wow-inequalities/scoping_review.qmd

---
bibliography: 02-data/supplementary/lib.bib
csl: /home/marty/documents/library/utilities/styles/APA-7.csl
papersize: A4
linestretch: 1.5
fontfamily: lmodern
fontsize: "12"
geometry:
    - left=2.2cm
    - right=3.5cm
    - top=2.5cm
    - bottom=2.5cm
lang: en
title: Scoping review on 'what works'
subtitle: Addressing inequalities in the World of Work
filters:
    - src/pandoc-to-zotero-live.lua
zotero:
    library: wow-inequalities
    client: zotero
    csl-style: apa
---

```{python}
#| echo: false
from pathlib import Path

from bibtexparser.model import Field
DATA_DIR=Path("./02-data")
RAW_DATA=DATA_DIR.joinpath("raw")
WORKING_DATA=DATA_DIR.joinpath("intermediate")
PROCESSED_DATA=DATA_DIR.joinpath("processed")
SUPPLEMENTARY_DATA=DATA_DIR.joinpath("supplementary")

## standard imports
from IPython.core.display import Markdown as md
import numpy as np
import pandas as pd
from matplotlib import pyplot as plt
import seaborn as sns
from tabulate import tabulate

sns.set_style("whitegrid")
```

```{python}
#| echo: false
# load and parse overall bibtex sample
import bibtexparser

bib_string=""
for partial_bib in RAW_DATA.glob("**/*.bib"):
    with open(partial_bib) as f:
        bib_string+="\n".join(f.readlines())
sample_raw = bibtexparser.parse_string(bib_string)

bib_string=""
for partial_bib in WORKING_DATA.glob("**/*.bib"):
    with open(partial_bib) as f:
        bib_string+="\n".join(f.readlines())
sample = bibtexparser.parse_string(bib_string)
```

# Introduction

This section will introduce the reader to the concern of inequality in the World of Work (WoW),
and present a discussion on why policy interventions are needed to address these disparities.

# Labour market policies: concepts, functions, typologies and actors

This section will present a typology of policies that directly or indirectly tackle inequalities in the WoW both within the labour market and outside this domain (e.g. education policy).
In order to define the typology of policy areas,
it will be critical to review previous ILO work, in particular de documents outlined by the ToR.
Based on this typology, we will then develop a theory of change to depict policy objectives, components, inputs and functions of distinct types of interventions outlined in the typology.
The section will also identify the theoretical mechanisms and channels through which policies are expected to impact inequalities in forms of work and labour market outcomes.

The ILO has a policy approach to reducing inequalities in the world of work segmented into five major focus areas:
employment creation, access to education, labour rights protection, formalization, gender equality and diversity, and social protection [@ILO2022b].
Each of these areas in turn rests on a variety of more specific emphases which further describe the potential implemented policy measures.

## Policy areas

The ILO has a policy approach to reducing inequalities in the world of work segmented into five major focus areas: employment creation, access to education, labour rights protection, formalization, gender equality and diversity, and social protection.
Each of these areas in turn rests on a variety of more specific emphases which further describe the potential implemented policy measures.
An exemplary typology of general policy area, related specified policy focus and related focus if any can be found in @tbl-policy-areas.

| area of policy           | focus                                              | related                          |
| ---                      | ----                                               | ----                             |
| employment creation      | pro-employment framework                           |                                  |
|                          | gender-transformative framework                    |                                  |
|                          | promotion of business sustainability               | productivity increases           |
|                          |                                                    | reduction in productivity gaps   |
|                          | promotion of digital infrastructure                | technology for decent work       |
|                          |                                                    | reducing digital divide          |
| access to education      | quality of education/training/skills development   | green transition                 |
|                          | relevance of education/training/skills development | digital transition               |
|                          | gender-transformative career guidance              |                                  |
|                          | improvements of public services/social protection  |                                  |
|                          | work-life balance                                  | juggle paid work and family care |
|                          | targeted support for disadvantaged groups          | targeted education               |
| labour rights protection | promotion of rights for all workers                | collective bargaining systems    |
|                          | minimum wage                                       |                                  |
|                          | inclusive labour market institutions               |                                  |
|                          | equal pay for work of equal value                  |                                  |
|                          | wage transparency                                  |                                  |
| formalization            | equality-driven approach to formalization          | gender-responsive                |
|                          | increase decent work in formal economy             | country-tailored                 |
|                          | absorb informal workers / economic units           | comprehensive                    |
|                          |                                                    | non-discriminatory               |
| gender equality          | removal of discriminatory practice                 | removal of stereotypes           |
| diversity                | promotion of equality of treatment                 | removal of discriminatory law    |
|                          | promotion of equality of opportunity               |                                  |
|                          | data collection improvements                       | gender-focus                     |
|                          | occupational gender segregation                    | age-focus                        |
|                          | unequal pay for work of equal value                | disability-focus                 |
|                          | gender-based violence                              | race-focus                       |
|                          | gender-based harassment                            | ethnicity-focus                  |
|                          | gender unequal division of unpaid care work        | migrant status-focus             |
| social protection        | extend reach of social protection schemes          |                                  |
|                          | reach those not adequately protected               |                                  |
|                          | ensure access to social protection                 | comprehensive social protection  |
|                          |                                                    | adequate social protection       |
|                          |                                                    | sustainable social protection    |

: ILO focus areas for inequality reduction {#tbl-policy-areas}

Source: Authors' elaboration based on ILO [-@ILO2022b].

## Existing reviews

Aside from the general typology by the ILO introduced above, there are a variety of differing approaches to the interplay of inequalities and outcomes,
outlined in the following section.

<!-- income, spatial, pre-existing -->
In a multi-disciplinary systematic review of the association between a person's income, their employment and poverty in an urban environment, Perez et al. [-@Perez2022] find that employment plays a significant role in the poverty of urban residents, with primary barriers identified as lack of access to public transport, geographical segregation, labour informality and inadequate human capital.
Many of their investigated barriers can be mapped onto channels of inequality:
gender inequality's impact, through traditional gender roles and lack of empowerment, a lack of childcare possibilities, or unequal proportions of domestic work;
spatial inequality, through residential segregation or discrimination, lack of access to transportation, and a limited access to work;
as well as pre-existing inequalities, here defined as the inter-generational persistence of poverty, larger household sizes, and its possible negative impacts on human capital.
They also identify potential policy interventions to be applied to counteract these inequalities:
credit programs, institutional support for childcare, guaranteed minimum income/universal basic income or the provision of living wages, commuting subsidies, and housing mobility programs.

<!-- gender -->
Zeinali et al. [-@Zeinali2021], in undertaking a systematic review of female leadership in the health-sector in low- and middle-income countries, take an intersectional approach and focus on the main barriers at the intersection of gender and social identity.
Here, they find that the main barriers limiting women's access to career development resources can be reduced access to mentorship and sponsorship opportunities, as well as a reduced recognition, respect, and impression of value at work for women in leadership positions.
The main channels of inequalities entrenching these barriers identified were the increased likelihood for women to take on the 'dual burdens' of professional work and childcare or domestic work, as well as biased views of the effectiveness of men's over women's leadership styles.

<!-- policy interv -->
Looking strictly at the impact of basic income interventions on labour market, health, educational, housing and other outcomes, Pinto et al. [-@Pinto2021] find that, while workforce participation is the primary outcome in most studies, the evaluations have shifted over time to include a wider array of outcomes, perhaps reflecting an understanding of lower health and social care spending offsetting some of the basic income investments.
Most of the studies investigating basic income perspectives focus on advanced economies such as the US.

<!-- gender -->
Finlay [-@Finlay2021] looks at the effects of female women's reproductive health on female labour force participation, especially career advancement, job quality and hours worked, to find a variety of responses differing between low-income, middle-income and high-income countries.
The main findings are that in low-income countries because of the prevalence of informal work, women are forced to adopt individual strategies of balancing child rearing and labour force participation through job type selection, reliance on other women in the household for child care, or birth spacing.
In middle-income countries, women have to juggle child rearing and labour force participation with an overall income inequality; here, early childbearing or lone motherhood especially can perpetuate poverty.
In high-income countries, social protection policies can assist in balancing child rearing and work but many underlying issues of gender inequality remain.
Throughout all countries, childbearing significantly interrupts career advancement.

<!-- gender/pre-existing -->
Chaudhuri et al. [-@Chaudhuri2021] conduct a systematic review to look at coping strategies and the effects of food insecurity, often through poverty, on social and health outcomes for women and children.
They find that one of the primary non-food coping strategies for women is to look for outdoor employment, mostly farm work, which can in turn lead to what the authors argue as *time* poverty when their time for childcare or personal nutrition is now cut short.
This in turn can, in combination with food-based coping strategies such as food rationing (in size or frequency), nutritional switches or food sharing, lead to negative health outcomes for children including disrupted socio-cognitive development as well as coping through dropping out of school, thereby furthering the rift of pre-existing inequalities.

<!-- gender -->
Chang et al. [-@Chang2021] use a qualitative systematic review to look at the linkages of breast-feeding and returning to paid employment for women and identify multiple barriers provided through inequalities discouraging continued breast-feeding after return to employment --- an experience often experienced as physically and emotionally difficult and potentially providing a barrier to full labour force participation.
Aside from individual motivation and support from employers, colleagues, and family members, women highlighted the importance of having workplace legislation in place to facilitate breast-feeding during employment, as well as access to convenient child care.
The review concludes indicating remaining gender and employment inequalities in accessing and receiving the support needed: gender role expectations viewing women as responsible for domestic work or childcare, with shorter maternity leave further discouraging breast-feeding especially of women not in managerial roles.

<!-- disability -->
Undertaking a systematic review to find the effects of brain tumours in individuals on their labour market outcomes, Silvaggi [-@Silvaggi2020] find an impact of neuropsychological functioning on work productivity, issues for their process of returning to work, and often an exit from employment (job loss) for long-term survivors of brain tumours
While the channels are primarily viewed as stemming from the high short-term mortality and depressive symptoms or cognitive deficits, environmental barriers are identified as one channel as well, with the review ending in the policy recommendation of increased vocational rehabilitation for affected persons.

<!-- basic income -->
De Paz-Banez et al. [-@dePaz-Banez2020] use a systematic review of empirical studies to look at the effects of universal basic income on labour supply to find that, with no evidence of significant reductions in labour supply, instead the labour supply would increase globally among adults, men, women, young and old.
The insignificant reductions they found they assumed functional, since they were in the categories of: children, elderly, sick, people with disabilities, women with young children, young people continuing their studies and were offset by the otherwise increased supply.

<!-- disabilities, gender -->
Looking at the impact of gender on the employment outcomes for young disabled adults, Lindsay et al. [-@Lindsay2018] find that while youth with disabilities are half as likely to be employed, gender inequalities may play a compounding role with men being more likely to be in employment than women, working longer hours and having higher wages.
The identified channels here are different social supports, gender role expectations, as well as women's lower job expectations and overprotection from parents or guardians discouraging their independence.

<!-- gender -->
Kumari [-@Kumari2018] looks at the relationship of both economic growth and gender disparity on the labour supply in investigating their effects on female work participation.
<!-- TODO explain U-shape -->
They see a U-shaped participation rate and some evidence of cross-sector gender pay disparity which is affected by demographic factors such as migration, marriage, child care and fertility, as well as economic factors such as per capita income, unemployment, infrastructure and the prevalence of non-farm jobs.
Ultimately, they argue that the labour supply inequalities are based on inequality between the sexes and, while regulatory measures such as adequate family and childcare policies, tax regimes and the presence of subsidized healthcare help, changes to the female labour force participation fundamentally require the replacement of such a traditional value system itself.

<!-- income -->
While undertaking a systematic review concerning the effects of adopting technology on employment in LICs or LMICs, Ugur and Mitra [-@Ugur2017] find when adoption favours product innovation positive effects are somewhat likely.
They also find, however, that existing income inequalities can make the possible positive effects of its adoption more ambiguous and may in turn widen the rift of demand for skilled versus unskilled labour.
Lastly, policies favouring green transition technologies may in turn reduce income inequality, providing another possible linkage.

<!-- disability -->
Lettieri and Diez Villoria [-@Lettieri2017] find that hiding mental illness is one of the primary strategies for improved employment outcomes in a meta-review looking at barriers to labour market inclusion for people mental disabilities.
This act of concealment of identity and self-stigmatization can seem necessary, they argue, due to the channels of workplace prejudices, perceiving them missing skills, as dangerous or unpredictable, or seeing the act of their hiring as charity due to expectations of lower productivity; but also due to discriminatory hiring practices and pre-existing inequalities leading to them being lower-skilled individuals due to prior discrimination, cultural and social barriers to training and work inclusion.
Here, relevant policies include interventions of supported employment (removing an environmental barrier), cognitive behavioural or computer-assisted therapies (cognitive barrier) or vocational rehabilitation programmes (human capital).

<!-- gender -->
Taukobong et al. [-@Taukobong2016] review various dimensions of female empowerment and their effects on a variety of health and development outcomes, including the access and use of financial services for the poor.
They find that, aside from gender inequalities being both highly contextual and intersectional, especially the channels of control over one's income, assets, resources, having decision-making power and individual education affected these outcomes across all dimensions, reflecting their position as channels of gender inequality.
Additionally, personal mobility, safety and equitable interpersonal relationships are associated with some health and family planning outcomes.
Ultimately, the review shows that due to the contextual nature, interventions need to identify the variations of inequality at their start, see where inequalities exist, overlap and work as barriers for an effective implementation.

<!-- disability -->
Ruhindwa et al. [-@Ruhindwa2016] review a variety of barriers to adequate workforce inclusion for people with disabilities, proposing an inclusive approach in which the individual is given space to take ownership of the solutions addressing challenges experienced in the employment sector.
Similarly, they view hiring discrimination and workplace stigmatization as the largest channels through which inequalities of disability manifest themselves.
They see especially employment support practices, with focus on enabling this, as relevant policy strategies, as well as national campaigns to ease disclosing one's disability in the labour market.

<!-- disability, gender, age -->
In looking at the various dimensions affecting the labour market outcomes of supported employment interventions for people with disabilities, Kirsh [-@Kirsh2016] finds that most literature still only regards the overall efficacy of the interventions without taking into account compounding intersectional characteristics.
They find that generally men are more likely to find employment through the intervention, possibly resting on current programmes focus on manual labour, as well as younger people generally finding better employment.
This highlights the intersectional nature of inequalities between disability, gender and age.
One relevant policy they see is that of vocational rehabilitation.

<!-- disability -->
Hastbacka et al. [-@Hastbacka2016] undertake a scoping review to find the linkages between societal participation and people with disabilities, looking at specific interventions for the identity of participants, types of participation analysed, and channels of effect.
They see most literature focusing on labour market participation and viewing disabled people as coherent group instead of intersectional.
The main channels of inequality providing barriers they identify are financial factors, attitudes of discrimination, health issues and unemployment, while the main driving mechanisms identified are legislation and disability policies, as well as support from people in close contact with disabled people and attitudes in society and the hiring process.

<!-- disability -->
In a systematic review looking at the effectiveness of workplace accommodations on employment and return to work, Nevala et al. [-@Nevala2015] find few studies with rigorous design leading to conclusive evidence.
They do find moderate evidence that employment in disability can be increased through workplace accommodations such as vocational counselling or guidance, education, self-advocacy, positive perception and help by others.
There is also low evidence for return to work being increased by education, work aids and techniques and cooperation between employers and other professionals (such as occupational health care, or service providers).

## The world of work

The policy areas and their respective focus perspectives are based in the conceptual understanding of the world of work, following the definition of work being activities performed by persons of any sex and age producing goods or providing services for "economic units [which] can be allocated mutually exclusively to one of the following sectors:" the formal sector, the informal sector, or the community and household own-use sector [@ILO2023c, 6].
This is the broader understanding of work which specifically separates itself from a more narrow conception of those in employment who are engaging in "production for pay or profit", whether for the informal or the formal market economy [see especially @ILO2023c, Point 18ff].
The key differentiations for these concepts are founded on an understanding of the production of goods or provision of services, as well as the distinctions between use by others for ultimate own-use and that of working for pay and/or profit – that is, as part of a market transaction in exchange for remuneration or in the form of profits derived from the goods or services.

Whether these services or goods are produced in what is defined as the informal economy, the formal economy or under informal employment outside the informal sector is, for the general encapsulation of no importance – they occur in the world of work.
Here, conceptually, it should be captured under one of the five mutually exclusive forms of work [@ILO2023, p. 4, Point 7c] to be understood as: own-use production work, performing "any activity to produce goods or provide services for own final use" [@ILO2013, p.5]; employment work comprising those performing work for others in exchange for pay or profit introduced above; unpaid trainee work, performing "any unpaid activity to produce goods or provide services […] to acquire workplace experience or skills" [@ILO2013, p.7]; and volunteer work, that being "any unpaid, non-compulsory activity to produce goods or provide services for others" [@ILO2013, p.8].

Any activity falling under work as defined above on the one hand, but not under any of these forms of work on the other, is instead designated as other work activities in the following considerations. The key concepts between these categories come down to a varying intensity of participation, the distinction of working for pay and/or profit mentioned above, whether it is for ultimate own-use or the use by others, and its compulsory nature.

## Inequalities in the world of work

Inequalities in the world of work have to be fundamentally conceptualized along two axes: On the one hand, vertical inequality captures the "income inequality between all households in a country" [@ILO2021].
Measurements of vertical inequalities is a perspective which focuses primarily on incomes as data, with debate of top income percentiles versus the remaining body of people often posing the primary area of debate [@ILO2021a].
Horizontal inequalities, on the other hand, occur when "some groups within the population find themselves disadvantaged and discriminated against on the basis of their visible identity, for example their gender, colour or beliefs, among others" [@ILO2021a].

Importantly, these inequalities do not act in a vacuum but create an interplay through overlaps and accumulations which take on their own driving dynamics for people belonging to multiple disadvantaged groups, captured in the idea of inequality’s intersectionality [@ILO2022b].
Here, especially horizontal inequalities may be hard to disentangle for impact finding, an important aspect of effective rigorous analysis in quantitative studies.

Thus, for a study on inequalities, or in turn a study on policies aimed at reducing inequalities in the world of work to be one of rigorous analysis, it must clearly define the type of policy taken as its object of analysis (its independent variable) as well as the types of inequalities targeted for reduction through the respective policy and measured as channels of impact.
Ultimately, then, the individual outcome measures need to be clearly specified and disentangled, most clearly reflecting in labour market outcome measures (dependent variables).
Only then can the targeted inequality be delineated as a clear channel.

In targeting an increase in equality, there are then two approaches to take: either levelling the playing field so that characteristics beyond an individual’s control can not influence their future perspectives, nor limit the potential of the powers they possess, through achieving equality of opportunity; or strive for an equality of outcomes, in factual observed resulting (in-)equalities.
As the ILO established, such a focus on equality of outcomes can be of great importance since "high levels of inequality today tend to reduce social mobility tomorrow" [@ILO2021a], making it that much more difficult to ultimately ensure equality of opportunity for following generation.
The key concepts here are thus the distinction of within-group and between-group inequalities, their overlapping characteristics, as well as policies enabling an equality of opportunity or of outcome.

Income inequality is still the primary lens of inequality that many approaches target, as well as the main focus point of many inequality measurements such as the Gini coefficient or ratios such as the Palma ratio [@DFI2023].
Following the ILO, "labour income is the main source of income for most households in the world [thus] unequal access to work and working poverty are major drivers of inequalities" [@ILO2021].
Income inequality, here, can be affected by a wide set of factors: status in employment, forms of work, the sector of activity, the respective occupation, type of enterprise, type of contract for those in waged work, and the status of formality among others [@ILO2019].
Income inequality should also not be seen as separate from, nor standing above, other inequalities, but closely linked to other inequalities.
As the ILO states, "income inequality, inequality of employment outcomes more generally and inequality of opportunities are intimately related" [@ILO2022b].
At the same time the exact linkages of these factors remain under-analysed, which is the reason why the channels of inequalities and the policies to reduce them will pose a fruitful space of analysis for this research.

While income inequality holds a primary position of importance for many analyses from a perspective of quantity, it should not be understood as carrying more importance qualitatively for itself compared to other inequalities but rather be understood "like a prism, which reveals many other forms of inequality, including those generated in the world of work" [@ILO2021a, p. 13].
It is the primary measure of vertical inequality, however, with other inequalities describing primarily the concept of horizontal inequality.

Here, of primary focus for the ILO, and many studies on inequality in the world of work, is gender inequality.
It describes the inequalities that arise because of an individual’s gender.
Generally, while the type and extend of other inequalities does vary substantially by global location and country, "gender inequalities, despite some progress over the past decades, remain persistent and pervasive" [@ILO2021].

Following a report on the gendered make-up of work globally, women are making up a larger part of those in underemployment, they primarily make up the service sector – a rising trend – while suffering a persistently substantial wage gap, tend to work shorter hours in employment but in turn have longer working days when including unpaid work, as well as contributing disproportionally to family work [@ILO2016].
The domestic area of work is also dominated by women, who make up 76.2 per cent of it, in addition to domestic work being overwhelmingly informal labour globally [@ILO2023a].

These inequalities in the world of work in turn also reflect in women being hindered in accessing adequate education, training, as well as the possibility for lifelong learning, and furthermore access to quality jobs, housing, mobility, capital, land, and adequate social protection – disparities which, on the basis of deeply rooted inequalities of gender roles, education and places of residence remain largely static if not on the rise.
These channels and outcomes, viewed intersectionally, must thus represent the primary lens through which to disentangle the gender inequality in the world of work today.

There are additional socio-demographic inequalities beyond gender which are based on the innate, most often visible, identification of a person.
These are made up of, though not limited to, ethnic and racial inequalities, those based on religion and beliefs, based on a person’s status as a migrant, a person’s age, sex, or disabilities [@ILO2021a].
For example, young people generally fare significantly worse in labour markets shown through outcomes such as a higher incidence of temporary employment throughout youth and the young labour force [@ILO2023b; @ILO2019].

As a report on the global conditions of work established, over "7% of workers felt they had been discriminated against in the 12 months prior to the survey on grounds of sex, race, religion, age, nationality, disability or sexual orientation" [@ILO2019] in the EU alone, making socio-demographic inequalities a prevalent and important to tackle angle of horizontal inequality.
Here, it will be especially important to correctly disentangle individual sources or contributing characteristics from each other in finding their linkages to relevant outcomes.

Another form of inequality are spatial inequalities, those that arise because of an individual’s location relative to other.
These inequalities exist primarily between different regions of a country: those between urbanity and rurality or more peripheral areas, but also between richer and poorer regions and, as the ILO established, can even lead to a ‘growing sense of fractured societies’ [@ILO2021].

One of the channels this can manifest itself is through an unequal access to decent work opportunities or economic opportunities more generally, an unequal access to financial resources, quality public services or even overall access to an essential social service infrastructure and digital infrastructure, as well as quality access to education or relevant training.
For spatial inequalities it will be especially important to take note of locally bound differences versus more generalizable results, with the dimensions and contributing factors to its outcomes potentially varying widely between different geographies and national contexts.

In mentioning unequal access to quality education or public infrastructure another important dimension of inequalities becomes highlighted: the dimension of pre-existing inequalities, that is, inequalities which exist prior to an individual’s interaction with the labour market and, though closely intertwined with socio-demographic inequalities, will prove useful to analytically differentiate between.
A differentiation which becomes especially important with a view on the inter-generational effects of inequality, highlighted in recognizing the difference between equality of opportunity and outcome.
The level of education, an individual’s poverty, productivity on the labour market and similar inequalities in opportunities are often the result of long-running pre-existing inequalities such as unequal access to health services, education, lacking property rights or clear ownership of assets, the lack of formal recognition as an individual, no access to formal banking [@ILO2021a].
Understanding such channels becomes difficult if not taking pre-existing inequalities into account as a separate category of inequality and long-term impacting channel.

Addressing these inequalities, in turn, is just as important to reducing inequalities within the labour market (as well as beyond) since they do play such a role for intergenerational social mobility and their impacts can be seen, once again, reflecting in the prism of subsequent income inequality.
For pre-existing inequalities, it will be especially important to understand the often delayed and more opaque nature of the roots of many outcomes, with channel being more difficult to identify and clearly label – especially in an intersectional context.
These five dimensions of inequalities – income inequality, gender inequality, socio-demographic inequality, spatial inequality and pre-existing inequalities – will thus provide the categorical anchors along which the reviewed studies will be analysed for their policy effects, each with a slightly different focus in linkages between inequality, policy and outcome.

# The search protocol

This section will discuss the systematic scoping review methodology that is proposed to conduct the review of the literature on policy interventions that are expected to address inequalities in forms of work and labour market outcomes.
Unlike purely systematic reviews which typically focus on specific policy questions and interventions, systematic scoping reviews focus on a wider spectrum of policies, where different study designs and research questions can be investigated.
Since scoping reviews allow both broad and in-depth analyses, they are the most appropriate rigorous method to make a synthesis of the current evidence in this area [@Arksey2005].

The scoping review allows broad focus to be given to a subject for which no unified path with clear edges has been laid out yet by prior reviews, as remains the case with policies targeting inequalities in the world of work.
It does so through a breadth-first approach through a search protocol which favours working through a large body of literature to subsequently move toward a depth-favouring approach once the literature has been sufficiently delimited.
Its purpose, clearly mapping a body of literature on a (broad) topic area, is thereby useful on its own or in combination with a systematic approach [@Arksey2005].
With an increasingly adopted approach in recent years, with rigorous dichotomy of inclusion and exclusion criteria it provides a way of charting the relevance of literature related to its overall body that strives to be free of influencing biases which could affect the skew of the resulting literature sample [@Pham2014].

<!-- TODO need correct above definitions -->
The search protocol will be carried out based on the introduced areas of policies as well as the possible combination of definitions and outcomes in the WoW.
For each dimension of definitions, a cluster containing possible utilized terms will be created, that is for: definitions of work and labour, forms of work, definitions of inequality, forms of vertical and forms of horizontal inequalities, labour market outcomes, and definitions of policy.
Each of the clusters contains synonymous terms as well as term-adjacent phrase combinations which are in turn used to refine or broaden the search scope to best encapsulate each respective cluster, based on the above definitions.

<!-- TODO Why WOS database? -->
The search protocol then follows a three-staged process of execution: identification, screening and extraction.
First, in identification, the above categorizations are combined through Boolean operators to conduct a search through the database repository Web of Science.
While the resulting study pools could be screened for in multiple languages, the search queries themselves are passed to the databases in English-language only.
<!-- TODO will we be using gray lit? -->
Relevant results are then complemented through the adoption of a 'snowballing' technique, which analyses an array of published reviews for their reference lists to find cross-references of potentially missing literature.

To identify potential studies and create an initial sample, relevant terms for the clusters of world of work, inequality and policy interventions have been extracted from the existing reviews as well as the ILO definitions.
Identified terms comprising the world of work can be found in @tbl-wow-terms,
with the search query requiring a term from the general column and one other column.

```{python}
#| label: tbl-wow-terms
#| tbl-cap: World of work term cluster
terms_wow = pd.read_csv("02-data/supplementary/terms_wow.csv")
md(tabulate(terms_wow.fillna(""), showindex=False, headers="keys", tablefmt="grid"))
```

The world of work cluster, like the inequality and policy intervention clusters below, is made up of a general signifier (such as "work", "inequality" or "intervention") which has to be labelled in a study to form part of the sample,
as well as any additional terms looking into one or multiple specific dimensions or categories of these signifiers (such as "domestic" work, "gender" inequality, "maternity leave" intervention).
At least one general term and at least one additional term have to be mentioned by a study to be identified for the initial sample pool.

For the policy intervention cluster, a variety of terms have been identified both from the ILO policy areas and guidelines as well as existing reviews, as can be seen in @tbl-intervention-terms.
Where terms have been identified from previous reviews outside the introduced ILO policy guidelines,
there source has been included in the table.
For the database query, a single term from the general category is required to be included in addition to one term from *any* of the remaining categories.

```{python}
#| label: tbl-intervention-terms
#| tbl-cap: Policy intervention term cluster
terms_policy = pd.read_csv("02-data/supplementary/terms_policy.csv")
# different headers to include 'social norms'
headers = ["General", "Institutional", "Structural", "Agency & social norms"]
md(tabulate(terms_policy.fillna(""), showindex=False, headers=headers, tablefmt="grid"))
```

Lastly, the inequality cluster is once again made up of a general term describing inequality which has to form part of the query results, as well as at least one term describing a specific vertical or horizontal inequality,
as seen in @tbl-inequality-terms.

```{python}
#| label: tbl-inequality-terms
#| tbl-cap: Inequality term cluster
terms_inequality = pd.read_csv("02-data/supplementary/terms_inequality.csv")
md(tabulate(terms_inequality.fillna(""), showindex=False, headers="keys", tablefmt="grid"))
```

A general as well as category-specific term from each cluster will be required, using a intersection merge (Boolean 'AND'),
as well as in turn a single of those from each of the three clusters using an intersection merge.
The resulting sample pool will thus include a term and specific dimension of inequality and of policy intervention within the world of work.

Second, in screening, duplicate results are removed and the resulting literature sample is sorted based on a variety of excluding characteristics based on: language, title, abstract, full text and literature supersession through newer publications.
Properties in these characteristics are used to assess an individual study on its suitability for further review.

Narrowing criteria are applied to restrict the sample to studies looking at i) the effects of individual evidence-based policy measures or intervention initiatives ii) attempting to address a single or multiple of the defined inequalities in the world of work.
iii) using appropriate quantitative methods to examine the links of intervention and impact on the given inequalities.
The narrowing process makes use of the typology of inequalities, of forms of work, and of policy areas introduced above as its criteria.

An overview of the respective criteria used for inclusion or exclusion can be found in @tbl-inclusion-criteria.
It restricts studies to those that comprise primary research published after 2000,
with a focus on the narrowing criteria specified in @tbl-inclusion-criteria.

```{python}
#| echo: false
#| label: tbl-inclusion-criteria
#| tbl-cap: Study inclusion and exclusion scoping criteria {#tbl-inclusion-criteria}

inclusion_criteria = pd.read_csv("02-data/supplementary/inclusion-criteria.tsv", sep="\t")
md(tabulate(inclusion_criteria, showindex=False, headers="keys", tablefmt="grid"))
```

To facilitate the screening process, with the help of 'Zotero' reference manager a system of keywords is used to tag individual studies in the sample with their reason for exclusion,such as ‘excluded::language’, ‘excluded::title’, ‘excluded::abstract’, and ‘excluded::superseded’.
This keyword-based system is equally used to further categorize the sample studies that do not fall into exclusion criteria, based on primary country of analysis, world region, as well as income level classification.
To that end, a ‘country::’, ‘region::’ and ‘income::’ are used to disambiguate between the respective characteristics, such as ‘region::LAC’ for Latin America and the Caribbean, ‘region::SSA’ for Sub-Saharan Africa; as well as for example ‘income::low-middle’, ‘income::upper-middle’ or ‘income::high’.
These two delineations follow the ILO categorizations on world regions and the country income classifications based on World Bank income groupings [@ILO2022].

Similarly, if a specific type of inequality, or a specific intervention, represents the focus of a study, these will be reflected in the same keyword system, through for example ‘inequality::income’ or ‘inequality::gender’.
The complete process of identification and screening is undertaken with the help of the Zotero reference manager, ultimately leaving only publications which are relevant for final full-text review and analysis.
Last, for extraction, studies are screened for their full-texts, irrelevant studies excluded with ‘excluded::full-text’ as explained above and relevant studies then ingested into the final sample pool.

Should any literature reviews be identified as relevant during this screening process,
they will in turn be crawled for cited sources in a 'snowballing' process,
and the sources will be added to the sample to undergo the same screening process explained above.

```{python}
#| echo: false
#| output: asis

sample_out_title = []
sample_out_abstract = []
sample_out_fulltext = []
sample_out_language = []
sample_relvant_done = []

for e in sample.entries:
    if "keywords" in e.fields_dict.keys():
        if "out::title" in e["keywords"]:
            sample_out_title.append(e)
        elif "out::abstract" in e["keywords"]:
            sample_out_abstract.append(e)
        elif "out::full-text" in e["keywords"]:
            sample_out_fulltext.append(e)
        elif "done::extracted" in e["keywords"] and "relevant" in e["keywords"]:
            sample_relvant_done.append(e)

t3 = "`" * 3
# FIXME use 02-data/supplementary undeduplciated counts to get database starting and snowballing counts
# from: https://github.com/quarto-dev/quarto-cli/discussions/6508
print(f"""
```{{mermaid}}
%%| label: fig-prisma
%%| fig-cap: "Sample sorting process through identification and screening"
flowchart TD;
    search_db["Records identified through database searching (n=1643)"] --> starting_sample;
    search_prev["Records identified through other sources (n=753)"] --> starting_sample["Starting sample (n=2396)"];

    starting_sample -- "Duplicate removal ({2396 - len(sample.entries)} removed) "--> dedup["Records after duplicates removed (n={len(sample.entries)})"];

    dedup -- "Title screening ({len(sample_out_title)} excluded)" --> title_screened["Records after titles screened (n={len(sample.entries)-len(sample_out_title)})"];

    title_screened -- "Abstract screening ({len(sample_out_abstract)} excluded)"--> abstract_screened["Records after abstracts screened (n={len(sample.entries)-len(sample_out_title)-len(sample_out_abstract)}"];

    abstract_screened -- "  Language screening ({len(sample_out_language)} excluded)  "--> language_screened["Records after language screened (n={len(sample.entries)-len(sample_out_title)-len(sample_out_abstract)-len(sample_out_language)})"];

    language_screened -- "  Full-text screening ({len(sample_out_fulltext)} excluded)  "--> full-text_screened["Full-text articles assessed for eligibility (n={len(sample_relvant_done)})"];
{t3}
""")
```

All relevant data concerning both their major findings and statistical significance are then extracted from the individual studies into a collective results matrix.
The results to be identified in the matrix include a study’s: i) key outcome measures (dependent variables), ii) main findings, iii) main policy interventions (independent variables), iv) study design and sample size, v) dataset and methods of evaluation, vi) direction of relation and level of representativeness, vii) level of statistical significance, viii) main limitations.

## Description of results

```{python}
#| echo: false

sample_size_all = len(sample_raw.entries)

sample_relevant = []
for e in sample.entries:
    if "keywords" in e.fields_dict.keys() and "relevant" in e["keywords"]:
        sample_relevant.append(e)

md(f"""
The query execution results in an initial sample of {sample_size_all} potential studies after the identification process.
This accounts for all identified studies without duplicate removal, without controlling for literature that has been superseded or applying any other screening criteria.
Of these, {len(sample_relevant)} have been identified as relevant studies for the purposes of this scoping review.
""")
```

The currently identified literature rises almost continuously in volume,
with small decreases between 2001 and 2008, as well as more significant ones in 2012 and 2016,
as can be seen in @fig-publications-per-year.
Keeping in mind that these results are not yet screened for their full relevance to the topic at hand, so far only being *potentially* relevant in falling into the requirements of the search pattern, an increased results output does not necessarily mean a clearly rising amount of relevant literature.

```{python}
# load relevant studies
from src import data
bib_df = data.from_yml(f"{PROCESSED_DATA}/relevant")

# load zotero-based metadata
reformatted = []
for e in sample_relevant:
    ed = e.fields_dict
    reformatted.append([
                        ed.get("doi", Field(key="doi", value=None)).value,
                        ed.get("times-cited", Field(key="times-cited", value=None)).value,
                        ed.get("usage-count-since-2013", Field(key="usage-count-since-2013", value=None)).value,
                        ed.get("keywords", Field(key="keywords", value=None)).value,
                        ])
zot_df = pd.DataFrame(reformatted, columns = ["doi", "cited", "usage", "keywords"])

bib_df["doi"] = bib_df["uri"].str.extract(r"https?://(?:dx\.)?doi\.org/(.*)", expand=False)
bib_df["zot_cited"] = bib_df["doi"].map(zot_df.drop_duplicates("doi").set_index("doi")["cited"])
bib_df["zot_usage"] = bib_df["doi"].map(zot_df.drop_duplicates("doi").set_index("doi")["usage"])
bib_df["zot_keywords"] = bib_df["doi"].map(zot_df.drop_duplicates("doi").set_index("doi")["keywords"])

bib_df["date"] = pd.to_datetime(bib_df["year"], format="%Y")
bib_df["year"] = bib_df["date"].dt.year

# only keep newer entries
bib_df = bib_df[bib_df["year"] >= 2000]

# Add WB country grouping definitions (income group, world region)
# TODO Re-enable for processed study pool
# WB_COUNTRY_GROUPS_FILE = Path(f"{SUPPLEMENTARY_DATA}/wb-country-groupings.xlsx").resolve()
# df_country_groups = pd.read_excel(WB_COUNTRY_GROUPS_FILE)
# bib_df["income group"] = bib_df["country"].map(df_country_groups.set_index("Economy")["Income group"])
# bib_df["region"] = bib_df["country"].map(df_country_groups.set_index("Economy")["Region"])
```

```{python}
#| label: fig-publications-per-year
#| fig-cap: Publications per year

# create dummy category for white or gray lit type (based on 'article' appearing in type)
bib_df["pubtype"].value_counts()
bib_df["literature"] = np.where(bib_df["pubtype"].str.contains("article", case=False, regex=False), "white", "gray")
bib_df["literature"] = bib_df["literature"].astype("category")

# plot by year, distinguished by literature type
ax = sns.countplot(bib_df, x="year", hue="literature")
ax.tick_params(axis='x', rotation=45)
# ax.set_xlabel("")
plt.tight_layout()
plt.show()
```

Anomalies such as the relatively significant dips in output in 2016 and 2012 become especially interesting against the strong later increase of output.
While this can mean a decreased interest or different focus points within academia during those time spans,
it may also point towards missing alternative term clusters that are newly arising, or a re-focus towards different interventions, and should thus be kept in mind for future scoping efforts.

Looking at the distribution between white and gray literature a strong difference with white literature clearly overtaking gray literature can be seen, a gap which should not be surprising since our database query efforts are primarily aimed at finding the most current versions of white literature.
The gap will perhaps shrink once the snowballing process is fully completed,
though it should remain clearly visible during the entire scoping process as a sign of a well targeted identification step.

@fig-citations-per-year-avg shows the average number of citations for all studies published within an individual year.
From the current un-screened literature sample, several patterns become visible:
First, in general, citation counts are slightly decreasing - as should generally be expected with newer publications as less time has passed allowing either their contents be dissected and distributed or any repeat citations having taken place.

```{python}
#| label: fig-citations-per-year-avg
#| fig-cap: Average citations per year
bib_df["zot_cited"] = bib_df["zot_cited"].dropna().astype("int")
grpd = bib_df.groupby(["year"], as_index=False)["zot_cited"].mean()
ax = sns.barplot(grpd, x="year", y="zot_cited")
ax.tick_params(axis='x', rotation=45)
plt.tight_layout()
plt.show()
```

Second, while such a decrease is visible in relatively recent years (especially 2019--2023), it is not a linear decrease throughout but rather a more erratically stable citation output.
This points to, first, no decrease in academic interest in the topic over this period of time,
second, no linearly developing concentration or centralization of knowledge output and dissemination,
and third potentially no clear-cut increase of *relevant* output over time either.

Lastly, several years such as 2001, 2002, 2005 and 2008 are clear outliers in their large amount of average citations which can point to one of several things:

It can point to clusters of relevant literature feeding wider dissemination or cross-disciplinary interest, a possible sign of still somewhat unfocused research production which does not approach from a single coherent perspective yet.
It can also point to a centralization of knowledge production, with studies feeding more intensely off each other during the review process, a possible sign of more focused knowledge production and thus valuable to more closely review during the screening process.

Or it may mean that clearly influential studies have been produced during those years, a possibility which may be more relevant during the early years (2000-2008).
This is because, as @fig-publications-per-year showed, the overall output was nowhere near rich as in the following years, allowing single influential works to skew the visible means for those years.

In all of these cases, such outliers should provide clear points of interest during the screening process for possible re-evaluation of current term clusters for scoping.
Should they point towards gaps (or over-optimization) of sepcific areas of interest during those time-frames or more generally, they may provide an impetus for tweaking the identification query terms to better align with the prevailing literature output.

```{python}
#| label: fig-intervention-types
#| fig-cap: Predominant type of intervention
#| column: page

interv_type_df = (
    bib_df["zot_keywords"]
    .str.replace(r"\_", " ")
    .str.extractall(r"type::([\w ]+)")
    .reset_index(drop=True)
    .rename(columns = {0:"intervention type"})
)

sort_order = interv_type_df["intervention type"].value_counts(ascending=False).index
fig = plt.figure()
fig.set_size_inches(12, 4)
ax = sns.countplot(interv_type_df, x="intervention type", order=sort_order)
plt.setp(ax.get_xticklabels(), rotation=45, ha="right",
         rotation_mode="anchor")
plt.show()
```

{{++ TODO: describe intervention types with complete dataset ++}}

```{python}
#| label: fig-inequality-types
#| fig-cap: Types of inequality analyzed
#| column: page

inequ_type_df = (
    bib_df["zot_keywords"]
    .str.replace(r"\_", " ")
    .str.extractall(r"inequality::([\w ]+)")
    .reset_index(drop=True)
    .rename(columns = {0:"inequality type"})
)

sort_order = inequ_type_df["inequality type"].value_counts(ascending=False).index
fig = plt.figure()
fig.set_size_inches(12, 4)
ax = sns.countplot(inequ_type_df, x="inequality type", order=sort_order)
plt.setp(ax.get_xticklabels(), rotation=45, ha="right",
         rotation_mode="anchor")
plt.show()
```

Income inequality is the primary type of inequality interrogated in most of the relevant studies.
This follows the identified lens income inequality can provide through which to understand other inequalities ---
many studies use income measurements and changes in income or income inequality over time as indicators to understand a variety of other inequalities' linkages through.

{{++ TODO: describe inequality types with complete dataset ++}}

# Synthesis of Evidence

This section will present a synthesis of evidence from the scoping review.
The evidence will be presented by type of policies and world regions.
The section will also present a discussion on the implications of the current evidence base for policy and underscore key knowledge gaps.

Since policies employed in the pursuit of increased equality can take a wide form of actors, strategy approaches and implementation details,
the following synthesis will first categorize between the main inequality (or combination of inequalities) a policy is aimed at improving.

{{++ sort by intervention? gender+generational=maternity:paid leave,education benefits; disability+others; education+others  ++}}

## Income inequality through a vertical lens

One of the primary lenses through which policy interventions to reduce inequalities in the world of work are viewed is that of income inequality, often measured for all people throughout a country or subsets thereof.

Income and gender

income and racial/ethnicity

income and disability

income and spatial

income and age

{{++ first describe income inequ as vertical then view it through horizontal lenses of other inequalities ++}}

## Gender inequality

Gender inequality is the second most reviewed dimension of workplace inequality in the study sample,
with a variety of studies looking at predominantly it through the lens of female economic empowerment or through closing gender pay gaps.

<!-- economic empowerment and wage gap -->
Looking at the returns of the Tanzanian 'Universal Primary Education' programme on consumption and on rural labour market outcomes, @Delesalle2021, finds outcomes that additionally differ along spatial and gender lines.
The programme both attempted to increase access to schools but also changed curricula to contain more technical classes, judged relevant to increase equity in rural areas.
Even though the programme aims to increase universal equality of access to education, the study finds that gender, geographical and income inequalities persist throughout, with individuals that complete primary education more likely to be male urban wage workers.
The study measures returns purely on consumption of households to show the estimated effect on their productivity ---
here, it finds generally positive returns but greatest for non-agricultural work, self-employed or as wage work.
Importantly, the introduction of more technical classes, however, also changes employment sector choices, with men working less in agricultural work and more in non-farm wage sectors and an increased probability for rural women to both work in agriculture and to work formally.
Limitations of the study include the inability to directly identify intervention compliers and having to construct returns for each household head only and a possibly unobserved 'villagization' effect by bringing people together in community villages for their education leading to other unobserved variable impacting the returns.

@Al-Mamun2014 conduct a study on the impacts of an urban micro-finance programme in Malaysia on the economic empowerment of women.
The programme introduced the ability for low-income urban individuals to receive collateral-free credit.
The study finds that the programme, though not specifically aimed at women, indeed increased women's economic empowerment with an increase in household decision-making, as well as increased personal economic security.
Primarily this is due to the increased access to finance, though it also functions thorugh an increase of collective agency established for the women in organised meetings and trainings.
It also finds, however, that the empowerment outcomes are constrained by the inability for individuals to obtain loans, with the programme only disbursing group loans which are harder to achieve through obstacles to collective organisation by different racial and socio-demographic backgrounds in each dwelling.
The study is somewhat limited in its explanatory power since even through its random sampling design it can not establish control for all factors required in experimental design.

In an observational study looking at the inclusive or exclusionary effects of infrastructure development, @Stock2021 analyses the 'gender inclusive' development of a solar park in India which specifically aims to work towards micro-scale equality through regional uplifting.
The project included a training and temporary employment to local unskilled/semi-skilled labour.
It finds that the development instead impacted equality negatively, creating socio-economic exclusion and disproportionately negatively affected women of lower castes.
While acquiring basic additional skills, none of the women participating in training remained connected to the operators of the solar park and none were hired.
An insignificant amount of women from local villages were working at the solar park, of which most belonged to the dominant caste, and the redistributive potential was stymied through capture by village female elites.
The author suggests this is an example of institutional design neglecting individual agency and structural power relations, especially intersectional inequalities between gender and caste.
The study is limited in explanatory power through its observational design, not being able to make causal inferences.

<!-- maternal intersection, children -->
A variety of studies also look at female economic empowerment outcomes through a more generational lens,
focusing on the effects of interventions aimed at maternity support ---
childcare programmes, paid leave and maternity benefits.

@Broadway2020 study the introduction of universal paid maternal leave in Australia, looking at its impacts on mothers returning to work and the conditions they return under.
It finds that, while there is a short-term decrease of mothers returning to work since they make use of the introduced leave period, over the long-term (after six to nine months) there is a significant positive impact on return-to-work.
Furthermore, there is a positive impact on returning to work in the same job and under the same conditions,
the effects of which are stronger for more disadvantaged mothers (measured through income, education and access to employer-funded leave).
This suggests that the intervention reduced the opportunity costs for delaying the return to work, and especially for those women that did not have employer-funded leave options, directly benefiting more disadvantaged mothers.
Some potential biases of the study are its inability to account for child-care costs, as well as not being able to fully exclude selection bias into motherhood.
There also remains the potential of results being biased through pre-birth labour supply effects or the results of the financial crisis, which may create a down-ward bias for either the short- or long-term effects.

@Clark2019 undertake an experimental study on the impacts of providing childcare vouchers to poor women in urban Kenya, estimating the impacts on their economic empowerment.
The empowerment is measured through disaggregated analyses of maternal income, employment probability and hours worked.
It finds that, for married mothers there was a significantly positive effect on employment probability and hours worked, suggesting their increased ability to work through lower childcare costs increasing personal agency.
For single mothers, it finds a negative effect on hours worked, though with a stable income.
The authors suggest this is due to single Kenyan mothers already working increased hours compared to married mothers, though the effect shows the ability of single mothers to shift to jobs with more regular hours, even if they are not compatible with childcare.
Minor limitations of the study are its restriction to effects within a period of 1 year, and a somewhat significant attrition rate to the endline survey.

@Hojman2019, in an experimental study looking at the effects of providing free childcare for poor urban mothers in Nicaragua under the 'Programo Urbano', examine the effects on inequality for mothers and children.
It finds that providing free childcare for young children of poor mothers significantly increases the employment probability of the mothers (14ppts) independently of the childcare quality.
It also finds significantly positive impacts on the human capital of the children, though dependent on the quality of childcare facilities.
This suggests childcare costs being removed through a quasi-subsidy reducing the required childcare time burden on mothers, increasing parental agency and employment choices.
Some limitations to the study include a relatively small overall sample size, as well as employment effects becoming insignificant when the effect is measured on randomization alone (without an additional instrumental variable).

## Spatial inequality

<!-- non-spatial policy but spatial effects -->
@Gilbert2001 undertake a study looking at the distributional effects of introducing a minimum wage in Britain, with a specific spatial component.
Overall it finds little effect on income inequality in the country.
It finds that the effects on rural areas differ depending on their proximity to urban areas.
While overall income inequality only decreases a small amount, the intervention results in effective targeting with remote rural households having around twice the reduction in inequality compared to others.
Rural areas that are accessible to urban markets are less affected, with insignificant impacts to overall income inequality reduction.
One limit of the study is that it has to assume no effects on employment after the enaction of the minimum wage for its results to hold.

In a study on the impacts of minimum wage and direct cash transfers in Brazil on the country's income inequality,
@SilveiraNeto2011 especially analyse the way the policies interact with spatial inequalities.
It finds that incomes between regions have converged during the time frame and overall the cash transfers under the 'Bolsa Familia' programme and minimum wage were accounting for 26.2% of the effect.
Minimum wage contributed 16.6% of the effect to overall Gini reduction between the regions while cash transfers accounted for 9.6% of the effect.
The authors argue that this highlights the way even non-spatial policies can have a positive (or, with worse targeting or selection, negative) influence on spatial inequalities,
as transfers occurring predominantly to poorer regions and minimum wages having larger impacts in those regions created quasi-regional effects without being explicitly addressed in the policies.
Some limitations include limited underlying data only making it possible to estimate the cash transfer impacts for the analysis end-line,
and minimum wage effects having to be constructed from the effects wages equal to minimum-wage.

@Kuriyama2021 look at the effects of Japan's move to decarbonise its energy sector on employment, especially rural employment.
It finds that, while employment in general is positively affected, especially rural sectors benefit from additional employment probability.
This is due to the renewable energy sector, while able to utilise urban areas for smaller scale power generation, being largely attached to rural areas for larger scale projects such as geothermal, wind power or large-scale solar generation.
The study also suggests some possible inequality being created in between the different regions of Japan due to the Hokkaido region having limited transmission line capacity and locational imbalance between demand and potential supplies.
Limitations include its design as a projection model with multiple having to make strong assumptions about initial employment numbers and their extrapolation into the future,
as well as having to assume the amount of generated power to increase as a stable square function.

Similarly, @Whitworth2021 analysis of the spatial consequences of a UK work programme on spatial factors of job deprivation or opportunity increases.
The programme follows a quasi-marketized approach of rewarding employment-favourable results of transitions into employment and further sustained months in employment.
The author argues, however, that the non-spatial implementation of the policy leads to spatial outcomes.
Founded on the approach of social 'creaming' and 'parking' and applied to the spatial dimension,
the study shows that already job-deprived areas indeed experience further deprivations under the programme,
while non-deprived areas are correlated with positive impacts, thereby further deteriorating spatial inequality outcomes.
This occurs because of providers in the programme de-prioritizing the already deprived areas ('parking') in favour prioritizing wealthier areas for improved within-programme results.

Highlighted by these studies, one issue of spatial inequality especially is that in many cases policies are crafted that are targeted without any spatial component, intended to function nationally.
These non-spatial policies will, however, carry effects on inequalities that are created or exacerbated by spatial inequalities themselves.
Ideally, policies can make use of spatial effects without having to include explicit spatial components,
as was the case with Brazilian social programmes [@SilveiraNeto2011].
Often however, spatial targeting considerations have to be explicitly invoked to not lose effectiveness or, worse, create adverse outcomes for specific spatial variations,
as may be the case for some regions in Japan infrastructure efforts [@Kuriyama2021].

<!-- explicitly spatial policies -->
@Blumenberg2014 look at the effects of a housing mobility intervention in the United States on employment for disadvantaged households,
and comparing its impacts to the ownership of a car for the same sample.
It follows the 'Moving to Opportunity' programme which provided vouchers to randomized households for movement to a geographically unrestricted area or to specifically to a low-poverty area (treatment group),
some of which are in areas with well-connected public transport opportunities.
The sample for the study is made up predominantly of women (98%).
No relationship between programme participation and increased employment probability could be established.
However, a positive relationship exists between owning an auto-mobile and improved employment outcomes for low-income households,
as well as including those households that are located in 'transit-rich' areas.
Access to better transit itself is related to employment probability but not gains in employment -
the authors suggest this reflects individuals' strategic relocation to use public transit for their job.
However, moving to a better transit area itself does not increase employment probability,
perhaps pointing to a certain threshold required in transit extensiveness before it facilitates employment.
Ultimately, the findings suggest the need to further individual access to auto-mobiles in disadvantaged households or for extensive transit network upgrade which have to cross an efficiency threshold.
Some limitations of the study are its models low explanatory power for individual outcomes, more so among disadvantaged population groups,
as well as some remaining possibility of endogeneity bias through unobserved factors such as individual motivation or ability.

@Adam2018 model the effects of transport infrastructure investments in Tanzania on rural income inequalities and household welfare inequalities, modelled through consumption indicators.
Generally it finds that the results of public investment measures into transport infrastructure largely depend on the financing scheme used.
Comparing four financing schemes when looking at the effects on rural households, it finds that they are generally worse off when the development is deficit-financed or paid through tariff revenues.
On the other hand, rural households benefit through increased income from measures financed through consumption taxes, or by external aid.
The general finding is that there is no Pareto optimum for any of the investment measures for all locations,
and that much of the increases in welfare are based on movement of rural workers out of quasi-subsistence agriculture to other locations and other sectors.
The study creates causal inferences but is limited in its modelling approach representing a limited subset of empirical possibility spaces,
as well as having to make the assumption of no population growth for measures to hold.

# Conclusion

The section with conclude with reflections on the implications of findings for policy.

{{< pagebreak >}}

# References

::: {#refs}
:::

{{< pagebreak >}}

# Appendix

## Full search query

```{python}
#| echo: false
#| output: asis
with open(f"{SUPPLEMENTARY_DATA}/query.txt") as f:
    query = f.read()

t3 = "`" * 3
print(f"""
```sql
{query}
{t3}
""")
```

{{< pagebreak >}}