Tonga Gets New IMF Forecasting Model to Tackle External and Domestic Economic Shocks

In a groundbreaking initiative led by the IMF’s Institute for Capacity Development, in collaboration with the Monetary and Capital Markets and Research Departments, a new macroeconomic forecasting tool has been crafted to meet the specific needs of the Kingdom of Tonga. The working paper titled “A Quarterly Projection Model for Tonga”, authored by Sam Ouliaris, Celine Rochon, and Daniel Taumoepeau unveils a customized Quarterly Projection Model (QPM) that simulates how Tonga’s economy would respond to various macroeconomic shocks. This model, tailored with great precision, provides the National Reserve Bank of Tonga (NRBT) with a robust framework to conduct forward-looking policy analysis and produce more consistent and structured economic forecasts. In an economy highly exposed to global developments, natural disasters, and structural inefficiencies, this model represents a critical advancement in policymaking capacity.

A Small Island Nation with Big Vulnerabilities

Tonga is a small and open Pacific Island economy, heavily dependent on remittances and foreign aid to balance its persistent trade deficits. On average, imports exceed 50 percent of GDP while remittances alone contribute nearly 40 percent. With such strong external reliance, Tonga remains extremely vulnerable to global macroeconomic shocks and climatic events. Compounding this fragility is the fact that Tonga’s domestic financial system is shallow, its monetary transmission mechanism is weak, and its financial markets are underdeveloped. Due to an overabundance of liquidity in the banking system, exceeding 30 percent of GDP, interest rate tools have limited influence. This has led the NRBT to rely primarily on managing the exchange rate, pegging the Tongan Pa’anga to a basket of currencies while retaining the ability to adjust it monthly by up to 5 percent.

Building a Model that Mirrors Tonga’s Economy

To reflect the country’s structural realities, the QPM was calibrated to capture Tonga’s monetary idiosyncrasies, including the predominance of the exchange rate as a policy tool, the persistence of inflation, and the low responsiveness of output to foreign demand. The model integrates a standard gap-based framework covering aggregate demand and supply, an uncovered interest parity (UIP) condition, and a modified Taylor rule. However, unlike traditional models, Tonga’s QPM adjusts for its exchange rate-focused monetary stance and its inflation dynamics, where core inflation is more persistent due to sticky wages in a government-dominated labor force, while food and energy prices are highly volatile and sensitive to external conditions. The model’s Taylor rule gives minimal weight to interest rates, appropriate given Tonga’s near-zero policy rate since 2012 and its inactive interbank market.

The QPM offers a flexible and dynamic platform for NRBT to conduct scenario-based forecasting. Prior to its development, Tonga relied on separate, spreadsheet-based forecasts from the NRBT and Ministry of Finance, often disconnected across sectors and driven by expert judgment. With this model, forecasts can now be integrated, policy shocks quantified, and risk assessments standardized, allowing for clearer communication and more informed decision-making.

What If a Bank Fails?

The model’s first simulation tests the impact of a domestic bank failure, an especially relevant risk given rising non-performing loans after repeated natural disasters and weak business recovery. The simulated shock assumes a persistent rise in the credit risk premium, with the troubled bank holding a significant market share. The result: a 1.25 percent permanent decline in real GDP relative to the baseline over a two-year horizon. The model’s monetary condition index (which incorporates both real interest rate and exchange rate effects) becomes tighter, leading to a negative output gap. Although Tonga’s interest rate channel is largely inactive, the exchange rate depreciates, driving up the cost of imports. Inflation, however, remains subdued, a consequence of the high persistence built into the core inflation framework.

Surviving a Natural Disaster

The second scenario models a typical natural disaster of the kind Tonga faces roughly every two years. The shock is designed as a temporary 5 percent decline in potential output and a concurrent staggered rise in inflation, especially in food prices. The initial result is a widening of the output gap and an increase in the policy interest rate within the model’s logic. In practice, though, NRBT’s tools would remain focused on exchange rate adjustments. Notably, the Pa’anga appreciates slightly in this simulation due to aid-driven recovery and higher interest rates. Inflation climbs moderately above baseline in the short term but stabilizes quickly, as domestic food production recovers. The model captures the short, sharp pain of a disaster followed by a quick adjustment period, consistent with Tonga’s history of aid-supported recovery.

The World Turns Against Tonga

In the third and most complex scenario, the model simulates a negative external shock, blending rising global inflation, declining global output, and exchange rate volatility. A one-time global inflation spike in early 2025 is followed by sustained output contraction abroad. These shocks reverberate through Tonga’s economy: initially, the Pa’anga appreciates as foreign inflation rises, suppressing domestic inflation. But this is soon reversed as Tonga’s trading partners raise interest rates to fight inflation, causing the Pa’anga to depreciate. The depreciation lifts the output gap and raises inflation, particularly through imported food and energy prices, which comprise more than half of Tonga’s CPI basket. The model shows how even with capital controls in place, global conditions can penetrate Tonga’s monetary fabric.

A Tool for Policy Preparedness

While the paper cautions that the findings are contingent on calibration assumptions, the simulations illustrate the QPM’s potential for enhancing policymaking in Tonga. It allows for an integrated view of how different shocks affect output, inflation, exchange rate dynamics, and policy responses. The model offers the NRBT a critical analytical edge, replacing fragmented spreadsheets with a rigorous framework grounded in economic theory and tailored to the country’s institutional realities. In doing so, it lays the groundwork for better-informed policy dialogue, more effective forecasting, and a potential blueprint for other Pacific Island nations facing similar vulnerabilities in an increasingly uncertain world.