A new IMF analytical paper argues that Georgia’s international reserves, while now broadly adequate by standard benchmarks, remain below the level required to fully insure the economy against external shocks. The report, “Reserve Adequacy in Georgia: How Much is Enough?”, concludes that the country’s optimal reserve level is significantly higher than conventional IMF metrics, estimating a range of 145–150% of the ARA (Assessing Reserve Adequacy) metric—well above the current level of around 100%.
The IMF notes that Georgia’s reserve position has improved markedly in 2025, after several years of falling below adequacy thresholds and experiencing pressure during the October 2024 parliamentary elections. By the end of 2025, reserves had recovered to the 100% ARA level, supported by strong services export inflows, higher gold prices, and continued de-dollarization in the economy. This accumulation, according to the report, helped restore market confidence and contributed to a successful Eurobond refinancing in early 2026. Despite recent external pressures—including higher energy prices linked to geopolitical tensions—the exchange rate has remained broadly stable, reflecting stronger buffers and improved confidence in the lari.
However, the IMF stresses that conventional reserve adequacy metrics do not fully capture Georgia’s country-specific vulnerabilities. Standard frameworks typically focus on short-term external liabilities, but the IMF expands its analysis using a cost-benefit approach that incorporates three additional channels: sovereign risk premia, private-sector dollarization, and the need for foreign exchange intervention (FXI) in a relatively shallow FX market. Together, these factors significantly increase the value of holding reserves beyond traditional thresholds.
In the baseline model, reserves serve as a form of self-insurance against sudden stops in capital inflows, allowing consumption smoothing during crises. Under this standard framework, the IMF estimates Georgia’s optimal reserves at around 130% of ARA, assuming a 10% probability of a sudden stop. However, this baseline excludes key structural features of the Georgian economy, particularly its sensitivity to external financing conditions and high levels of foreign currency usage in the private sector.
The first major extension of the model introduces sovereign risk premia. The IMF finds that higher reserves reduce sovereign borrowing costs by improving investor perceptions of liquidity and repayment capacity during stress periods. Empirical literature suggests that a 10 percentage point increase in reserve-to-GDP ratios can reduce emerging market spreads by 30–70 basis points. For Georgia, this creates a feedback effect: higher reserves reduce borrowing costs, which in turn lowers the opportunity cost of holding reserves. Depending on crisis probabilities, this channel can modestly increase optimal reserves, although the effect diminishes at higher reserve levels due to diminishing returns.
The second extension focuses on dollarization, which remains structurally important in Georgia. In highly dollarized economies, private agents often hold foreign currency as self-insurance against macroeconomic instability and exchange rate risk. The IMF argues that public reserves can partially substitute for this behavior by increasing confidence in the central bank’s ability to supply foreign exchange during stress. As reserves rise, private-sector demand for FX assets declines, reducing balance sheet mismatches and lowering the severity of potential crises. This channel has the strongest quantitative effect in the report, raising optimal reserves by around 18% of ARA at a 10% crisis probability, particularly at lower reserve levels where confidence effects are most pronounced.
The third extension introduces the role of foreign exchange intervention. Because Georgia operates in a relatively shallow FX market, exchange rates can occasionally overshoot fundamentals due to shifts in sentiment, even outside crisis episodes. The National Bank of Georgia (NBG) therefore uses FX intervention to smooth volatility. However, such interventions can significantly reduce reserve buffers, as seen in the pre-election period of 2024, when around USD 750 million—approximately 10% of ARA—was used to stabilize the lari. The IMF estimates that maintaining sufficient “working liquidity” for FXI requires an additional reserve buffer of up to 10% of ARA, though it cautions that some of this may overlap with crisis-related needs.
When all extensions are combined, the IMF concludes that Georgia’s optimal reserve level rises materially to around 145–150% of ARA. At this level, reserves provide not only crisis insurance but also broader macro-financial stability benefits, including lower borrowing costs, reduced dollarization risks, and greater capacity for FX intervention. While current reserves at just over 100% of ARA are considered broadly adequate under moderate stress scenarios, they remain below the optimal range implied by higher-risk environments.
The report also notes that the composition of reserves matters for their effectiveness. A portion of the recent increase reflects valuation gains, particularly from higher gold prices, which may not be as liquid or readily deployable in stress situations compared to foreign currency assets. This limits the immediate usability of part of the buffer during external shocks.
Finally, the IMF recommends continued opportunistic reserve accumulation, especially given elevated global uncertainty. It supports the National Bank of Georgia’s current price-based approach to foreign exchange purchases while emphasizing that interventions should remain limited and carefully calibrated. The exchange rate, the report stresses, should continue to function as the primary shock absorber, with reserves acting as a backstop rather than a substitute for adjustment.
Overall, the IMF’s analysis presents a more demanding benchmark for external resilience in Georgia, suggesting that while the country has rebuilt buffers to adequate levels, significant additional accumulation would be required to reach what it defines as the true optimal reserve position.
