Data limitations
GHS Tracking aggregates publicly reported financial disbursements from funders to recipients for activities related to global health security, from January 2014 and December 2022. This effort draws from a large number of data sources, each with varying reporting standards, methods of summarizing, and levels of metadata quality. Given the lack of consistency in reporting, we believe these data are the best currently available. The limitations of these data are detailed below.
For more information see the Technical appendix or contact us at globalhealthsecurity@georgetown.edu.
Funding totals
The site does not capture a complete accounting of funds distributed, but rather a reporting of those funds for which data were publicly available and readily accessible using the project methods.
Specifically, GHS Tracking is generally unable to track dollars “end-to-end” due to a lack of standardized and transparent public reporting about aid. For this reason, the total values reported for “Funds disbursed” and “Funds received” often diverge, and funds reported in the “top 10 JEE core capacities” charts may not represent total capacity-building funds. For example, while $2.5B disbursements from the Inter-American Development Bank (IDB) are reported for total capacity building, only $756.5M in funds disbursed align with specific JEE core capacities.
Summing across charts may not yield intuitive results due to rounding. Precise values can be derived directly by downloading the data from the data access page.
Intermediary stakeholders
The flow of funds from the initial donor to the final recipient is rarely reported, particularly when there are intermediaries facilitating fund distribution. In some cases, intermediaries may be described in the data as funders or recipients in publicly reported data. Data are described on the site based on public reporting and were de-duplicated wherever possible. For example, Abt Associates Inc. and DAI International are shown as funders for a handful of projects even though the organizations contract with USAID to implement global health-related projects. The GHS Tracking team does not have visibility into the details of contracts or implementation plans, and thus cannot identify the original funding entity; therefore, these transactions are left “as reported.” Similarly, disbursements listed as disbursed to Switzerland may have been, in practice, disbursed to an international organization headquartered in Geneva.
Given these challenges, we have addressed double-counting by screening for duplicates: searching for any projects within 10% of the same amount disbursed in the same year to the same recipient and/or by the same funder. In addition, transactions >$100M are manually reviewed by the research team to identify any known duplicates of large value. Where it was unclear if a transaction or project was duplicated (due to vague description, dissimilar amounts, etc.), the team erred on the side of inclusion.
Incomplete data
Occasionally, incomplete transaction data are publicly reported. As a minimum, transactions included in GHS Tracking data must provide at least a funder name or a recipient name, the funding year, and a description of the project's purpose (either as a project name, description, key activities, etc.) For transactions that involve a group of recipients or a group of funders (e.g., Canada provides support to all Caribbean nations to respond to Zika), data were only included where specific amounts could be attributed to each party. If no specific information was provided about how the funding was allocated, the data were excluded.
Erroneous data
All data captured by GHS Tracking is from publicly reported sources. Most of this reporting is manually entered by stakeholders, which can introduce human errors. To address the most egregious of these errors, we perform manual reviews for transactions >$100M. This enables us to flag substantial errors, such as instances where reported figures differ by orders of magnitude. For instance, manual review flagged a record from IATI that showed a $200B contribution from Japan to Morocco for COVID-19 response. Upon further investigation, a press release showed that the actual contribution was $200M and that the reporting party likely accidentally added three zeroes. In this case, the research team manually corrected the value captured in our database and notified the data owners about the mistake.
Domestic spending
Particularly as the COVID-19 pandemic unfolded, national governments invested heavily in national public and medical health response often as a complement to or in place of foreign assistance. Domestic health spending (including national or subnational spending) is not captured in the tool, as national health account expenditures are not routinely or standardly reported.
Commitments and In-kind contributions
We did not capture committed funds for which disbursements were not also publicly reported. In-kind contributions are also not included, as the monetary value is not well-defined.
