Questions from
economists and researchers.

Because it proposes a mechanism and makes predictions that can be tested — but, like any theory, it stands or falls on those tests.

A framework organises variables. RCG goes further: it proposes why capital behaves as it does — that it follows conductance (how freely it can flow), not expected return alone — and it predicts where stimulus will end up, how effective it will be, and when capital is likely to leave an economy.

The claim is falsifiable, meaning evidence could show it to be wrong. The main test is a regression — a statistical method for estimating how much one factor explains another — run across 25–30 past stimulus events:

m_i = β₀ + β₁·G_i + β₂·ln(P_i) + β₃·ΔR_i + β₄·X_i + ε_i

Here m_i is the stimulus multiplier (how much output each £1 of stimulus produced), G_i is conductance, P_i the size of the stimulus, ΔR_i the return differential and X_i other controls. RCG predicts β₁ is positive: lower friction, bigger multiplier.

That is correct. The work is a working paper: a proposed framework, preliminary evidence and a specified test.

We see this as a research programme with a clear route to being proved wrong, and would welcome collaborators.

It builds on that literature but makes a narrower, separately testable claim.

Work by Acemoglu & Robinson, Rodrik and North shows that institutions — laws, regulators, norms — matter for a country’s long-run growth. That work is mainly about national, long-run outcomes.

RCG claims something more specific: that the same frictions reduce the effect of each additional pound of stimulus, even within a stable country, and that they vary by sector.

Rajan & Zingales focus on the cost of finance. RCG looks at a wider set of frictions, several of which are not about cost at all.

Rajan & Zingales showed that industries that depend more on outside finance grow faster in countries with deeper financial markets, and explained this through cheaper finance. RCG adds four kinds of friction:

• Regulatory friction — capital is delayed, not just made more expensive
• Information gaps — capital faces uncertainty, not just higher prices
• Market structure barriers — capital is blocked by established firms
• Capture — capital is diverted before it reaches productive use

The test: RCG predicts that friction still explains outcomes after controlling for the cost of finance. If it does not, the cost-of-finance explanation is sufficient.

The two can overlap, but they describe different causes and point to different policies.

Secular stagnation — associated with Summers and with Eggertsson & Mehrotra — describes a long period of weak demand, where people and firms want to save more than they invest. RCG is about transmission: even when demand exists, capital may struggle to reach productive uses.

One possibility we want to test is that weak demand is partly a result of high friction: if capital does not reach productive uses, wages grow slowly, spending weakens and low growth persists.

Those studies are independent findings that pre-date RCG. We cite them as patterns RCG would need to explain, not as proof.

• Ioannou & Wójcik: across 614 cities, finance and growth follow an inverted U — finance helps up to a point, then holds growth back. RCG’s interpretation is that finance first reduces friction, then becomes a source of it.
• Centre for Cities: many UK cities with strong output growth have weak household income growth. RCG’s interpretation is that high local friction stops output gains feeding through to households.

Offering an explanation for existing patterns is a normal step in building a theory. Whether the explanation is right is exactly what the back-test is for.

They are not estimated from data. They are an initial, openly stated choice, which the back-test will either support or lead us to change.

The Omega Index combines the four sources of friction into one score:

Ω = 0.25·RF + 0.20·IO + 0.25·MSB + 0.30·CC

• Regulatory friction (RF), 25% — can be addressed through planning and process reform
• Information gaps (IO), 20% — through transparency and standardisation
• Market structure barriers (MSB), 25% — through competition policy
• Capture (CC), 30% — the highest weight, as it can be addressed most directly through conduct-based incentives

The weights will be judged by how well they predict outcomes rather than by statistical factor analysis. If the back-test shows weak predictive power (for example, β₁ well below 0.3), we will recalibrate.

We propose measuring both from published, auditable indicators rather than judgement.

Information gaps (IO) can be estimated from:

• Bid-ask spreads — the gap between buying and selling prices in financial markets, a standard sign of uncertainty
• The cost of due diligence
• How standardised contracts are
• Market liquidity measures, such as trading volume and depth

Capture (CC) is measured through PRICI, using audited company accounts:

• Price Intensity — price growth compared with the sector median
• Investment Compliance — (capital expenditure + R&D + wage growth) ÷ (operating profit − dividends − buybacks)

The audit rules set out in the PRICI-P specification allow only factual checks against audited financial data, not qualitative judgements.

This is a real concern — economists call it endogeneity — and the back-test is designed to address it.

If struggling sectors attract stimulus because they are struggling, a simple correlation between friction and weak results could be misleading. Three strategies help separate cause from effect:

• Lagged friction: use friction scores from five years before each stimulus. If they predict later results, reverse causality is less likely.
• Neighbouring countries: use friction scores for the same sector in comparable countries. If a sector has high friction everywhere, it is more likely structural than a response to the event.
• Natural experiments: look at regulatory changes that reduced friction in one sector but not another, and check whether the effect of stimulus then improved.

The framework’s credibility depends on these tests, not only on a simple regression.

You shouldn’t treat it as evidence. It is an illustration of how the calculation works.

The example uses documented features of the sector — around 24-month permitting, long grid-connection queues, a concentrated utility market and consultancy costs of 30–40% — to show how an Omega Index score would be built. The method and data sources are set out in the working paper.

Evidence will come from the back-test across many events, not from a single example.

There is a sketch, not yet a full derivation. Developing one is a priority.

Microfoundations means deriving an economy-wide relationship from the decisions of individual people or firms. The sketch is this: an investor chooses between destinations, each offering a return R_i, but reaching destination i involves deployment frictions. If those frictions reduce the realised return proportionally, the investor picks the destination with the highest R_i·G_i — which is the RCG equation.

What is not yet shown is why the frictions should be proportional (multiplicative) rather than a fixed cost (additive), or how investors anticipate friction in advance. We plan to develop this with academic collaborators once the back-test is complete.

Multiplication is the simplest form that captures the core idea: if any term is close to zero, flow is close to zero.

With addition, a very large amount of stimulus could make up for a nearly blocked channel. With multiplication, it cannot. The working paper also sets out alternative forms, and the back-test will compare them:

• Log-linear:ln(F) = α + β₁·ln(P) + β₂·ln(ΔR) + β₃·ln(G) + ε
• Exponential friction:F = P · ΔR · exp(−λΩ)
• Threshold:F = P · ΔR · G if (ΔR · G) > 0, otherwise 0

The central idea — that friction systematically diverts or reduces capital flow — holds under each of these.

They can see it and adjust for it, but seeing a barrier does not remove it.

Friction acts as a discount on the return an investor actually receives. A simple example:

B looks more attractive on paper, but A delivers more once friction is taken into account, so a rational investor chooses A. Investors pricing in friction is therefore consistent with RCG — it is the mechanism by which friction redirects capital.

No. A tax applies to activity regardless of behaviour. PRICI is designed as an incentive that responds to behaviour.

A general tax raises revenue but gives firms little reason to act differently. PRICI targets one specific pattern — high payouts without matching investment — and rewards the alternative. Firms that invest are not penalised. The aim is to change incentives, not to raise revenue.

It is designed to do the opposite: to encourage investment when firms are most inclined to hold back.

In a prolonged slowdown, firms often hold on to cash or return it to shareholders rather than invest. That can be sensible for one firm but damaging for the economy as a whole.

PRICI-Lite, the low-inflation version of the score, drops the pricing element and focuses on payouts versus verified investment (capital expenditure, R&D and wages). It switches on only when median price growth in a sector has been flat or falling (PI ≤ 1.0) for four consecutive quarters, and does not apply during normal growth or inflation.

Through three design features that protect firms which invest.

• Safe harbours: firms with high Investment Compliance (IC above 0.15) are exempt from enforcement
• Clear thresholds: firms whose prices have not risen faster than the sector (PI ≤ 1.0) are exempt from the penalty tier
• Pre-clearance: firms can register planned investment projects in advance so they count towards their score, subject to later verification

The intention is to discourage extraction without reinvestment, not innovation — and the Investment Compliance and Price Intensity components are how that distinction is made in practice.

No. The examples are mostly from OECD countries because the data is easier to obtain, not for theoretical reasons.

Zimbabwe is a useful contrast. During the 2008 hyperinflation, capital flows effectively collapsed. Adopting the US dollar in 2009 was a single, large reduction in monetary friction, and activity recovered quickly — consistent with RCG in a setting very different from the OECD cases (see Further Reading & Working Papers).

The initial back-test is limited to OECD economies for data reasons. Extending it to developing economies is a priority for the next phase.

The framework offers a specific explanation of the middle-income trap, a central problem in development economics.

The middle-income trap describes countries whose growth stalls after an early period of rapid catch-up. RCG explains it as follows:

• As a country grows, wages rise and its low-cost advantage shrinks, so the return premium (ΔR) falls
• At the same time, friction (Ω) often rises, through regulatory complexity, rent-seeking and market concentration
• Capital leaves when the friction-adjusted return at home falls below that of advanced economies — (ΔR · G)_developing < (ΔR · G)_advanced — even if headline returns at home are higher

The implication is that escaping the trap needs both higher returns through innovation and action to stop friction building up.