Skip to main content

Uncertainty, policymaking and the role of research

Type
Speech
Subtype
BIS management speech
Date delivered
16 November 2026
Opening remarks by Mr Alexandre Tombini, Chief Representative of the Office for the Americas of the BIS, Lima, 16 November 2026.

Key messages

Although economic and policy uncertainty has long been a feature of the economic environment, it has risen in recent years. Shocks have come one after another. They now spread faster and further across closely connected economies. They can also interact with vulnerabilities, such as high public debt. For central banks, uncertainty complicates both policymaking and communication. In advanced economies, uncertainty often acts like a demand shock. In emerging market economies, it can act more like a supply shock. Research is essential to improve our understanding of uncertainty and its implications. This research allows us to develop measures of uncertainty, understand its effects and identify what makes economies more resilient in the face of uncertainty. The 2025–26 CCA Research Network Conference brings together different data, methods and perspectives to advance this work.

It is a great pleasure to welcome you to the 2025–26 CCA Research Network Conference on “Heightened economic and policy uncertainty and its implications for the macroeconomy and financial stability”. This event also marks the 16th Annual BIS CCA Research Conference.

I would like to express our sincere thanks to the Central Reserve Bank of Peru (BCRP) for hosting us here in Lima. In particular, I would like to thank Adrián Armas and all those at the BCRP who contributed to making this conference possible. We are pleased to welcome all participants. Thank you for joining us and for sharing your expertise and perspectives in the discussions ahead.

The topic of this conference is very timely. In recent years, we have faced a series of major shocks. These include the pandemic; the war in Ukraine, now in its fifth year; and the rise in US tariffs to levels not seen for a century. This year has also brought conflict in the Middle East. It is therefore not surprising that measures of economic policy uncertainty remain elevated.

Uncertainty has always been part of economic life. But two developments may be changing how shocks affect the economy. First, shocks now spread further and more quickly. Economies are more closely connected than in the past. Production depends on global value chains. Finance is integrated across borders. And information moves instantly.1

Second, several shocks are now occurring at the same time or in close succession. This can make their effects stronger, as the impact of one shock can add to that of another. They are also hitting economies at a time when some macro-financial vulnerabilities are becoming more apparent. One important vulnerability is high public debt. Public debt is close to record levels in many countries. This limits the room that governments have to respond when shocks occur. It can also turn a shock that might otherwise be manageable into a much larger problem. Consider, for example, an energy shock that raises inflation. Higher inflation may then imply that central banks need to keep interest rates high. Higher rates in turn increase the cost of servicing public debt. Concerns about fiscal positions may then unsettle bond markets. This can happen even more quickly today, as price-sensitive non-bank investors hold a growing share of government debt (BIS (2026)).

We can see both forces at work in financial markets today: shocks are spreading quickly, and they are interacting with existing vulnerabilities. Higher oil prices due to the conflict in the Middle East have added to inflation in many countries. At the same time, long-term government bond yields have risen. In several cases, they are at their highest since the turn of the century. In advanced economies, several forces are behind this increase. Markets are concerned about fiscal sustainability and the risk of more persistent inflation. They also seem to be responding to expectations about artificial intelligence (AI). At least at the margin, AI investment may be competing for funds. Or higher expected productivity could also raise equilibrium interest rates.

We may be entering a period in which the forces that shape growth, inflation and interest rates are themselves changing. The fragmentation of the global economy is reshaping trade and investment links. More frequent extreme weather events are disrupting harvests, supply chains and infrastructure. AI may be the most far-reaching of these forces. It is already affecting financial markets, contributing to high valuations and a surge in investment (Tombini (2026)). Yet these valuations could also adjust sharply. AI may reshape labour markets and change the skills required of workers. It also creates new cyber risks. We do not yet know the scale or timing of these effects. But they could emerge quickly.

What does this mean for central banks? Two aspects of their work are directly affected: policy decisions and communication.

Monetary policy works with a lag. Decisions must therefore be based on an assessment of where the economy is heading. Uncertainty makes that assessment more difficult. Central banks face uncertainty about the policies that different authorities could implement. They also face uncertainty about how households, firms and markets will respond. 

The effects of uncertainty are not the same in all economies. In advanced economies, uncertainty often acts like a fall in demand. It reduces both activity and inflation (Leduc and Liu (2016)). Recent BIS research finds a different pattern in Latin America. There, economic policy uncertainty acts more like a supply shock. Output falls, but inflation rises as the currency weakens and risk premia rise (Aguilar et al (2026)). More broadly, in emerging market economies, higher global risk has long been linked to sudden stops in capital flows (Forbes and Warnock (2012)). 

Uncertainty also makes communication more difficult. A BIS survey conducted in 2025, covering 12 central banks, shows how communication practices have changed since the pandemic and the inflation surge that followed (Amaral et al (2025)). Almost all of the surveyed central banks now use scenario analysis regularly. They also find that forward guidance works best when uncertainty is low. When uncertainty is higher, some central banks do not use forward guidance. Others use it to signal possible actions over a shorter horizon or make it conditional on future developments. All report that it is difficult to explain uncertainty to the public, as audiences often focus on central forecasts rather than on the range of possible outcomes. Uncertainty therefore calls for humility from all of us, researchers and policymakers alike, as we must recognise the limits of our ability to foresee how events may unfold.

This is where research comes in. It enables us to measure uncertainty. It deepens our understanding of how uncertainty affects the economy and the decisions of households and firms. And it helps us identify what makes economies more resilient in the face of uncertainty.

Uncertainty cannot be observed directly. The ways in which we measure it have changed over time. Early measures focused on volatility in financial markets. Later measures included indices based on news coverage and forecast errors in economic data. More recently, surveys have asked firms directly about the range of outcomes they expect. Artificial intelligence can also be used to analyse large volumes of text and assess the uncertainty they contain. Because these approaches capture different dimensions of uncertainty, the resulting measures do not always move in tandem.

Research has long examined the macroeconomic effects of uncertainty. It shows that uncertainty weighs on economic activity and can also affect inflation, risk premia and exchange rates. More recently, data from surveys of firms and from experiments have complemented this work. They provide a clearer picture of how and why households and firms respond to uncertainty. Much depends on their incentives. For example, when decisions are costly to reverse, uncertainty increases the incentive to wait. Firms may therefore delay investment and hiring, while households may postpone major purchases. These incentives vary across firms and households, depending on their level of debt, their access to credit and the cost of delaying a decision. Microdata help us understand these differences.

Finally, research helps us identify what makes economies more resilient. The effects of uncertainty seem to be smaller where financial markets are deeper (Carrière-Swallow and Céspedes (2013)). Economies with more policy space are also better able to respond to shocks (Bergant and Forbes (2023)). The experience of Latin American economies illustrates this point. Since the turn of the century, they have managed shocks much better than in the past, even though uncertainty has remained high. Research suggests that this improved resilience largely reflects stronger policy frameworks.2

Rigorous research on uncertainty helps to build and strengthen sound policy frameworks. This is why I am very pleased that the CCA Research Network has taken up such an important topic. The research presented at this conference will help central banks deepen their understanding of uncertainty and its implications for policymaking. The papers use different types of data and a range of methods. They measure uncertainty, examine how it affects the economy and consider the implications for policy.

I am positive that the discussions over the next two days will further advance this important work. I wish you all a very productive and engaging conference. Thank you.

References

Aguilar, A, R Guerra, C Müller and A Tombini (2026): “The macro-financial impact of economic policy uncertainty in Latin America”, BIS Working Papers, no 1324.

Alfaro, L (2026): “From resilience to growth: opportunities for Latin America and the Caribbean”, paper prepared for the ECB Forum on Central Banking, Sintra.

Amaral, E, T Ehlers, I Shim and A Tombini (2025): “Monetary policy decision-making and communication under high uncertainty: insights from a survey of central banks in the Americas and beyond”, in E Amaral, F Avalos and I Shim (eds), BIS Papers, no 163.

Bank for International Settlements (BIS) (2026): “High public debt and shifting financial markets: challenges for central banks”, Annual Economic Report 2026, Chapter II.

Bergant, K and K Forbes (2023): “Policy packages and policy space: lessons from COVID-19”, European Economic Review, vol 158, 104499.

Carrière-Swallow, Y and L F Céspedes (2013): “The impact of uncertainty shocks in emerging economies”, Journal of International Economics, vol 90, no 2.

Forbes, K, J Ha and M A Kose (2026): “Heaven or earth? The evolving role of global shocks for domestic monetary policy”, NBER Working Papers, no 34806. 

Forbes, K and F Warnock (2012): “Capital flow waves: surges, stops, flight, and retrenchment”, Journal of International Economics, vol 88, no 2.

Leduc, S and Z Liu (2016): “Uncertainty shocks are aggregate demand shocks”, Journal of Monetary Economics, vol 82.

Rose, J (2023): “Understanding the speed and size of bank runs in historical comparison”, Federal Reserve Bank of St Louis, On the Economy, 26 May.

Tombini, A (2026): “Navigating financial stability in the age of AI”, speech at the Mexican Banking Association (ABM) Seminar on Financial Stability and Markets, 3 November.

1 Shocks originating abroad now explain about half of the variation in interest rates in advanced economies, far more than before the late 1990s (Forbes et al (2026)). Financial stress also moves faster: in 2023 a bank lost a quarter of its deposits in a single day, whereas an outflow of similar size took 10 days in 1984 (Rose (2023)).
2 Crises in Latin America and the Caribbean were about 70% less frequent in 2000–16 than in the 1980s and 1990s, which Alfaro (2026) attributes to stronger fiscal and monetary frameworks, better supervision and deeper markets.