Benchmarking Mauritius’ growth performance
Revisiting Mauritius’ economic miracle with help from AI on fact-checking
Jaime de Melo
31/03/26
Benchmarking Mauritius’ growth performance
Mauritius aspires to high-income status, but its per capita income growth over the past 16 years has been lower than the average for its group, the upper-middle-income (UMI) category in the World Bank’s income classification. The same slippage is observed for indicators of governance and port efficiency, but not for doing business indicators. Taking the more specific case of Port-Louis port, its efficiency has remained low compared to other ports in East and South Africa. However, Mauritius performs well in Doing Business indicators. These examples demonstrate that Mauritius should utilize more extensively available data from comparator countries when evaluating its performance.
South Korea is the poster child on the World Bank’s per capita income ladder. South Korea climbed from the Low- Income Category (LIC) to the High-Income Country (HIC) group in about 35-40 years, Mauritius’ success story also sits up there as it too started from the LIC status at independence, reaching temporarily the HIC briefly in 2020 (based on 2019 data), before returning now to the UMI category following the Covid-19 crisis.
For the first 20 years following independence, Mauritius’ performance was hailed “the Mauritian miracle” as early as 2001. This is a long enough period to claim a ‘miracle’ as Mauritius’ growth outperformed Sub-Saharan Africa (SSA) peers by 3-4 percentage points yearly. (Over 1973-1999, Mauritius clocked an average GDP growth of 6.0 percent a year against 2.5 percent for Sub-Saharan Africa (SSA), a sufficient overperformance to merit the praise).
Figure 1 extends the stretch to 50 years using Purchasing Power Parity (PPP) to correct for inflation and for differences in cost of living across countries. Estimates are available for 25 SSA countries over 1970-2022. Mauritius averaged 4.7% per capita income growth over the period, but its growth slowed to 3.2% over 2010-2023 (annually compounded growth rates in the notes to the figures show that only Botswana caught up to Singapore over the period). Mauritius had the same average yearly GDP growth of 4.0 percent as the average for SSA. Mauritius is no longer an exceptional performer. The post-2010 period marks a clear deceleration in Mauritius’ growth trajectory, reinforcing the view that the country has transitioned from a high-growth outlier to an average performer among SSA countries.
Figure 1: GDP per capita trajectories: Botswana, Mauritius Seychelles
Note: PPP estimates measure what income can actually buy domestically. It corrects for cheaper non-tradable goods in poorer countries at the cost of relying on infrequent surveys and reliance on interpolation. Preferred over measures at current exchange rates for welfare and long-term performance analysis. Top 5 performers (excluding Equatorial Guinea selected from the sample of 25 countries.
Semi-log scale: slope shows GDP per capita growth rate. Annual per capita compounded growth rates: Botswana (5.6%); Singapore (4.8%); Cape Verde (3.2%); Mauritius (3.2%); Seychelles (1.6%).
Source: Our World in Data
Economists observing an outstanding performance followed by a performance close to the average call it‘a reversion to the mean’ or ‘a regression to mediocrity’, to use the expression coined by Francis Galton in the 19th century. But reversion to the mean is not the case here because performance is measured over two decades or more, and performance is not randomly distributed.
So, is Mauritius slipping down the WB’s income classification ladder? Or perhaps Mauritius will benefit from a rising tide that will lift all boats (the number of countries in the LIC category fell from 45 in 2010 to 28 in 2024/25 while the number in the HIC category rose from 70 to 83)? Let’s take a peep under the hood to see how Mauritius is performing across three dimensions: governance, time to set up a business, and performance of the port of Port-Louis.
Declining Governance performance in the UMI group
While Mauritius continues to outperform the average UMI country across all 6 dimensions of the widely used Worldwide Governance Indicators (WGI), the margin has narrowed significantly. This pattern suggests a process of institutional convergence, with Mauritius moving from a high-governance outlier toward the mean of its income group. Significantly, the erosion of Mauritius’ governance advantage is most pronounced in government effectiveness, regulatory quality, and especially control of corruption—precisely the dimensions most closely associated with infrastructure performance and policy implementation.
Figure 2: Mauritius’ governance performance is worsening
Note: Governance indicators are from the World Bank Worldwide Governance Indicators (WGI). Percentile ranks range from 0 to 100, with higher values indicating better governance outcomes. The number of sources (in parentheses) for each indicator refers to the approximate number of underlying data sources (surveys, expert assessments, and institutional indicators) used to construct each governance dimension. For percentile value ranges, ChatGPT uses the triangulation method
WGI percentile ranks 0-100.A higher percentile is a higher rank. The spider chart shows :(i) Mauritius is above the average for the UMI group (54-56 countries) for all WGI categories, both in 2010 and 2024; (ii) For Mauritius, the 2024 circles are closer to the origin for all categories, indicating a decline in governance. (iii). No decline for the UMI average on any category. (iv) Largest decline for the corruption indicator
Source: ChatGPT extraction from WB data to query asking for a comparison of 2010 and 2023/4 data.
The WGI has shortcomings. First, each one of the 6 indicators is an average of 30-40 subcomponents. Second, the WGI includes measures of institutional capacity, perceptions and corruption outcomes. This makes it difficult to distinguish between flaw in design and flaw in implementation. Third, the WGI also suffers from perception bias as country correspondents differ in perceptions and are likely to be influenced by the latest ‘home news’. So, can one attribute a change in the index value to a change in bribes, elite capture, state capture (i.e. rent extraction), or public sector integrity, even if the decline in the ranking on control of corruption is the greatest?
Mauritius outperformed peers in time to start a business.
Starting a business requires filing 7-8 procedures in UMI countries. The time required to start a business is an outcome easily measured. In Mauritius, it took approximately 46 days in the mid-2000s. By 2020, the time was reduced to about 6 days, placing Mauritius among the fastest globally. While Upper Middle-Income (UMI) countries also improved, reducing average start-up time to around 20–25 days, Mauritius significantly outperformed its peers, remaining the star in the UMI group (table 1). The sharp improvement in business entry procedures highlights the success of targeted regulatory reforms.
Table 1: Mauritius in the Doing Business rankings
Notes: UMI sample of 55 countries.
Source: ChatGPT calculations from Doing Business data
The port of Port-Louis is losing ground on the East African coast
In a summary report of a recent seminar of senior experts, Blin (2026) reviews Port-Louis performance. She notes the poor performance ranking on the CPPI (Container Port Productivity Index) of efficiency container handling (rank of 327/348 ports in 2022). Noting the growing challenge for transshipment and bunker fuel activities (bound to grow with the improved Toamasina port), she suggests a list of policy recommendations to improve the port’s performance. So do the 2024 Report to the Competition Commission and the 2025 report from the Mauritius Port Authority (MPA). These reports would be strengthened if their recent vs. past performance included a regional dimension as in figure 3.
Figure 3: Port-Louis is losing ground to regional competitors.
Note: 2010 values are harmonized from ADR (2010) indicators; 2023 values are constructed using recent port throughput data and CPPI-consistent turnaround estimates. This measure captures efficiency and scale, while the CPPI controls for scale. Note that neither measure accounts for the open sea facing of Port-Louis (see here for a critique of CPPI)
Figure 3 compares a measure of efficiency across all 8 East and South African ports for 2010 and 2023. These ports are in competition for transshipments and bunker fuel activities. Four clear patterns emerge. Durban and Cape Town dominate in terms of volume (throughput) and turnaround time for both years. Second, Mombasa and Djibouti also improve along both indicators. Third, Port-Louis is losing ground as volume and turn-around time do not change, so it is losing ground. Fourth, Dar Es Salaam has not overcome the deleterious effects of extensive rent-seeking activities. in the early 2010s.
Port-Louis is still probably losing ground if one considers annual downtime due to weather-related disruptions. [1]When benchmarking is across all UMI ports using the CPPI index that controls for scale, Port-Louis is still underperforming relative to the UMI group (see table 4).
Table 2: Benchmarking Port-Louis Port Productivity for containers in UMI sample*
Notes: * Source: World Bank (2025) report on container performance. Data for 2023 Sample has 43 UMI ports including Port-Louis.
CPPII (Container Port Performance Index) rank. The CPPI does not rely on throughput data (million TEUs). CPPI = Simple average of: (i) time in port (port stay / berth time); (ii) vessel size and call structure; (iii) productivity=efficiency of container moves calculated as productivity=number of container moves/crane hours worked. This measure avoids mixing scale and efficiency It is a measure of port, not berth, performance. Over 2023/4, the Mauritius score ranks 369/403 in the sample, i.e. 368 ports did better.
Port benchmarks: High performance based on time in port: [Top:(12-14 hrs.; mid-tier: 24-40 hrs: low-performance: 40-70+ hrs]
Productivity benchmarks: World class (Singapore, Shanghai) =30-40 moves per hr; Efficient (Djibouti, Colombo) =20-30 moves per hr. Mid-tier (Port-Louis, Mombasa) =15-20 per hr.
Efficiency measures are constructed using harmonized throughput and turnaround time estimates based on ADR (2010) and World Bank (2023) benchmarks. Values are indicative and designed for cross-port comparison.
Source: ChatGPT elaboration from World Bank (2025, annex).
Costing delays at the port. For Mauritius, freight costs include about 7.5% of the total costs (cif) of imports. Direct regulated Port costs only account for about 5% of the total shipping costs. Delays at the Port create indirect costs, beyond the running costs of ships standing idle. Applying the widely-used estimate of Hummels and Schaur (2013) of one day lost in maritime shipment is equivalent to a tariff of 1.3% at destination to Mauritius with Djibouti as the frontier, gives a tariff equivalent of 1.1% due to additional waiting.
Dashboard
Figure 4 pulls together these three performance indicators. All three compare performance across countries and time to see how the relative position has evolved since the slowdown in growth around 2010 shown in figure 1. Mauritius is losing ground on the governance front while improving its already top-ranking on the time required to open a business.
Figure 4: Dashboard: Composite institutional performance
(Mauritius vs UMI (2010–2023).
Notes: Each indicator is normalized to 100. Higher values reflect higher score. UMI is average for 55 countries.
Governance: WGI indicator is average of the 6 components in figure 1 included in WGI.
Doing Business: Time to complete the 7 steps to open a business
Ports: Thruput/average turnaround time from figure 2 extended to all ports in UMI group.
The Way Ahead: Taking monitoring of economic activity seriously by benchmarking with the competition
After 25 years of outlier growth, Mauritius’ growth returned to the average for SSA over the period 201-2023. This was more than statistical reversion to the mean--often observed in sports—when an athlete’s (or broker’s) performance returns to the average in the sample. Digging deeper by comparing performance with other Upper Middle Income (UMI) countries, Mauritius maintained an outstanding ranking on the ease of Doing Business indicator (as reflected in the time needed to complete the necessary steps to set up a business). Mauritius experienced a gradual decline in the subjective indicator of governance and on the easy-to-measure (and less controversial) performance of the port of Port Louis. Significantly, this decline in port performance has taken place relative to its competitors on the East and Southern coasts of Africa and is robust to weather-related disruptions.
This decline, typically observed in rent-seeking societies, was recently alluded to by the respected weekly, The Economist. Speaking of the Mont Choisy Golf and Beach Estate, The Economist writes “Where African slaves and Indian indentured labourers used to toil, French and South African pensioners now mull over the merits of a five-iron on the 14th hole”. This Estate is among the symbols of Mauritius’ sustained performance since independence (for others, see Stiglitz on health care and education or Poncini (2018) on the success of the first example of supply chain trade in the Micro Jewels company). Could pensioners’ preoccupations also be spreading to a Mauritian elite and governing class engaging in rent-seeking activities?
Vignettes like those presented here should help towards better policy choices through more fact-based information for policy deciders. In an era where information is increasingly ubiquitous, Mauritius would benefit from more benchmarking. For example, why doesn’t Mauritius participate in the regular OECD PISA (Programme for International Student Assessment) surveys on literacy and reading performance in high school? Mauritius does not participate. According to a query to ChatGPT on participation by UMI countries in PISA “approximately 30 to 35 UMI countries participate in the PISA assessments [in the 2018-2022 wave], representing around 60 percent of the income group. Participation is selective, however, and tends to reflect countries with stronger institutional and statistical capacity”. Mauritius is unlikely to have weak institutional and statistical capacity like, say Equatorial Guinea, so why not participate?
References
Africa Development Report (2010) Ports, Logistics and Trade in Africa, Oxford University Press, https://www.afdb.org/sites/default/files/documents/publications/african_development_report_2010.pdf
Blin, M. (2026) “Towards a More Competitive and Greener Port Louis Harbor”, Charles Telfair Policy Brief, https://charlestelfaircentre.com/towards-more-competitive-and-greener-port-louis-harbour
Competition Commission (2024) Port Market Study: Final Report , https://competitioncommission.mu/wp-content/uploads/2024/11/MS011-Final-Report.pdf
Hummels, D. and G Schaur (2013) “Time as a Trade Barrier”, American Economic Review, 103(7), 2935-59, https://www.aeaweb.org/articles?id=10.1257/aer.103.7.2935
Mauritius Port Authority (MPA) (2025) Port Trade Performance: FY23/24 vs. FY24/25, http://www.mauport.com/sites/default/files/public/port_trade_performance_fy23-24_versus_fy24-25_2.pdf
Morisset, J., C. Moret, J. Regolo “How to Push Efficiency Enhancing Reforms at the Port of Dar Es Salaam?”, Africa Trade Policy Notes, World Bank, https://documents.worldbank.org/en/publication/documents-reports/documentdetail/515201468194638167
Poncini, José (2018) Bâtir sur ses rêves , éditions Vizavi, Port Louis
Stiglitz, J. (2011), “The Mauritius miracle”, Project syndicate, https://www.project-syndicate.org/commentary/the-mauritius-miracle-2011-03
The Economist (2025) “The Mauritian Miracle is losing its sheen: the island economy is less of a model than it used to be”, https://www.economist.com/middle-east-and-africa/2025/04/24/the-mauritius-miracle-is-losing-its-sheen
World Bank (2025) The Container Port Performance Index: 2020 to 2024: Lessons Learned,https://openknowledge.worldbank.org/server/api/core/bitstreams/695e8bdc-eb9a-439a-a8d5-228593831ce8/content
1
[1]A thorough evaluation would consider estimates of annual downtime/disruption days in the group. For Djibouti and Dar Es Salam, disruptions are not weather-related. For the other ports, ChatGPT gives the following ranges in days: Port- Louis and La Réunion [5-15]; Cape Town [15-25]; Toamasina [10-20]; Durban [5-10]. Estimates via a search by ChatGPT.