Mapping Unemployment: Spatial Patterns and Clusters in Jakobstad with Finland's Fine-Grained Statistical Data

National or municipal unemployment figures are useful, but they flatten space. Once you look at unemployment on a 250-meter grid, the picture gets more interesting.

In this post I used a sample dataset from Jakobstad (Pietarsaari), based on 2017 data. The case itself is small, but the method scales well. The same approach can be applied much more broadly with Statistics Finland’s regularly updated grid dataset.

Jakobstad works well as a case area because it is small enough to inspect closely, but still large enough to show meaningful variation inside the city.

Why The Grid Matters

Most unemployment analysis stops at municipal or regional units. The RTTK grid database lets you work at 250-meter resolution instead. That makes it possible to see neighborhood-scale differences that disappear in larger administrative averages.

In this sample, the study area covers 394 grid cells, of which 160 contain valid unemployment data.

Figure 1: Overview of unemployment patterns across Finnish 250m grid cells, showing spatial distribution, LISA clusters, and key statistics

One Clear Finding

The most interesting result in this sample is that urban grid cells (defined here as cells with at least 1,500 inhabitants/km²) show clearly higher unemployment than rural ones:

• Urban areas: 12.1% average unemployment (median 10.7%)
• Rural areas: 7.7% average unemployment (median 6.7%)

There is an important caveat here. Statistics Finland suppresses values in cells where the population is too small for privacy reasons. In practice, that means sparse rural areas are underrepresented, so the urban-rural comparison should be read with some caution.

Still, the result is worth looking at. It suggests that the concentration of jobs in urban areas does not automatically translate into lower visible unemployment.

Figure 2: Distribution of unemployment rates showing the contrast between urban and rural areas

Unemployment Is Spatially Clustered

The second main result is that unemployment is not randomly distributed across the city. Moran’s I shows clear positive spatial autocorrelation:

• Moran’s I: 0.248 (p < 0.001)
• Interpretation: Areas with similar unemployment rates cluster together

In plain terms, areas with similar unemployment levels tend to sit near each other.

Interactive Maps

The maps below make that pattern easier to inspect.

Hotspots, Coldspots, and Outliers

Local Indicators of Spatial Association (LISA) analysis identified distinct spatial clusters:

Hotspots (19 cells):

• Average unemployment: 20.1%
• Meaning: High unemployment surrounded by other high-unemployment cells

Coldspots (12 cells):

• Average unemployment: 1.7%
• Meaning: Low unemployment surrounded by other low-unemployment cells

Outliers (9 cells):

• Mixed local patterns where a cell differs clearly from its neighbors

Method In Brief

The workflow was fairly standard:

1. Data Processing: Cleaned Statistics Finland’s RTTK 250m grid data, calculating unemployment rates as unemployed/(unemployed + employed)

2. Urban-Rural Classification: Applied 1,500 inhabitants/km² threshold consistent with Finnish urban planning standards

3. Spatial Weights: Constructed 8-nearest neighbor weights to account for geographic relationships

4. Global Analysis: Computed Moran’s I to test for overall spatial clustering

5. Local Analysis: Applied LISA to identify specific hotspots and coldspots

Why This Matters

If this kind of analysis is used in practice, it can support more targeted discussion around where unemployment problems are concentrated and whether they are isolated or spatially persistent.

For example:

• hotspot areas can be prioritized for closer local investigation
• urban-rural comparisons can be tested instead of assumed
• repeated analysis over time could show whether the same clusters persist

Limits

This is still a small demonstrator, so there are obvious limitations:

• Temporal snapshot: The biggest one, this sample dataset represents 2017 conditions; patterns have most certainly evolved substantially since then.
• Missing data: 22.8% of cells lack employment data, likely rural areas with very low population
• Privacy constraints: Small area statistics subject to data suppression for privacy protection
• Causation vs correlation: Spatial patterns suggest relationships but don’t establish causation

If I continue this analysis, the next steps would be time-series comparison and combining unemployment with other socioeconomic indicators.

Conclusion

The main point is not that Jakobstad is unique. It is that spatial pattern matters, and municipal averages hide a lot.

Once you move to fine-grained spatial data, unemployment stops looking evenly distributed and starts looking local, clustered, and uneven. That is exactly the kind of thing GIS is good at making visible.

That does not solve the policy question on its own, but it gives a much better starting point for it.

Data Source and Acknowledgments

This analysis uses Statistics Finland’s RTTK (Ruututietokanta) 250-meter statistical grid database. This sample dataset was downloaded from an online data portal managed by Esri Finland Oy.

For questions about methodology, please don’t hesitate to contact me.