Opinion by Leonardo Badea, First Deputy Governor of the National Bank of Romania
Generally, when discussing price developments in a broad sense, the public thinks of inflation developments, which denote the annual change in the consumer price index (or the harmonized index of consumer prices – HICP, according to the European methodology). Thus, the consumer price index (simple or harmonized) reflects the price developments for a representative basket of goods and services, weighted according to the structure of household consumption. Another index for price developments in the economy is the GDP deflator, which reflects how the prices for all domestically produced goods and services, referring here to consumption, investment, government spending or exports, change over a period, usually a year. On the other hand, imports directly affect the level of the HICP, but not that of the GDP deflator. I will give a simple example in this regard. The increase in the price of a domestically produced loaf of bread affects both the HICP and the GDP deflator. However, an increase in the prices of the machinery required to produce bread affects the GDP deflator only if this machinery is produced domestically. If the bread discussed earlier is imported rather than domestically produced, its price will affect only the HICP.
At the international level, it was empirically noticed that periods in which the GDP deflator growth rate was higher than the HICP based inflation are associated with increases in the prices of domestically produced goods and services, including investments and exports, against the backdrop of tensions in supply chains, rising energy costs, and wage adjustments. In contrast, in economies where inflation was predominantly driven by price shocks to imported goods, the HICP dynamics temporarily exceeded that of the GDP deflator, reflecting the direct transmission of external inflation to consumers.
In the figure below we find the annual change in the harmonised index of consumer prices or the inflation rate (according to Eurostat statistics, the annual average of the monthly HICP index is reported, so its change represents the average annual inflation rate), respectively that of the GDP deflator. As can be seen from a simple visual inspection, the annual dynamics for the two indicators were very close in certain periods of time, but there were also periods characterized by significant decoupling. The correlation between the annual change in the HICP and that of the GDP deflator for the period 2005-2025 is approximately 72.3%, therefore a relatively high one. We note that in the period before the financial crisis of 2008-2009, the annual dynamics of the GDP deflator were systematically and significantly above that of the HICP. If we look at the post-crisis period, i.e. the period 2012-2025, the correlation between the two dynamics increases to around 85.4%. On the other hand, if we exclude the period after the outbreak of the pandemic crisis and consider the period 2012-2019, the correlation between the annual change in the HICP and the GDP deflator decreases to around 67.1%. Therefore, the link between the annual change in the HICP and the GDP deflator depends on elements such as the phases of the economic and financial cycles, the nature of the shocks affecting the economy as a whole, or different measures in terms of monetary and fiscal policy. An annual change in the GDP deflator above that of the HICP translates into the fact that prices in the whole economy are increasing faster than those in the consumer basket.
Therefore, the link between the annual change in the HICP and the GDP deflator depends on elements such as the phases of the economic and financial cycles, the nature of the shocks affecting the economy, or different measures in terms of monetary and fiscal policy. An annual change in the GDP deflator above that of the HICP translates into the fact that prices in the whole economy are increasing faster than those in the consumer basket. Also, if we look at the evolution of these indices, given the current international context, we can appreciate that the relationship between dynamics of the harmonized index of consumer prices and those of the GDP deflator has become increasingly dependent on the nature of global shocks and the positioning of economies shaped by the developments occurred within the international value chains. Thus, certainly, episodes characterized by large external shocks such as the global financial crisis, the pandemic, or the energy crisis have amplified the differences between the two indicators, especially in economies with a high degree of trade openness and a significant share of imports in final consumption.
We can thus conclude that the differential between the HICP and the GDP deflator provides relevant information on domestic inflationary pressures relative to imported ones, but also on macroeconomic adjustment mechanisms, providing important information regarding the national income and public debt sustainability channels.
Depending on the nature of the factors driving their evolution, HICP-based inflation and changes in the GDP deflator may have different implications for macroeconomic policies. More precisely, headline inflation represents the compass of monetary policy, particularly in countries that explicitly adopt an inflation-targeting regime. Taylor-type rules, used in a stricter or more flexible form to guide monetary policy, focus directly on the deviation of headline (or policy -relevant) inflation from the inflation target.
At the same time, in benchmark models employed in academia, by central bankers, and by analysts, the standard representation of an economy consists of a relationship describing aggregate demand, another describing aggregate supply—where the key variable is HICP-based inflation—and a Taylor rule to characterize the behavior of the monetary authority. Such a representation can be found in the article The Science of Monetary Policy: A New Keynesian Perspective (1999), authored by the renowned economists Richard Clarida, Jordi Galí, and Mark Gertler, which represents one of the foundational contributions to modern theoretical and applied monetary policy.
By contrast, if we shift our attention to fiscal policy, the key variable becomes the GDP deflator. More specifically, in the quantitative relationships used to monitor public debt sustainability—such as those employed by the International Monetary Fund—the essential element for the evolution of public debt is the ‘r–g’ differential, where ‘r’ denotes the implicit or effective cost of public debt and ‘g’ represents the nominal GDP growth rate. Furthermore, the nominal GDP growth rate is influenced by inflation measured through the GDP deflator. In other words, an increase in the prices of domestically produced goods leads to an increase in nominal GDP, thereby easing the public debt burden, given that IMF-type debt sustainability analyses focus on the ratio of public debt to nominal GDP.
Romania’s example in 2025 may be illustrative in this regard. In 2025, nominal economic growth amounted to 8.62% (according to AMECO data), while real economic growth was only 0.72%. Thus, when assessing recent economic performance, we observe a very low real growth rate—below 1% – which points toward a situation approaching stagflation. On the other hand, nominal growth was relatively high, close to 9%.
In this context, the implications for monetary and fiscal policy differ. From the perspective of the Clarida, Galí, and Gertler (1999) framework, as well as other New Keynesian models, the aggregate demand equation refers to the evolution of real GDP (or the output gap based on real GDP), while the aggregate supply relationship and the monetary authority’s reaction function, described by a Taylor-type rule, respond to developments in real economic activity. By contrast, a substantial increase in nominal GDP mitigates the rise in the public debt ratio by affecting the denominator of the debt-to-GDP ratio. Therefore, in situations characterized by modest real economic growth alongside strong nominal growth (driven by price effects), the implications for monetary and fiscal policy can be significantly different, or at least highly nuanced.
A scenario – whether counterfactual or not – in which HICP-based inflation diverges from inflation measured through the GDP deflator may arise in the context of a substantial increase in the price of imported energy, combined with a decline in real economic activity. In such a scenario, higher energy import prices would exert upward pressure on HICP, but would not directly enter the GDP deflator, except insofar as they are passed through into domestically produced prices. The resulting rise in inflation would reduce purchasing power, which in turn could lower fiscal revenues, subsequently increasing the budget deficit and public debt. Under these circumstances, the absence of direct effects on the GDP deflator would not contribute, through this specific channel, to an increase in the debt-to-nominal-GDP ratio via changes in the denominator. Naturally, this scenario isolates the effects of an energy shock, abstracting from other contemporaneous shocks affecting macroeconomic balances.
Although the difference between HICP inflation and inflation measured by the GDP deflator has received limited attention in the literature, it is evident that in emerging economies—small, open, and financially vulnerable—such as Romania and other Central and Eastern European countries, the two macroeconomic variables may diverge significantly during certain periods. In such circumstances, the implications for macroeconomic policies may differ. Moreover, in scientific approaches that analyze the interaction between monetary and fiscal policy—particularly within the framework of the Fiscal Theory of the Price Level, as developed in works such as those by Christopher Sims (2011) or John Cochrane (2026)—the differential between HICP-based inflation and GDP-deflator-based inflation may also have implications for the results and their associated conclusions.
Moreover, Ricardo Reis, a prominent economist whose influential work focuses primarily on monetary and fiscal policy, highlights such concerns in his paper Which r-star, public bonds or private investment? Measurement and policy implications (2022), and, to some extent, in The Constraint on Public Debt when r < g but g < m (2021).
In conclusion, we can say that the nature of the differential between the HICP and the GDP deflator provides a key to understanding the content of the macroeconomic adjustment required in a global environment characterized by persistent shocks.