Network analysis of inter-industry payment flows reveals structural economic relationships invisible to traditional bilateral measurement approaches, with significant implications for real-time economic monitoring. Analysing 532,346 UK payment records (2017--2024) across 89 industry sectors, we demonstrate that graph-theoretic features which include centrality measures and clustering coefficients improve payment flow forecasting by 8.8 percentage points beyond traditional time-series methods. Critically, network features prove most valuable during economic disruptions: during the COVID-19 pandemic, when traditional forecasting accuracy collapsed (R2} falling from 0.38 to 0.19), network-enhanced models maintained substantially better performance, with network contributions reaching +13.8 percentage points. The analysis identifies Financial Services, Wholesale Trade, and Professional Services as structurally central industries whose network positions indicate systemic importance beyond their transaction volumes. Network density increased 12.5% over the sample period, with visible disruption during 2020 followed by recovery exceeding pre-pandemic integration levels. These findings suggest payment network monitoring could enhance official statistics production by providing leading indicators of structural economic change and improving nowcasting accuracy during periods when traditional temporal patterns prove unreliable.
Figures & tables
Rank
Industry (SIC)
Volume (£T)
Share (%)
1
Financial Services (64-66)
4.18
18.9
2
Wholesale Trade (46)
3.07
13.9
3
Manufacturing (10-33)
2.43
11.0
4
Real Estate Activities (68)
1.77
8.0
5
Professional Services (69-71)
1.52
6.9
6
Retail Trade (47)
1.28
5.8
Table 1: Top 10 Industries by Inter-Industry Payment Volume (2017-2024)
Feature Set
R 2
RMSE (%)
MAE (%)
vs. Traditional
Traditional Features Only
0.324 ± 0.028
6.89
4.85
Baseline
Network Features Only
0.298 ± 0.031
7.02
4.96
-2.6 pp
Combined (Network + Traditional)
0.412 ± 0.026
6.42
4.43
+8.8 pp
Diebold-Mariano Test
DM = 3.42
p = 0.0008
Table 2: Network-Enhanced Model Performance Comparison
Period
Traditional R 2
Enhanced R 2
Improvement
Pre-Pandemic (2017-2019)
0.378
0.438
+6.0 pp
Pandemic (2020-2021)
0.186
0.324
+13.8 pp
Recovery (2022-2024)
0.342
0.421
+7.9 pp
Full Sample
0.324
0.412
+8.8 pp
Table 3: Network Performance Improvements by Economic Period
Figure 1: Network Topology Evolution of UK Inter-Industry Payment Flows (2017-2024). Each panel displays the payment network structure for a given year, with nodes representing industry sectors coloured by economic category: Financial & Business (red), Manufacturing (blue), Trade & Distribution (green), Public & Social (orange), Primary Industries (purple), and Other Services (grey). Network density increased systematically from 0.689 (2017) to 0.775 (2024), indicating growing economic integration. The COVID-19 disruption is visible in 2020, where density temporarily fell to 0.693 before recovery exceeded pre-pandemic levels. Average path length decreased from 1.31 to 1.22, suggesting increasingly direct inter-industry relationships.
Year
Density
Edges
Avg Path Length
Clustering
2017
0.689
5,321
1.31
0.72
2019
0.725
5,595
1.28
0.74
2020
0.693
5,312
1.31
0.71
2022
0.736
5,486
1.26
0.74
2024
0.775
5,767
1.22
0.76
Change 2017-2024
+12.5%
+8.4%
-6.9%
+5.6%
Table 4: Evolution of UK Payment Network Structure (2017-2024)
Financial forecasting uses market, company, and economic data to predict prices, returns, volatility, and business outcomes for decisions and risk management.
Sol Price School of Public Policy, University of Southern California, Los Angeles, USA · Boston University, Boston, USA · Johns Hopkins University, Washington, USA