Executive summary

Widespread violence between two gangs started on Sunday 24 April 2022, spread over six communal sections in the north of Port-au-Prince before the situation progressively stabilised after 06 May (sources: media and humanitarian stakeholders). Between 23 April and 20 May, IOM's Displacement Tracking Matrix (DTM) and the Haitian Directorate General of Civil Protection (DGPC) estimated 34,600 individual displacement movements. At least 188 deaths and 113 people injured were reported (OCHA), and at least 81 houses burned (RNDDH).

Based on data from the mobile network operator (MNO) Digicel Haiti up to 18 July 2022, and using 21 January to 21 April 2022 as baseline period, this report provides further evidence of large-scale displacement and additional information on these population movements:

  • At least 27,200 Digicel subscribers left the affected area between 24 April and 8 May, with most of the likely internally displaced persons (IDPs) in the department of Ouest, Centre and Artibonite.
  • The 1ère section of St-Martin (Delmas commune) is the section which hosted the largest numbers of likely IDPs among Digicel subscribers, at least 6,500 as of 08 May;
  • A large share of the likely IDPs (60%) are spread across Haiti in small numbers, particularly to the north of Port-au-Prince;
  • Within the affected area, the 3ème Section de Bellevue (Tabarre Commune) experienced the largest drop of residents amongst Digicel subscribers (at least 7,700) and for the longest period, as the number of residents remained below the baseline for 50 days;
  • Flows of Digicel residents out of the affected area take place in two stages: initially mainly from the 3ème Section Bellevue (Tabarre), and later also from the 1ère Section des Varreux (Croix-des-Bouquets) and the 2ème Section des Varreux (Croix-des-Bouquets).

Figures provided in this report are conservative estimates because:

  • They are based on the analysis of movements of Digicel SIM cards only, not of the whole population: for reference, in 2018 in Port-au-Prince, only 76% of the adult population owned a phone (source: DHS Haiti 2016/2017) and the latest public figures (2016) on Digicel market shares nationally are 74% (CONATEL, 2016).
  • Only the subset of Digicel SIM cards with sufficient activity for our analyses are included;
  • Therefore, the figures reported here are lower-bound estimates.

Considerations

For the sake of simplicity, Digicel SIM cards are often referred here as people, however: the results are not representative of the affected area’s population because:

  1. Not everyone uses a Digicel SIM cards nor uses it enough to be included in the analysis,
  2. Some SIM cards are shared within households, 
  3. Not all movements are observed as the approximate location of a SIM cards is registered only when a call is made or received.

The delimitation of the affected area does impact the results of the analysis. Seasonal effects may explain part of the results toward the end of the study period. Only significant changes are reported here. As significance thresholds vary between cell clusters and as the count of Digicel SIM cards of some cell clusters outside of the affected area is not detrended (see annex), the significant decrease in Digicel SIM cards within the affected area (21,500) differs from the significant increase in Digicel SIM cards outside of the affected area (27,200).

Read the report (PDF)

Authors

  • This report was authored by the Flowminder Foundation, a non-profit foundation specialising in the analysis of anonymous mobile operator data, satellite imagery and household survey data for humanitarian and development purposes.

Acknowledgements

  • This report was made possible thanks to the pseudonymised Call Detail Records (CDR data) provided by Digicel Haiti on the Flowminder Foundation server hosted behind Digicel Haiti firewalls in Port-au-Prince, Haiti.

  • This work was coordinated and funded by the International Organization for Migration (IOM), thanks to funding received by IOM from the European Union's Instrument contributing to Stability and Peace.

  • This work builds on Flowminder’s existing work in Haiti funded by the Agence Française de Développement and the William and Flora Hewlett Foundation which focuses on enabling the increased access to and use of mobile operator data, in low- and middle-income countries, in ways
    that are ethically sound, financially viable, and sustainable.

  • The United Nations Office for the Coordination of Humanitarian Affairs (OCHA) in Haiti is acknowledged for feedback on earlier drafts of this report.

Back to Publications