Inflation, the Tenge Exchange Rate and Non-Fundamental Appreciation: March 2026

In March 2026, inflation in Kazakhstan slowed significantly to 0.6% month-on-month. This is a very encouraging result, and I expect April's figures to be comparable, or perhaps even better.

At the end of 2025, I analyzed the impact of changes in the tenge/US dollar exchange rate (with a one-month lag) on monthly inflation, starting from August 2015. Based on the latest data through November 2025 inclusive, the Pearson correlation coefficient (lag 1) is r = 0.47, which is statistically significant (p < 0.001, 123 observations). This is a strong relationship for macroeconomic data.

The regression pass-through coefficient is 0.08: a 1% depreciation of the tenge adds about 0.08 percentage points to monthly inflation with a one-month lag (R² = 0.22). It is important not to confuse correlation with the pass-through coefficient: they are different quantities. The cumulative effect over three to four months (lags 0–3) is 0.23 percentage points for each percentage point change in the exchange rate. The rapid appreciation of the tenge by roughly 15% (from 550 to 464 tenge per dollar) between October 2025 and April 2026 produces a total disinflationary effect of about 2.0–2.5 percentage points in annual terms. This is exactly what we see in the latest inflation data, even after accounting for the 4-point increase in the VAT rate on 1 January 2026.

Many people believe that the tenge is strengthening because of rising oil prices. I cannot agree with this, and here is why.

First, the oil sector's foreign currency revenue fell sharply because of the Tengiz outage in January 2026, and production has still not recovered. Second, there is a lag between the receipt of export revenue and its effect on the exchange rate: the real effect of any oil price increase will not arrive before May. Third, and this is the key point, the additional foreign currency flowing into the oil sector is practically not converted into tenge and does not reach the market. Taxes from the oil sector accumulate in the National Fund, from which no more than the guaranteed transfer goes to the budget. Foreign and local companies exchange exactly as much currency as they need for operating expenses, and this amount does not change significantly. In this context, rising prices for gold, uranium and copper have had a greater effect on the tenge than oil.

What worries me far more is the non-fundamental nature of the tenge's appreciation. I propose a hypothesis that I consider reasonably well founded, although it requires verification at the balance of payments level: the appreciation of the tenge is driven by a record increase in external debt and by structural restrictions on currency purchases by legal entities.

According to the National Bank of Kazakhstan, as of 1 January 2026 Kazakhstan's external debt reached USD 181.8 billion, an increase of USD 17.2 billion over the year, or 10.4%, the fastest rate of growth in 17 years. The key point is the chronology of this growth. Over the first three quarters of 2025, debt rose moderately, from USD 165.6 billion (January) to USD 171.4 billion (October), an increase of USD 5.8 billion. The fourth quarter, however, brought a jump of USD 10.4 billion at once: this is when the bulk of external borrowing took place. And it was precisely in the fourth quarter of 2025 and the first quarter of 2026 that the tenge strengthened most rapidly, from 550 to 464 tenge per dollar. The timing of the peak in borrowing and the peak in appreciation coincides too closely to be accidental.

The mechanism is as follows: foreign currency raised abroad is converted into tenge on the domestic market, creating additional supply. Demand for foreign currency from legal entities, however, is structurally constrained: a number of currency control mechanisms introduced during the fight against the tenge's depreciation remain in place to this day. The result is an asymmetry: the supply of foreign currency is free, while demand is restricted. This is what leads to overappreciation.

Non-fundamental appreciation carries a direct risk: as soon as the borrowed foreign currency begins to flow out into imports and the dividend payment season arrives, the tenge will come under serious pressure. A return of the exchange rate to KZT 520 per dollar (a moderate scenario, a 12% depreciation) would produce a cumulative inflationary effect of about +2.8 percentage points over a quarter. With current baseline monthly inflation of 0.6%, this means a peak of up to 1.5–1.6% per month, effectively a return to the levels of early 2025 and a threat that annual inflation will break out of single digits. Restraining demand for foreign currency while supply remains free is a symmetrical trap.

Sustainable appreciation of the tenge is possible only through growth in labor productivity, and in no other way. Productivity growth requires investment in productive and tradable sectors, and those in turn require investment lending. To understand the scale of the problem: lending to business in Kazakhstan amounts to about 12% of GDP, many times lower than in countries at a comparable level of development, where this indicator is typically in the range of 40–80% of GDP. There are many questions about Baiterek in this context, but Baiterek does not crowd out bank lending; it compensates for its structural absence. Without lending to the real sector, it is extremely difficult to reduce inflation in the long term: productivity must rise and the supply of goods must expand.

If we analyze Baiterek National Managing Holding in figures to determine its impact on inflation, we see the following.

Overall, the entire portfolio of Baiterek holding as of 31 December 2025 amounts to about KZT 10 trillion. The largest share is mortgage lending, at approximately KZT 4 trillion. Next comes the Industrial Development Fund (IDF), which handles all leasing in the Baiterek system and issues corporate loans: the leasing portfolio is KZT 2.4 trillion, and corporate loans, issued mainly by the Development Bank of Kazakhstan (DBK), amount to KZT 3.0 trillion. Approximately KZT 1 trillion is concentrated in agriculture (mainly KazAgro's "Kendala", working capital).

Annual inflation is affected not by the outstanding stock of debt but by the growth of the portfolio and the sources of that growth. Viewed through this lens, the picture is fundamentally different from the conventional one.

Concessional funding through the Agrarian Credit Corporation (ACC) is effectively not growing: the portfolio remains at around KZT 1 trillion, and adjusted for annual inflation, the ACC's real portfolio is shrinking. From an inflation perspective this matters: financing for agriculture expands the supply of domestic food and in the long term works as an anti-inflationary instrument. A decline in the ACC's real portfolio is a risk to food supply, not to inflationary demand.

The IDF and DBK provided the bulk of the growth in corporate financing in the Baiterek system, approximately KZT 1.5 trillion, mainly through external borrowing, and it was directed to investment projects. This is a critically important point: investment loans mostly go toward purchasing equipment from abroad, and where there is a corresponding foreign currency funding source, they produce practically no inflationary impulse. In 2025, DBK placed Eurobonds worth USD 500 million and raised loans from the EIB and Banco Santander, which is precisely the model in which the volume of external borrowing matches the volume of imported equipment.

In 2025, Damu Fund made a dramatic change to its business model: it abandoned subsidies for all new loans, effectively raised the rate to 12.6% (about 13.5% per annum including guarantees), and switched to guaranteeing loans to business at market rates. More than doubling the rate on concessional loans is a direct anti-inflationary measure. Taken together, the impact of concessional lending in the Baiterek system in 2025–2026 has been minimized.

Otbasy Bank added about KZT 0.5 trillion (+15%) to its portfolio, roughly in line with inflation. At the same time, mortgage lending finances construction with a high share of domestic labor and materials, and the supply of housing is inelastic in the short term, so this channel requires separate monitoring. Otbasy Bank accounted for all of the growth in the country's mortgage market in 2025: its share reached 51%, while commercial banks effectively withdrew from the mortgage market after the base rate was raised to 18%.

Thus, in 2025 Baiterek holding had, if anything, an anti-inflationary effect on the economy, while increasing corporate financing by approximately KZT 1.5 trillion, which in the long term should benefit labor productivity and the expansion of the supply of goods.

At the same time, as I have written many times, more than a third of banking profit in Kazakhstan is generated not by taking credit risk but by regulatory distortions. In 2024, regulatory arbitrage amounted to KZT 912 billion, or 36% of the sector's total profit of KZT 2,555 billion (in 2023 it was 43%, or KZT 940 billion out of KZT 2,183 billion). As long as risk-free income on this scale exists, second-tier banks have no economic incentive to lend to the real sector. This, rather than a lack of liquidity, is the main structural barrier to growth in labor productivity and, consequently, to a sustainable reduction in inflation.