First-Digit Test: What Benford’s Law Reveals About Numbers Behind Next Crisis
Russia’s wartime economy has been called a miracle and a mirage, sometimes by the same analyst in the same paragraph. GDP grew at over 4 per cent in 2023–24, then collapsed to 1 per cent in 2025. The government now forecasts 0.4 per cent for 2026. A Kremlin-aligned think tank has declared the banking system in systemic crisis. The defence minister admits actual military spending runs a full percentage point of GDP above the published budget. And a 2024 study by the Bank of Finland’s Institute for Emerging Economies, applying Benford’s first-digit law to 31 Russian economic data series, found that while Rosstat’s core statistics remain broadly reliable, the monthly and quarterly budget data published by the Finance Ministry and the Federal Treasury ceased to follow Benford’s Law after the February 2022 invasion of Ukraine.
[https://en.thebell.io/can-we-trust-russian-economic-statistics/]
That finding should trouble anyone who takes geopolitical risk seriously. Not because it proves fraud — Benford deviations are circumstantial, not conclusive — but because of what happens when official numbers no longer track the probability distributions that govern the natural world. When data is fabricated or politically adjusted, variance is artificially compressed. The tails — the region where crashes, defaults, and strategic miscalculations live — are not eliminated. They are hidden. And hidden tails, as any finance professional knows, do not stay hidden. They detonate.
Benford’s Law, first observed by Simon Newcomb in 1881 and formalised by Frank Benford in 1938, states that in datasets spanning several orders of magnitude the leading digit follows a logarithmic distribution: the digit 1 appears about 30 per cent of the time, while 9 appears under 5 per cent. The pattern holds across river lengths, stock prices, city populations, tax returns, and national accounts. When a dataset deviates significantly from this expected distribution, it raises a specific and testable suspicion: someone has been editing the numbers.
The forensic application is well established in accounting and tax compliance. Its geopolitical application is newer, and the international system has been unconscionably slow to institutionalise it. This is the argument I want to........
