Finland’s economic forecasts have repeatedly failed to predict how the country’s economy would actually perform, according to a new study by the Kalevi Sorsa Foundation. The findings raise a wider question for Finnish policymakers: how much should government decisions depend on forecasts when those forecasts can be significantly wrong?
The study looked at economic forecasts from some of Finland’s most important forecasting institutions, including the Ministry of Finance, Bank of Finland, Etla, Pellervo Economic Research, Labore, Nordea and OP Pohjola.
Researchers compared what these institutions predicted with what eventually happened to Finland’s economy. They found that spring forecasts for the following year’s economic growth were off by an average of 1.5 percentage points between 2016 and 2025.
That is a significant gap when compared with Finland’s actual economic performance. The country’s economy grew by an average of only about 1 percent a year during the same period.
In simple terms, the forecasts were often missing the size and direction of changes in the economy by more than the economy was growing in the first place.
One of the more striking findings is that the institutions were not simply making different mistakes.
Their forecasts were often similar, and they frequently moved in the same direction. When economic conditions changed unexpectedly, several institutions tended to miss the turning point at roughly the same time.
Lauri Finér, executive director of the Kalevi Sorsa Foundation and one of the study’s authors, said the institutions had broadly similar levels of forecasting error and often made mistakes in the same direction.
That matters because it suggests that the problem may not be limited to one institution or one forecasting model. There may be wider problems in the way economic forecasting is done.
Economic forecasts are not guesses made at random. They are built using large amounts of data, economic models and assumptions about what will happen in Finland and abroad. But even a sophisticated model can produce a poor forecast if the information going into it is incomplete or later turns out to be wrong.
This is particularly important for Finland, whose economy is strongly affected by international trade, interest rates, energy prices, industrial demand and developments in major European economies. A sudden change in any of these areas can quickly make an earlier forecast look outdated.
One of the most important points in the study is that forecasters are not always starting with a completely accurate picture of the economy.
Statistics Finland publishes early estimates of economic growth. Those figures are later revised as more information becomes available.
The Kalevi Sorsa Foundation study found that the first GDP estimate differed from the final confirmed figure by an average of 0.7 percentage points between 2011 and 2025.
That is a large difference considering that Finland’s average annual economic growth over the same period was also only 0.7 percent.
This creates a difficult situation for forecasters. They are being asked to predict what will happen next while the figures describing what has already happened may still change considerably.
Jussi Systa, an economic policy specialist at the foundation and co-author of the study, argued that forecasts are often built on economic statistics that have not yet been confirmed.
The problem is easy to understand. If the starting point is wrong, even a well-built forecast can move in the wrong direction.
The revisions to Finland’s GDP figures for 2024 and 2025 provide a useful example.
Earlier data indicated that Finland’s economy had contracted in 2024. Later revisions showed that the economy had actually grown by close to 1 percent.
For ordinary people, the difference between a shrinking economy and an economy that is growing may seem technical. For government, businesses and investors, it is not.
A recession can lead policymakers to consider support measures, spending cuts or other responses. A growing economy may require a different approach.
When the underlying economic data changes significantly, decisions based on the earlier figures can also look very different in hindsight.
The study gives particular attention to the Ministry of Finance because its forecasts have a direct role in the government’s budget planning.
Researchers examined the ministry’s forecasts over a longer period, from 2011 to 2025.
During those years, the ministry’s spring forecasts made one year before the economic outcome predicted average growth of 1.7 percent. Actual average growth was only 0.7 percent.
The difference is substantial.
On average, the ministry was expecting economic growth at more than twice the rate that was eventually recorded.
This does not mean that the Ministry of Finance deliberately overstated growth. Economic forecasting is inherently uncertain, and unexpected events can change economic conditions quickly.
But the size of the gap becomes important when forecasts are used to make decisions involving billions of euros in public money.
The study also examined forecasts for Finland’s public deficit.
Two-year forecasts for the deficit as a share of GDP were wrong by an average of 2.2 percentage points between 2011 and 2025.
That percentage may look small on paper. At the level of national finances, however, it represents a substantial amount of money.
The foundation calculated that an error of this size, applied to the projected size of Finland’s economy in 2031, would correspond to roughly EUR 7.5 billion.
This illustrates why economic forecasting is not simply an academic exercise.
A forecast can influence how much a government believes it can spend, how much revenue it expects to collect and how quickly it thinks public debt can be brought under control.
A forecasting error of several billion euros can therefore have real consequences for public services, taxation and government spending.
Finland’s forecasting institutions do not simply pick a growth figure and publish it.
They usually begin by looking at previous forecasts and then update their assumptions based on changes in the global economy, Finnish economic policy and other conditions.
They examine different parts of the economy separately. Household consumption, investment, exports, imports, employment, government spending and other factors are assessed before the information is brought together into a wider economic model.
The method is detailed, but the final result is still dependent on assumptions.
If interest rates move differently than expected, exports fall, consumer confidence weakens or international demand changes, the forecast can quickly become less reliable.
The study also points to another problem. Forecasts often assume that the economy will eventually move back towards its historical growth rate.
That assumption can be reasonable in some circumstances. But if Finland is facing long-term problems such as weak productivity, an ageing population, labour shortages or structural changes in major industries, simply expecting the economy to return to its old growth pattern can produce overly optimistic forecasts.
The researchers are not calling for Finland to stop producing economic forecasts.
That would not be realistic.
Governments need some idea of what could happen in the future when they prepare budgets, plan public services and make decisions about taxes and investment.
The bigger issue is how much confidence policymakers place in those forecasts.
Jussi Systa has argued that forecasts are necessary because governments have to make decisions about the future. The problem comes when the uncertainty surrounding those forecasts is not properly reflected in policy decisions.
That distinction is important.
A forecast should be treated as an estimate with a margin of uncertainty, not as a fixed description of what the economy will look like.
If a government plans a major spending programme or a large fiscal adjustment based on one economic projection, it needs to understand what happens if that projection turns out to be wrong.
The issue has become more important because economic forecasts are deeply connected to Finland’s fiscal policy.
Government budgets and programmes rely on Ministry of Finance forecasts when estimating tax revenue, spending needs and the development of public debt.
Forecasts can also influence decisions about research and development funding and annual adjustments to income taxation.
Their importance is expected to increase further under Finland’s new fiscal framework and planned debt brake.
A preliminary parliamentary agreement reached in February 2026 for EUR 8 billion to EUR 11 billion in public-finance adjustment measures was calculated using the Ministry of Finance’s economic forecast.
That makes the reliability of the forecast more than a technical question for economists.
If the economic outlook is weaker than expected, government revenue can fall while spending pressures increase. If growth is stronger than expected, the fiscal situation can look better.
Either way, the assumptions used at the beginning can influence decisions for years.
The study recommends that policymakers pay greater attention to economic indicators that become available faster and depend less on long-term modelling assumptions.
Real-time tax information is one example.
Tax data can provide a relatively quick indication of what is happening with economic activity because it reflects actual transactions and income rather than only projections.
Employment statistics can also provide useful information about the direction of the economy.
These indicators will not replace economic forecasts. They serve a different purpose.
Forecasts attempt to estimate what may happen in the future. Real-time indicators can help policymakers understand what is happening now.
Using both could give decision-makers a better picture of the economy and make it easier to respond when conditions change.
The central lesson from the Kalevi Sorsa Foundation study is not that economic forecasting is useless.
Forecasting remains an important part of running a modern economy. No government can make serious long-term decisions without considering what might happen in the years ahead.
The problem begins when a forecast is treated as more certain than it really is.
Finland’s experience between 2011 and 2025 shows why caution is necessary. Economic growth was frequently overestimated, forecasts often missed turning points and even the initial data describing past economic performance was later revised.
That does not make forecasting a failure. It shows its limits.
For Finland, the better approach may be to build economic policy around a wider set of evidence. Forecasts can provide one view of the future, while tax receipts, employment data, business activity and other real-time indicators can show what is happening on the ground.
That would give policymakers more room to adjust when reality does not follow the forecast.
For a country facing difficult questions about public debt, economic growth and government spending, that flexibility could become increasingly important.
