Inflation in Kazakhstan continues to ease in annual terms. Headline inflation slowed to 10.3% year-on-year in June, down from 10.4% in May. At the same time, monthly price growth accelerated to 0.8%, up from 0.7% a month earlier. In other words, the disinflation trend remains intact, but monthly price dynamics suggest that inflationary momentum has not yet fully faded.
June’s inflation profile was uneven. Food prices rose 10.4% year-on-year and 0.6% month-on-month, non-food goods increased 11.7% and 0.9%, while services climbed 9.0% and 1.1%, respectively. This confirms that non-food goods and services remain the main sources of price pressure. Higher transport, logistics and fuel costs continue to weigh on non-food prices, while services are increasingly reflecting the gradual pass-through of accumulated business costs.
The National Bureau of Economic Research had previously forecast June inflation at 10.2-10.4% year-on-year and 0.6-0.8% month-on-month. The actual outcome landed squarely within that range: annual inflation came in at 10.3%, while monthly inflation reached 0.8%.
Forecast accuracy by component was as follows:
- Food: 10.2% forecast vs. 10.4% actual (0.2 percentage point deviation)
- Non-food goods: 11.4% vs. 11.7% (0.3 percentage point deviation)
- Services: 9.1% vs. 9.0% (0.1 percentage point deviation)
Overall, June’s data validated the previous assessment of the inflation trend. However, the composition proved slightly firmer than expected. Food inflation eased more slowly, while non-food goods maintained strong monthly momentum. Services were marginally softer in annual terms, although monthly growth continues to point to persistent cost pressures.
Another important factor is the gradual easing of monetary conditions. On June 5, the National Bank lowered the base rate from 18.0% to 17.0%, with one additional cut of 50-100 basis points still possible later this year. This does not automatically imply faster inflation, but it does make price dynamics more sensitive to domestic demand, inflation expectations and the speed at which businesses pass higher costs on to consumers.
July also marks the beginning of a period when official statistics will more fully reflect the expiration of earlier price-monitoring measures. Monitoring of utility tariffs ended at the close of the first quarter of 2026, while fuel price monitoring concluded at the end of the first half. As a result, the second half of the year should provide a clearer picture of how fuel and tariff increases feed through to transport, logistics, utilities and service prices.
What’s changed in the forecast
Compared with the June outlook, the projected inflation path has been revised lower starting in July. This does not mean inflation risks have disappeared. Instead, they are increasingly shifting away from the headline index toward the composition of inflation in the second half of the year.
The July forecast has been lowered from 10.4% to 10.3%, August from 10.1% to 9.9%, and September from 9.8% to 9.6%.
The year-end outlook has also improved:
- October: 9.9% (previously 10.1%)
- November: 9.7% (previously 10.0%)
- December: 9.6% (previously 9.9%)
The revision reflects lower-than-expected inflation in May, the accurate June forecast for headline CPI and an updated seasonal profile for food prices during the third quarter. The model assumes relatively moderate monthly food inflation in July and August, allowing headline inflation to move into single digits more quickly.
Still, lower headline inflation does not imply broad-based easing across all categories. Non-food inflation is expected to cool much more gradually, while services are likely to become the most inflation-sensitive segment in the second half of the year. Fuel and utility price adjustments typically pass through the economy with a lag, first affecting transport and коммунal services, before spreading into logistics, repairs, retail trade and other service industries.
July forecast and the outlook through year-end
According to the updated baseline forecast, July inflation is expected to remain broadly stable while monthly price growth moderates:
- Headline inflation: 10.3% year-on-year (0.6% month-on-month)
o Food: 10.0% year-on-year (0.4% month-on-month)
o Non-food goods: 11.5% year-on-year (0.7% month-on-month)
o Services: 9.6% year-on-year (0.9% month-on-month)
The July profile suggests that food prices will be the main driver of disinflation. Slower food inflation should reduce the overall monthly pace, while non-food goods and services continue to keep inflation elevated. Services, in particular, are expected to accelerate in annual terms, reflecting the gradual accumulation of cost pressures.
The baseline scenario assumes headline inflation will return to single digits in August and remain there through the end of the third quarter. However, summer disinflation is expected to remain uneven.
Food inflation is projected at:
- 10.0% in July
- 10.1% in August
- 9.8% in September
Non-food inflation is forecast to slow from:
- 11.5% in July
- 11.3% in August
- 10.5% in September
Services inflation is expected to rise to 9.6% in July before easing to 8.9% in both August and September.
Looking further ahead, headline inflation is projected to remain in single digits throughout the rest of 2026:
- October: 9.9%
- November: 9.7%
- December: 9.6%
Even so, the year-end composition will be less favorable than during the summer. By December, food inflation is projected at 9.9%, non-food goods at 10.7%, and services at 10.9%. In other words, the decline in headline inflation will be driven largely by food prices and favorable base effects, while underlying inflation across goods and services is likely to remain more persistent.
The updated forecast suggests that Kazakhstan will return to single-digit inflation as early as August, with inflation remaining below 10% through the end of 2026. The downward revision reflects a softer starting point after the May data, an accurate June headline forecast and a more favorable seasonal outlook for food prices in the third quarter.
The key question for the second half of the year, however, is not simply how low headline inflation falls, but what is happening beneath the surface.
If fuel and tariff effects remain contained and monthly inflation slows across most categories, single-digit inflation is likely to become a durable trend, creating room for further cautious monetary easing. But if rising domestic costs spread more broadly into services and non-food goods, headline inflation may stay below 10% while underlying inflation proves far more persistent.
Forecast methodology
The forecast is built on a framework designed to maximize statistical consistency and ensure comparability across different data regimes.
- Data coverage. The models use monthly observations from January 2011 through the latest available data, providing a sufficiently long history to identify seasonal patterns and recurring inflation dynamics.
- Stationary time series. All models are estimated using stationary versions of the data, meaning the series are transformed so that their key statistical properties remain stable over time. This is essential for reliable econometric estimation, as non-stationary data increase the risk of spurious relationships and unstable coefficients, particularly over long samples.
- Consensus of independent econometric models. The final forecast is an aggregate of multiple model specifications rather than the output of a single model. This approach offers several practical advantages:
o Lower model risk. If one model temporarily over- or underestimates the impact of seasonality, one-off events or inflation persistence, averaging across models reduces that error.
o Greater resilience during unusual periods. Different models respond differently to outliers and structural breaks, making the consensus forecast less prone to overfitting.
o Better insight into inflation composition. Forecasting both headline inflation and its individual components improves confidence in identifying where price pressures remain concentrated – whether in food, services or non-food goods.
Forecasting inflation components is therefore not simply an analytical add-on. It is an essential way to determine whether disinflation is broad-based and sustainable or whether it is driven by only one part of the consumer basket.
Forecast accuracy
A key measure of the model’s reliability is its recent forecasting performance.
- During 2026, the average monthly forecasting error amounted to 0.2 percentage points for annual inflation and -0.03 percentage points for monthly inflation.
- For June 2026, the deviation from the forecast was 0.0 percentage points for annual inflation and 0.1 percentage points for monthly inflation.
These results indicate that the model captures both current inflation dynamics and short-term inflation persistence with a high degree of accuracy. Nevertheless, all econometric forecasts have inherent limitations. Unexpected supply shocks, disruptions to logistics or administrative policy changes may still produce temporary deviations from the projected path.
Disclaimer: This publication is intended solely for analytical purposes and does not constitute investment or financial advice. Actual inflation outcomes may differ from the forecast due to supply shocks, changes in tariff policy, external price developments or shifts in the behavior of economic agents.
National Bureau of Economic Research specifically for EconomyKZ.org
