WP-134 · 3 October 2026
ECONOMICS · POLICY

The Growth Imperative, Tested

A sequel to WP-133. Part I is satire in a conjured Swift’s voice; Part II splits the claim that the economy must grow without end and tests each part by derivation or by the sources read.

Satire in Part I · evidence in Part II · English and Português

Part I. A Modest Proposal for Preventing the Economy from Ever Standing Still, Lest It Be Observed to Be Standing

Satire. The narrator is conjured, yet again, from the Deanery of St Patrick’s. His voice is not ours; he is the target. Part II is the record, with every figure traced to its source and every derivation shown.

It is a settled opinion of this Kingdom that the economy is a bicycle, which if it cease to advance must fall over, and that every rider upon it, being persuaded of this, will pedal harder in proportion as he is told the road has no end. I have examined the bicycle. It is, I find, in several of its parts, a chair.

The debtor is told that his interest requires growth. I have computed the matter, and find that his creditor is satisfied equally by a surplus: a plain portion of the yearly income, carried across the table in the ordinary way, without any need that the table itself be enlarged. The tradesman is told that his rival’s growth requires his own. I find that the shares of a market sum to one, that what the rival gains the tradesman loses, and that nothing in the arithmetic obliges the market to swell. The labourer, I confess, is in a worse case: for the more ably a man works the fewer men are required to do the work, unless they work fewer hours, a remedy proposed by many and adopted by few.

I therefore propose only that the bicycle be observed to be partly a chair, and that those who sit in it be permitted, for once, to choose which of its legs to lengthen. I have no interest in the matter, being dead, and having never ridden a bicycle.

Part II. The record

Abstract

Following WP-132 and WP-133, this paper tests the claim that the economy has a growth imperative: that interest-bearing debt, competition and employment force output to rise without end. We split the claim and test each part by arithmetic or by the sources read. On the public debt, the standard identity d′ = d(1+r)/(1+g) − p shows that a debt ratio can be held constant at zero growth by a primary surplus equal to d(r−g)/(1+g); for the US ratio of 122.59 per cent of income (1) and an illustrative 3 per cent interest rate, that is about 3.7 per cent of income a year. Growth lowers the surplus needed; it is not required by the arithmetic. On private credit, two peer-reviewed model studies find that neither credit creation nor interest charges create a growth imperative in themselves (2; 3). On employment the claim holds as arithmetic: US 2025 productivity grew 2.2 per cent (5), so at constant output labour input falls about 19.6 per cent in a decade unless hours per worker fall. Competition for market share needs no aggregate growth, since shares sum to one. These are results about models and identities, not proofs about how any actual economy behaves. The author’s proposal, to grow what needs to grow and cut back the rest, is not tested here.

How to read this part

Each source is numbered and linked in order of first use and marked in the reference list as read in full, summary page, or not read. Where a figure or a derivation is ours, we say so and show it. The Portuguese text is a translation that needs native review.

1. The claim, split into six

“The economy has a growth imperative: debts carry interest, firms compete for market share, and people need jobs” and its companion “GDP demands growth” contain six claims. G1: interest on public debt forces growth. G2: bank credit and interest charges force growth. G3: employment forces growth. G4: competition between firms forces growth. G5: GDP demands growth. G6 is the author’s proposal, to grow what needs to grow and cut back on all else, which is a proposal and not a claim that can be tested by the arithmetic below.

2. Public debt (G1)

Let d be debt as a share of income, r the interest rate, g the growth rate of income and p the primary surplus (revenue less non-interest spending) as a share of income. Next year’s ratio is d′ = d(1+r)/(1+g) − p. Setting d′ = d gives the surplus that holds the ratio constant: p = d(r−g)/(1+g). At g = 0 this is d × r. The identity is our own statement of standard debt arithmetic; the check is that it reproduces its own constant path, which the verify script runs.

The US ratio of total public debt to GDP was 122.59 per cent in the first quarter of 2026 (1). The table gives the surplus needed to hold that ratio constant at illustrative rates (they are not the US effective interest rate, which we did not read).

Interest rate rGrowth gSurplus p needed, % of income
2%0%2.45%
2%2%0.00%
3%0%3.68%
3%2%1.20%
4%0%4.90%
4%2%2.40%

So G1 fails as a mathematical necessity: at zero growth a constant surplus holds the ratio flat, as the middle line of Figure A shows. What is true is narrower. If r exceeds g and the primary balance is in deficit, the ratio rises without limit (the upper line); if g exceeds r, a ratio can fall even with no surplus (the lower line). Growth therefore reduces the surplus a government must collect. Whether that surplus is politically available is a separate question, and the arithmetic does not answer it.

3% interest, no growth, no surplus3% interest, no growth, surplus of 3.7% of GDP2% interest, 3% growth, no surplus
0%100%200%300%400%500%010203040503% interest, no growth, no surplus3% interest, no growth, no surplus: 537%537%3% interest, no growth, surplus of 3.7% of GDP3% interest, no growth, surplus of 3.7% of GDP: 123%123%2% interest, 3% growth, no surplus2% interest, 3% growth, no surplus: 75%75%Years
Figure A. Debt as a share of income over 50 years, from 122.6% (the US ratio in the first quarter of 2026), under three scenarios. A flat line needs no growth, only a surplus equal to interest on the debt. Scenario rates are illustrative, not the US effective rate. Starting ratio: FRED, series GFDEGDQ188S (reference 1). Paths are our arithmetic from d′ = d(1+r)/(1+g) − p.

3. Private credit and bank money (G2)

A stronger form of the claim is that when banks create money by lending, interest can only be paid by new lending, so the stock of credit must grow. We did not derive this; we read the abstracts of two peer-reviewed model studies. Jackson and Victor (2015), using a system-dynamics model, report that neither credit creation nor the charging of interest on debt creates a growth imperative in and of themselves, and that a transition to a stationary state is feasible (2). Richters and Siemoneit (2017) analyse five post-Keynesian models and conclude that a debt-money system with private banks does not lead to an inherent growth imperative; they find that a stationary economy with positive interest is attainable if household consumption out of wealth exceeds a threshold that rises with the interest rate, that models claiming the opposite contain accounting inconsistencies, and that instability, where it appears, comes from behavioural choices (3). A 2026 paper in the American Journal of Economics and Sociology asks the same question in its title; we could not read it (4). These are models. Their results depend on their assumptions, and they show that a growth imperative is not implied by the structure, not that no real economy is driven to grow.

4. Employment (G3)

Output is productivity times labour input, so (1+gY) = (1+gA)(1+gL). In 2025 US nonfarm business productivity rose 2.2 per cent, output 2.6 per cent and hours worked 0.4 per cent, and the 2.2 per cent matches the average since 1947 (5); the three numbers satisfy the identity to rounding. If output is held constant (gY = 0) while productivity grows 2.2 per cent a year, labour input must fall 2.2 per cent a year: about 19.6 per cent in ten years and to about 34 per cent of today’s level in fifty (Figure B, our arithmetic). The same release reports that labour’s share of output reached its lowest level since 1947, 53.8 per cent, which we do not use here but note for the question of who owns the machines. G3 therefore holds as arithmetic for employment at given hours. The escapes are fewer hours per worker or a shift of output toward labour-intensive work; the sources read do not show either happening.

025507510001020304050Labour neededLabour needed: 33.733.780.4 at year 1080.4 at year 10Years
Figure B. Labour needed to produce a constant output when productivity grows 2.2% a year (index, today = 100). After ten years, about 80; after fifty, about 34, unless hours per worker fall or output moves to labour-intensive sectors. Growth rate: US nonfarm business, 2025 (reference 5). Our arithmetic: 100 × (1.022)−t.

5. Competition (G4)

If firms’ market shares si sum to one, then one firm’s gain in share is another’s loss, and competition for share alone does not require the market to grow. That is an identity, not a finding about firms. The related claim that firms must reinvest profits and expand to survive is a different argument, which we did not examine. We therefore mark G4 as not shown.

6. GDP, and what decoupling shows (G5)

GDP is a measure of output, so it cannot demand anything; the claim that it does is a claim about the incentives built around it, treated under G1 to G4 above. The unresolved empirical question is whether output can rise while resource use and emissions fall enough. A study of 200 years of data (1820–2022) finds that episodes of absolute decoupling, with GDP growing while emissions fall, are uncommon and short, the longest lasting seven to eight years, and that 60 per cent of cumulative fossil CO2 reductions in the period came under recessions; under a stricter test, 2.6 per cent growth with emissions falling 3.4 per cent a year, only 13 per cent of reductions qualify (6). A 2019 report for the European Environmental Bureau argues against green growth as a sole strategy (7); we read only its title and publication details. These sources bear on whether growth and a shrinking footprint have gone together in the past; they do not test the proposal in G6.

7. The proposal (G6)

“Grow what needs to grow, and cut back on all else” is a policy proposal. It needs a criterion for what needs to grow and a decider, and the employment arithmetic in section 4 shows a cost it would have to answer: if overall output is cut or held flat, labour input falls with productivity unless working time is shortened. We do not test the proposal. The sections above show only that three of the usual reasons given for why it cannot be done are, respectively, not a mathematical necessity (G1), not shown by the models read (G2) and not shown (G4), and that the fourth (G3) is an arithmetic constraint.

8. Status of each claim

ClaimStatus
G1. Interest on public debt forces growthFails as a mathematical necessity (our derivation); growth lowers the surplus needed
G2. Bank credit and interest force growthNot supported by the two model studies read; model results, not proofs about real economies
G3. Employment forces growthHolds as arithmetic: at 2.2% productivity growth and flat output, labour input falls about 19.6% in ten years unless hours fall
G4. Competition forces growthNot shown: shares sum to one; the reinvestment argument was not examined
G5. GDP demands growthNot shown; GDP is a measure
Absolute decoupling has been rare and shortReported by one study read (summary page); a second report not read
G6. Grow what needs to grow, cut the restThe author’s proposal; no criterion or decider given; not tested
US public debt was 122.59% of GDP in Q1 2026Established (FRED series page read)
US 2025 productivity +2.2%, output +2.6%, hours +0.4%Established (BLS release page read)

9. Limits and open items

The scenario interest rates are illustrative; we did not read the US effective rate or the maturity structure of the debt. The two model studies were read as abstract or summary pages only, not in full, and a third could not be read. The BLS figure covers the nonfarm business sector, not the whole economy. The decoupling literature was sampled by one study and one title, not surveyed. Nothing here tests whether an economy that cuts some output and grows other output would hold together politically or financially. The relation to WP-131, WP-132 and WP-133 is one of form and theme, not proof.

References

Style: ABNT NBR 6023. Bracketed notes give the status of each source.

  1. UNITED STATES. Office of Management and Budget; Federal Reserve Bank of St. Louis. Federal debt: total public debt as percent of gross domestic product (GFDEGDQ188S). St. Louis, updated 25 June 2026. Available at: https://fred.stlouisfed.org/series/GFDEGDQ188S. Accessed: 3 Oct. 2026. [Series page read (Q1 2026: 122.59%); official data via FRED.]
  2. JACKSON, Tim; VICTOR, Peter A. Does credit create a ‘growth imperative’? A quasi-stationary economy with interest-bearing debt. Ecological Economics, [S. l.], v. 120, p. 32–48, 2015. Available at: https://econpapers.repec.org/RePEc:eee:ecolec:v:120:y:2015:i:c:p:32-48. Accessed: 3 Oct. 2026. [Abstract read via a bibliographic summary; full paper not read; peer-reviewed.]
  3. RICHTERS, Oliver; SIEMONEIT, Andreas. Consistency and stability analysis of models of a monetary growth imperative. Ecological Economics, [S. l.], v. 136, p. 114–125, 2017. Available at: https://ideas.repec.org/a/eee/ecolec/v136y2017icp114-125.html. Accessed: 3 Oct. 2026. [Abstract read via a bibliographic summary; full paper not read; peer-reviewed.]
  4. FUDERS [first name not confirmed]. Is there a growth imperative inherent in our financial system? The American Journal of Economics and Sociology, [S. l.], 2026. Available at: https://onlinelibrary.wiley.com/doi/10.1111/ajes.70036?af=R. Accessed: 3 Oct. 2026. [Title only; the page returned an access error (403).]
  5. UNITED STATES. Bureau of Labor Statistics. Productivity and costs, fourth quarter and annual averages 2025. Washington, DC, 5 Mar. 2026. Available at: https://www.bls.gov/news.release/archives/prod2_03052026.htm. Accessed: 3 Oct. 2026. [Release page read; official source; title as given in the search listing.]
  6. INFANTE-AMATE, Juan; TRAVIESO, Emiliano; AGUILERA, Eduardo. Green growth in the mirror of history. Nature Communications, [S. l.], 2025. Available at: https://www.nature.com/articles/s41467-025-58777-4. Accessed: 3 Oct. 2026. [Summary page read; full paper not read; peer-reviewed.]
  7. PARRIQUE, Timothée; BARTH, Jonathan; BRIENS, François; KERSCHNER, Christian; KRAUS-POLIK, Alejo; KUOKKANEN, Anna; SPANGENBERG, Joachim H. Decoupling debunked: evidence and arguments against green growth as a sole strategy for sustainability. [S. l.]: European Environmental Bureau, 2019. Available at: https://research.wu.ac.at/en/publications/decoupling-debunked-evidence-and-arguments-against-green-growth-a/. Accessed: 3 Oct. 2026. [Title and publication details only; the abstract was not on the page; the report was not read; advocacy-adjacent publisher.]

Satirical frame: SWIFT, Jonathan. A modest proposal for preventing the children of poor people in Ireland from being a burden to their parents or country, and for making them beneficial to the publick. Dublin, 1729 (public domain). Part I echoes its title, cadence and closing disclaimer of interest.

3 October 2026.