and for example, the local office of the World Bank that,

04:05 - there was, really poor adherence to health and safety guidelines. There was ineffective supervision and, substandard technical planning, that led to, tangible hazards such as these unsafe open ditches and increased health risks, to the extent in which there were cases of children drowning in these ditches that got full of stagnant water, or also some grounds that, the open,

04:34 - they just became landfills, and, and became, breeding ground for vectorial diseases. In this paper, I assess the impact of the poor implementation of sea rich infrastructure on infant and under five mortality rates. And these are exactly the outcomes which were expected to improve, following the, the completion of such projects. so to this end, I create a novel link of several sources

05:02 - of administrative data, vital registries, census and spatial data, and that generated a panel data set for every year and every district municipality, which is the lowest jurisdictional level in Peru, encompassing around, 1,400 district municipalities for every year between 2005 and 2015. Now, through budgetary reports of these projects, I identified the timing of the initiation of the implementation phase across districts,

05:34 - and essentially this implementation phase is set as the period when at least one re project remains incomplete within a district. now, the estimation strategy here, the methodology exploits the fact that the infrastructure rollout across district municipalities throughout the national territory, that was pushed by the, the National Sanitation Plan was mostly based on engineering considerations, and that I find that

06:04 - before this, the initiation of this project, there was no break in the trends in mortality. There's only a break right after the start of the construction phase. So I leveraged this quasi random variation in the timing of the initiation of the implementation phase, of seaward projects across different district municipalities throughout the whole national territory, and the fact that around one fourth, one quarter of municipalities were never intervened.

06:32 - So they can serve, let's say, as a pure control group, that, that I, that I use in this methodology, my findings revealed an increase in mortality rates during the implementation of seal rich infrastructure in Peruvian district municipalities between 2005 and 2015. So that's the study period. Specifically, I find that infant mortality rates rose by an average of 0.74 deaths per thousand infants, while under five mortality rates increased

07:03 - by zero point 16 deaths per thousand children under five. Now, these effects translate to 5% and 4% increase in the, respectively in the mortality rates when compared to the pre-construction average, rates. But these are average effects, and they seem to be kind of a lower bound, because when we look at effects of the length of exposure, the the coefficients intensify. So, so essentially for the infant mortality rate,

07:35 - they range between one additional death to three additional deaths per thousand infant. And, and for the under five mortality rate, they range between 0.26 and 0.77 additional rates per thousand under five, children, which essentially they ident, they, they intensify over time. So between the fourth to the seventh year after the project initiation, the increase is equivalent to about 13% compared to the mortality rates.

08:03 - Pre-construction, the average mortality rates across districts pre-construction, and in the eight to 10 years after initiation, it surges close to 20% compared to the mortality rates pre-construction. now the magnitude of this estimated increase, it is not worthy, especially considering that at the beginning of the study, the average mortality rates stood, for instance, at 1.5 per thousand infants due to malformations. and for, under five, children,

08:38 - it stood at 0.06 per thousand children due to cancer, which is something that we have a lot of research on. And yet the mortality effects that I find due due to, this implementation, this bad implementation of infrastructure project is four times larger Than what we find, in terms of cancer for under five, children, at baseline. now you might be wondering, I gave you a little bit, of, of flavor of what was happening and why these projects, post such hazards.

09:12 - so, so I find that the, increase in mortality rates are, are mostly driven by these construction related health and safety hazards that I was talking about at the beginning. So rough, roughly 80% of the estimated impact can be ascribed to waterborne mortality, which encompasses deaths due to infections, parasitic diseases, and gastrointestinal complications, as well as deaths,

09:41 - that take place during the perinatal period, which most of them tend to be due to infectious diseases. or problems or, or, or further complications due to, wash related, so water hygiene and sanitation related, hazards. now the remaining mortality effect is ascribed to accidents and external causes, and these deaths include traffic related incidents, which is consistent with, local reports of, really bad signaling in these open ditches

10:18 - that divert traffic chaotically into previously quiet areas where children are used to run freely. There's also, within this category, there are also deaths caused by false exposure to mechanical forces, and drowning and submersion, which is consistent with this, media, this newspaper article talking about this really striking case of children drowning in this open ditches that get full of, filled with water.

10:47 - now notably I observe no discernible impact, so there's no no breaking in the trend of mortality rates after the initiation of the implementation phase in, the deaths that are unrelated to construction works, such as mortality caused by congenital mar malformations or non-communicable diseases, such as, complications of other organs, and, and, and neoplasms, for example. now this, let's say this placebo test of looking at other types of mortality actually,

11:24 - helps in, bolstering our confidence in the causal interpretation of the results that, that I give to the main findings. and finally, I examine heterogeneous effects by the intensity of construction activity. And for this, I exploit the fact that districts with greater geographic suitability for, local seaward systems started more projects, but also, completed fewer resulting in a more intense construction activity

11:58 - during the study period. So I leveraged this, leveraged this variation, and I find that the effect on under five mortality is eight times greater in districts where infrastructure activity was more intense. So my findings serve as a cautionary message, to the government effectiveness agenda of policy makers and also within the field of research on state capacity. they stress the need for, stricter enforcement of health and safety regulations in public projects

12:31 - to prevent accidents and the spreading of diseases, during the construction phase. And this can include low hanging fruits such as containment, ears and fences. These are things that are fully preventable. now in line with previous literature, I also find a significant decline in mortality once all infrastructure, works were completed in a district, meaning that we will have to get there quickly, we'll have

13:00 - to get to the point in which all open digits are closed as quickly as possible. so these results also underscored the importance of avoiding unnecessary halting and delays that can expose the local populations to hazards for extended periods. Now, it's worth noting that in, in the study, in this study context, only 13% of municipalities that started projects successfully completed all of them by the end of the National Sanitation Plan in 2015.

13:32 - And furthermore, nearly 85% of district municipalities that initiated sewage projects stole them for at least one year, and with an average cost per project lasting approximately two and a half years. now districts that experience this project holds endured the implementation phase for over double the duration. and on average, these halted projects had dispersed 40%

14:00 - of the contractual sum, meaning that it's not that they were halted just when the paperwork was initiated or on the verge of completion, let's say no, when these ditches were, were, were closed, no, they were actually halted mid-construction. So when these ditches are, are open, and, and moreover, by the last year of the study in 2015, over 40% of all initiated projects remain halted, and thus at risk of abandonment in mid-construction if they are

14:31 - not restarted. It is of course, worrying when considering the opportunity cost, cost of these dispersed funds, but it is even more distressing the fact that mortality rates increase with lens of exposure to, to, to the implementation phase, which was, delayed and was was expanded more than, than what it was supposed to. Given this inefficiencies, such as stalling projects or just abandoning them.

Faculti AI can make mistakes