Under an inflation-targeting regime, a central bank has to look at inflation more broadly than just consumer price data. Behind rising retail prices there are often deeper processes – at the level of production, raw materials, and intermediate goods. The key question is how directly higher producer costs are passed on to households. This question is addressed in the economic study “The Impact of Producer Prices on Consumer Inflation in Kazakhstan”prepared by Erzhan Islam and Altyn Kulkayeva at the National Bank of Kazakhstan.
The authors show that a link between producer prices and consumer inflation in Kazakhstan does exist. However, this link is neither linear nor stable over time. The main conclusion is clear: the overall Producer Price Index (PPI) is not a reliable leading indicator of inflation. Narrower indices – especially in manufacturing and food production – can play that role, but only during certain periods.
At first glance, the logic seems simple. Higher prices for raw materials and intermediate goods raise production costs. These costs should then be passed on to final prices, increasing inflation. But Kazakhstan’s economy works in a more complex way. The authors show that the structure of the PPI is distorted by the large weight of the mining sector. According to national statistics, about 54% of the PPI in 2021 came from mining and quarrying.
These prices are closely linked to global markets and oil prices, but they have little connection to goods and services in the consumer basket. As a result, causality tests do not find a stable link between the overall PPI and the Consumer Price Index (CPI). Price growth in mining may never reach consumers at all.
When the analysis moves from broad aggregates to narrower indices, the picture changes. The econometric results show that:
- producer prices in manufacturing act as a leading indicator for consumer prices;
- producer prices for food products affect food inflation;
- in some periods, producer prices for consumer goods depend on CPI movements, pointing to demand-driven inflation.
This means that inflation in Kazakhstan is formed through different channels at different times, and the direction of causality can change.
Using VAR models and Granger causality tests, the authors find that:
- there is no causal link between the overall PPI and CPI;
- manufacturing PPI leads CPI;
- food PPI leads food inflation;
- feedback from CPI to producer prices for consumer goods reflects demand pressure.
At the same time, classical methods do not identify stable two-way relationships. This is an important result. Standard models capture only part of the real dynamics and fail to reflect changes over time.
Traditional tests use the full sample and assume stable relationships. But inflation rarely behaves that way. Between 2011 and 2021, Kazakhstan experienced currency devaluations, periods of price controls, the pandemic, and strong external shocks. To capture changing relationships, the authors apply wavelet analysis – a tool that examines both time and frequency at the same time.
Wavelet analysis reveals a very different picture. The relationship between producer prices and consumer inflation:
- appears and disappears;
- changes direction;
- depends on the time horizon.
In the short term (up to 6–8 months), producer prices may lead inflation. In the medium term (16–32 months), the situation often reverses: consumer prices begin to influence producer prices. This reflects how businesses adjust their costs to an inflationary environment. This pattern is especially clear in manufacturing and consumer goods.
The period after 2018 deserves special attention. Despite a weaker tenge and rising producer prices, inflation slowed. Wavelet analysis shows a weakening link between PPI and CPI. The authors explain this by falling profitability. Producers and retailers temporarily absorbed higher costs by reducing margins. During this period, supply-side inflation pressure remained limited, even as producer prices increased. Later, other factors became more important: tariff policy, the pandemic, pent-up demand, and the global price shock of 2021.
The food segment is particularly interesting. Kazakhstan covers a large share of its food demand through domestic production. This strengthens the link between producer prices for food and food inflation. Classical tests show that food PPI leads food CPI. Wavelet analysis confirms this relationship but also shows that it is not permanent. In the medium term, demand-driven inflation often becomes dominant.
The main practical conclusion is straightforward: no single producer price index can be used as a universal inflation signal. For inflation analysis and forecasting, the National Bank needs to:
- consider the structure of the PPI;
- separate mining from manufacturing;
- recognize that relationships change over time;
- combine classical econometrics with time-frequency methods.
Inflation in Kazakhstan is not a simple cost pass-through. It is a process in which different factors dominate at different stages.
The study by Islam and Kulkayeva shows a mature analytical approach. Instead of simple conclusions, it offers a layered and nuanced analysis. Producer prices do matter for understanding inflation – but only when properly decomposed and analyzed over time.
For inflation targeting, the key lesson is clear: the quality of analysis matters more than simple indicators.
Sultan Valikhanov, expert of the EconomyKZ.org portal


