The Institute of Economic Research and Public Policy (IERPP) has conducted a critical review of the Ghana Statistical Service (GSS) Consumer Price Index (CPI) and Inflation Report for August 2025.
While the report is detailed and comprehensive, IERPP has identified several shortcomings that call into question its accuracy, neutrality, and clarity.
These concerns raise important issues about the integrity of official statistics and their communication to policymakers and the public. The key findings are outlined below.
Misleading framing of disinflation as falling prices
The report repeatedly states on page 18 under the key takeaways that “inflation is falling steadily” and even suggests “falling prices.” This is misleading.
Lower inflation does not mean prices are falling. An inflation rate of 11.5% still means prices are 11.5% higher than a year ago. Only the month-on-month rate of 1.3% shows a temporary fall between July and August. This confuses disinflation with deflation.
Disinflation is when prices are still rising but at a slower pace than before. For example, if inflation drops from 20% in 2024 to 11% in 2025, it means prices are still higher than last year, just not increasing as fast. Think of it like a car slowing from 120 km/h to 80 km/h; it’s still moving forward, only slower.
Deflation, on the other hand, is when prices actually fall year-on-year. For instance, if inflation is -2%, it means goods and services cost 2% less than the previous year. Here, the car shifts into reverse and starts moving backward.
Overstating policy success
In page 5, the report claims the drop in inflation “proves Ghana’s stabilization programme is delivering results.” This is policy advocacy rather than statistical reporting. Data alone cannot prove causation since external factors such as exchange rates, global prices, and harvests also influence inflation. Such framing risks compromising the neutrality of the GSS.
Data inconsistencies and errors
Several inconsistencies are evident. Transport inflation is shown as 7.7% in July and 5.2% in August, yet its contribution to MoM inflation is positive (+1.9%). Negative inflation should reduce overall CPI, not increase it.
Food weights are listed at 42.7%, but contributions in some tables do not add up to headline inflation. The Top 20 Contributors table includes negative contributions larger than item weights, suggesting errors or unclear methodology.
Regional inflation contradictions
Page 7 reports Bono East as the lowest inflation region at 6.1%, but page 20 shows Upper East as the lowest at –3.2% MoM. The difference between year-on-year and month-on-month is not explained clearly, making the report confusing and potentially misleading.
Headline MoM vs Food/Non-Food inconsistency
Food carries a weight of 42.7%, and Non-Food 57.3%. Reported MoM inflation is –2.5% and –0.1% respectively. Weighted together, this equals –1.12%, not the reported –1.3%. To match the headline figure, Food would need to be about –2.9%.
Headline MoM vs Goods/Services inconsistency
Goods have a weight of 72.5%, and Services 27.5%. Reported MoM inflation is -1.6% for Goods and +0.3% for Services. Weighted together, this equals –1.08%, not –1.3%. For the headline to hold, Goods would need to be about –1.9%.
Local vs imported MoM consistency
Locally produced items (68.5%) are reported at –1.7% and imported items (31.5%) at –0.3%. The weighted result is –1.26%, which is close to –1.3% and can be explained by rounding. By contrast, the Food/Non-Food and Goods/Services discrepancies are too large to be due to rounding alone.
Weak recommendations
The recommendations on page 21 are generic slogans such as “shop smarter,” “lock in disinflation,” and “green light to invest.” These are not evidence-based or statistically grounded. By offering prescriptive advice instead of neutral data interpretation, the report risks undermining its credibility.
The headline MoM figure of –1.3% does not reconcile with the subgroup MoM rates given their weights. The inconsistencies of around 0.2 percentage points suggest either subgroup rates were rounded incorrectly, transcribed wrongly, or based on different weights than published. Presenting subgroup data to two decimals or correcting the figures would resolve these issues.
The report confuses disinflation with deflation, overstates policy success, contains data inconsistencies, mixes up year-on-year and month-on-month analysis, and ends with weak recommendations. Improvements in accuracy, neutrality, and clarity are necessary to maintain confidence in official statistics.
Author: Prof. Isaac Boadi, Dean, Faculty of Accounting and Finance, UPSA, and Executive Director, Institute of Economic and Research Policy, IERPP