Measuring ODA Spending to Achieve Target 8.7
A growing body of evidence suggests that underdevelopment renders individuals vulnerable to forced labour, modern slavery, human trafficking and child labour (Target 8.7 exploitation). There is also some evidence to suggest that the presence of Target 8.7 exploitation may negatively influence wages, labour productivity and economic growth by perpetuating poor access to education, training and opportunities to participate in the economy. Both factors suggest an important role for development assistance and programming in addressing Target 8.7 exploitation.
How much are states spending on development assistance programming to achieve Target 8.7? Where are they spending their resources, and on what kinds of interventions? In a report recently published by UNU-Centre for Policy Research as a contribution to the Delta 8.7 – The Alliance 8.7 Knowledge Platform, we (myself and co-author James Cockayne) seek to begin answering these questions, by generating a baseline of Official Development Assistance (ODA) spending on Target 8.7 issues for 2000–2013.
Prior investigations of official aid flows targeting particular development issues or SDGs have often relied on pre-established codes used by the Development Assistance Committee (DAC) of the Organisation for Economic Co-operation and Development (OECD) that numerically identify the core issue area of each ODA project. Unfortunately, as the DAC coding system predates the establishment of the SDGs, coding is not often well-aligned with either the 17 overarching sustainable developments goals nor the 169 individuals target indicators of the SDGs. When DAC codes do not line up well with a particular topic or SDG target of interest, researchers have turned to the help of human coders to read through project descriptions, determine their relevance to a particular target or SDG and manually assign appropriate codes to the data. This is precisely the aim of the AidData project when compiling the Financing the SDGs Core dataset, which matches ODA projects and associated USD commitments across the 17 SDGs, though target-level spending is not available.
To address questions specific to SDG Target 8.7 spending we adopted a different approach to data collection that did not require us to rely on manual searching of millions of ODA project descriptions. Starting from the AidData Core Research Release 3.1, we constructed a sample of aid projects that referred to a set list of key root words relating to Target 8.7, in their project descriptions. This list of root terms was designed to create broad categories to capture seven overarching forms of exploitation that fall within Target 8.7: forced labour, child labour, child soldiering, human trafficking, forced marriage, modern slavery and servitude. We constructed this sample through a simple form of Natural Language Processing (NLP) known as entity extraction, using an algorithm to crawl mechanically through the 1,252,036 project descriptions available through AidData. The method produced a sample of 6,408 ODA projects relevant to the forms of exploitation that fall under SDG Target 8.7.
There are limitations both in our methodology and in the utility of the resulting analysis. Our research sizes aid spending to address forced labour, modern slavery, human trafficking and child labour by OECD countries between 2000 and 2013. It does not incorporate spending by other significant donors who were not OECD DAC members at the time (such as Qatar). It also does not cover non-ODA assistance, domestic expenditure or the growing flows of charitable giving directed at these concerns.
Our analysis explores what this data tells us about how much is being spent over time, by which donors, to which countries, to address what forms of Target 8.7 exploitation.
How much is being spent:
We found that ODA commitments between 2000 and 2013 totaled USD 4,128,037,703, or an average of just under USD 295 million per year, though commitments varied a great deal over time peaking at almost 500 million in 2009, plateauing at around 450 million per year between 2010 and 2013.
Who is contributing:
The US stands as the top donor (2.5 billion total), committing almost 10 times as much ODA as the next highest donor, Canada (263 million), followed by Norway (216.7 million), Australia (204.8 million), Sweden (138.9 million) and the UK (120.3 million). The report also presents a breakdown of each donor’s spending over time.
Where is the aid flowing:
A handful of countries received the bulk of average yearly ODA commitments relevant to Target 8.7, including Afghanistan (23.5 million), India (19.3 million), Colombia (19 million) and the Democratic Republic of Congo (14.5 million). Comparing US spending to all other OECD country spending reveals that recipients received the majority of assistance either from the US or from the rest of the OECD countries combined.
To what issues:
We find significant variation in spending between forms of Target 8.7 exploitation. In 2000 ODA commitments to projects aimed at eradicating child labour vastly outpaced commitments dealing with other forms of exploitation. By the last few years of the sample, the largest aggregate pool of ODA commitments targeted human trafficking. Funding directed at child labour gradually declined since 2004, hitting a low in 2011, with a subsequent uptick from 2012 to 2013.
Overall, the analysis presented in the report is intended to be descriptive rather than prescriptive. Nonetheless, we offer a few recommendations that will allow us to expand on what we know about aid contributions to Target 8.7 issues. First, future analysis would benefit substantially from the availability of DAC CRS data with more comprehensive and precise purpose codes to tag and track ODA flows to Target 8.7 issues. Second, the research we present does not provide an analysis of non-ODA contributions as there is no reliable, robust data source of aid that does not count as ODA, domestic spending or private charitable giving. Establishing a centralized database to systematically capture this information can move us a great deal further than the baseline provided in our report.
Kelly Gleason is the Delta 8.7 Data Science Lead.
This article has been prepared by Kelly Gleason as a contribution to Delta 8.7. As provided for in the Terms and Conditions of Use of Delta 8.7, the opinions expressed in this article are those of the author and do not necessarily reflect those of UNU or its partners.