The demand side of the programme rests on Rightmove property-search enquiries, aggregated into a family of origin-to-destination search-flow indices for each travel-to-work area. The analysis input is the enquiry itself; the nine indices are constructed from those enquiries.
Variables
- Enquiry record (input): each enquiry gives the enquirer’s location, the property’s location and type, the tenure, and the date. These are the raw inputs from which every index below is built; the released area-period aggregate summarises them as search volume.
- Search volume: the total number of enquiries originating in the area.
- Outflow: the number of cross-boundary enquiries an area’s residents make about property elsewhere.
- Inflow: the number of enquiries about the area’s property that come from residents of other areas.
- Retained: the number of enquiries in which the searcher and the property both sit inside the area.
- Leaving rate: the share of residents’ own search that points outside the area, from 0 to 1.
- Net-flow: whether more search arrives than departs, running from −1 (a pure source) to +1 (a pure sink).
- Openness: the share of all the area’s search that crosses its boundary in either direction.
- Reach: the mean distance, in kilometres, that residents look when they search beyond home.
- Concentration: a Herfindahl index from 0 to 1 recording whether residents’ outward search is spread widely or funnelled into a few destinations.
- Search volume: the total number of enquiries originating in the area.
Data source
Rightmove property-search enquiry micro-data, about 294 million enquiries from 2010 to 2025. Each row is one enquiry, giving the enquirer’s location, the property’s location and type, the tenure, and the date. The enquirer’s own area is resolved from a postcode-to-area lookup built from the data. Enquirer-location coverage is high from 2018 and thinner before; England is the reliable canvas.
Summary statistics of the analysis variables
| Variable | Role | Mean | SD | Minimum | Maximum |
|---|---|---|---|---|---|
| Search volume (enquiries) | input | 67,843 | 233,532 | 32 | 5,235,274 |
| Outflow (enquiries) | constructed | 22,237 | 54,202 | 32 | 1,123,909 |
| Inflow (enquiries) | constructed | 22,237 | 40,493 | 0 | 845,958 |
| Retained (enquiries) | constructed | 45,606 | 182,911 | 0 | 4,184,455 |
| Leaving rate | constructed | 0.444 | 0.159 | 0.145 | 1.000 |
| Net-flow | constructed | 0.021 | 0.244 | -1.000 | 0.660 |
| Openness | constructed | 0.616 | 0.132 | 0.251 | 1.000 |
| Reach (km) | constructed | 113.3 | 60.7 | 31.7 | 748.1 |
| Concentration (HHI) | constructed | 0.116 | 0.070 | 0.025 | 0.558 |
Grain: one observation per travel-to-work area, year, and tenure, 2016 to 2025 (both tenures pooled). The only raw input in this released aggregate is search volume; the analysis inputs proper are the enquiry-level micro-data (about 294 million records), from which the eight indices are constructed.
Method
Counts are aggregated first and ratios formed from the totals, never by averaging area-level ratios. Distances use the great-circle distance between area centroids. The gravity distance-decay elasticity is estimated with origin and destination fixed effects on a balanced panel of corridors.