← Back Econ Musings DRAFTS · UPDATED OCCASIONALLY
05 / Side Quest

Econ Musings

Lorem ipsum — a working scratchpad for half-formed ideas at the intersection of economics, AI, and how value gets produced and priced. Most of these are toy models, sketches, or stray paragraphs that aren't quite ready for a proper blog post. Sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.

01

Cobb-Douglas AI

Sketch · In Progress

The instinct, when adding a new input, is to bolt a third exponent onto Cobb-Douglas: Y = A · Lα · Kβ · Tγ where T is tokens, or tasks, or "intelligence-on-demand". It's tempting and it's almost certainly wrong in a way worth being precise about. The interesting question isn't whether T belongs in the function — it's how, and that depends on the regime of AI capability you're modelling. I see two avenues, and most of the current debate confuses them.

⚠ Pushback · The functional form is doing unearned work

Cobb-Douglas assumes a constant elasticity of substitution between every pair of factors, and that elasticity equals 1. That assumption breaks immediately for AI. AI is a near-perfect substitute for labour in some tasks (ticket triage, first-draft copy, boilerplate code) and a near-zero substitute in others (judgment under genuine ambiguity, trust-laden sales conversations, escalations). Compressing that range into a single γ is dishonest.

Two more honest forms: (a) nested CES with task-level elasticities — σ < 1 between AI and human labour within a task, σ > 1 across the bundle; or (b) a task-based production function in the Acemoglu-Restrepo tradition, built up from tasks rather than imposed from above. I'll keep using "Cobb-Douglas" as shorthand because the α, β, γ labels are useful, but the functional form is a fiction. What matters is the story about the exponents.

A · Labour Productivity Lens · Pre / Limited-Agentic

In this regime AI isn't a factor at all. It's a technology shifter — either inside the Hicks-neutral A term, or, more usefully, a multiplier φ on the effective units of labour:

Y = A · (φ · L)α · Kβ,   φ ≥ 1

Tokens get bought; the buying just lets each worker get more done per hour. Model and inference flow are a complement to labour, not a substitute. This is where most of the current empirical evidence actually sits.

  • Brynjolfsson, Li & Raymond (2023) — generative AI in customer support raised resolutions per hour by ~14%. Crucially, the largest gains went to novice workers. The model didn't replace them; it compressed the experience curve. NBER WP 31161.
  • Noy & Zhang (2023, Science) — writing tasks completed ~40% faster with ChatGPT, output quality steady-to-improved. Same shape: augmentation, not substitution.
  • Peng et al. (2023) — GitHub Copilot, ~55% speedup on a controlled coding task. Developer remains in the loop throughout.
  • Anthropic Economic Index (Feb 2025) — task-level decomposition showing the augmentation share materially exceeds the automation share across most occupations sampled in the first release.

The policy-relevant question inside Avenue A is the distribution of φ. Compression (novices catch up) or bifurcation (experts pull further away)? The evidence leans compression — which is itself a real result, and an uncomfortable one for the "AI will hollow out the middle" narrative that dominated 2023-2024.

⚠ Pushback · Treating φ as free

"AI as A-shifter" elides cost. Tokens have real marginal cost — paid per task, in opex, often a non-trivial fraction of gross margin. Modelling φ as pure TFP misrepresents the income statement. The cleaner formulation has each unit of task-labour as a bundle of "human-time + AI-tokens" jointly, with the bundle entering as effective L. That matters because it changes what the firm is optimising: not "how productive is my worker?" but "what's the cheapest mix to complete this task?"

B · Factor-of-Production Lens · Agentic Regime

When the agent runs autonomously — closes the ticket, writes the function, drafts the contract, joins the meeting — the production function genuinely changes shape and AI earns its own exponent. But "AI as capital" needs care, because three very different things get conflated under that label:

Model Weights
Durable produced asset, depreciating as better models ship. → Capital.
Inference Flow
Tokens consumed per task, variable cost, no stock to hold. → Intermediate input (closer to electricity than to capital).
Agent Service
Autonomous completion of a unit of work — the thing the firm is actually buying. → Labour, priced as a service.

A Series B SaaS firm spending $400K/month on Anthropic and OpenAI isn't holding capital in the sense a factory holds a CNC machine. It's buying a service flow whose underlying asset sits on someone else's balance sheet. So the sharper Avenue B question isn't "is AI capital?" — it's "for whom is AI capital?" Foundation labs hold the capital. Customer firms buy a service that competes with labour at the task level. The aggregate story has to be assembled from both views, and a one-sector model that just swaps "AI" in for K will get the long-run incidence wrong.

⚠ Pushback · The "agents = capital" leap

The popular framing — agents are workers, agents are owned, therefore agents are capital — skips a step. What the firm owns is a contractual right to inference, not a stock of agents. Reclassifying that as capital risks importing capital-deepening dynamics (declining marginal product, accumulation paths, depreciation schedules) that don't actually apply to a service subscription. The economics of the lab and the economics of the customer firm need to be modelled separately before they're aggregated.

Arc · The Token Budget and the Hiring Tell

Here's where the question gets sharp. Picture a firm with a monthly token budget B and a headcount L. B triples over 18 months. What else changes is the tell:

Scenario A · Augmentation
B↑, L steady-or-up, output ↑ roughly proportionally. φ rose. AI is multiplying labour.
Scenario B · Substitution
B↑, L↓, output steady or modestly up. Factor substitution at the margin.
Scenario C · Mixed
B↑, L mixed by role, output ↑. The empirically common case in 2025-2026.

Now the tension the user is pointing at. The accounting tells one story; the hiring decisions tell another. Token spend is booked as opex — cost of revenue, often — which behaves like a variable input. Labour-adjacent. But the headcount decisions — slowed hiring in customer support, junior copywriting, paralegal review, BPO-style code maintenance — behave like capital-labour substitution at a long-run margin.

Firms are treating AI as labour on the income statement and as capital in the org chart.

That gap is the research question. The accounting choice isn't a typo — opex is genuinely the right book entry for an inference subscription. But the decision it enables looks much more like the kind of multi-year factor swap that capital deepening describes. The two views can both be right at the same time, and that's what makes the dynamics worth modelling separately from either pure Avenue A or pure Avenue B.

My current hypothesis: firms haven't decided which lens they're operating in, and the inconsistency between the two is doing real damage to workforce planning. Token spend grows because each engineer "needs" it (Avenue A logic — productivity tool), and headcount freezes happen in parallel (Avenue B logic — substitute available). Both decisions feel locally rational. The aggregate is incoherent.

? Open Questions · What Would Discriminate

Empirical work I'd want to see, roughly in order of tractability:

  • Token spend per FTE across cohorts of comparable firms, correlated with role-level hiring slowdowns. Cross-firm panel, plausibly assemblable from public filings + LinkedIn data + provider disclosure.
  • Within-firm task allocation — are tokens deployed in support of existing workers (A) or to retire tasks entirely (B)? The Anthropic Economic Index gets at this at the task level; the firm-level cut is harder and more revealing.
  • Wage dispersion within roles. Avenue A predicts compression (novices catch up). Avenue B predicts bifurcation (the workers who remain are the ones whose tasks resist agent substitution).
  • The γ-vs-α-fall test. Functionally similar, theoretically distinct: γ rising says AI is a new input; α falling says human labour just got better. The cleanest discriminator is whether AI spend co-moves with output (γ) or substitutes for L holding output constant (α↓).
⚠ Pushback · Macro modest, micro brutal

Acemoglu (2024, "The Simple Macroeconomics of AI") argues aggregate impact is modest — well under 1% TFP per decade under reasonable parameter choices. That may be right at the macro level while role-level substitution is severe. "Macro modest, micro brutal" is a real possibility, and aggregate TFP measures will badly hide it. Anyone using top-line productivity numbers to dismiss labour-market concern is reading the wrong gauge.

Production Theory CES & Task-Based Labour Augmentation Factor Substitution Empirical
02

Agentic WTP

Sketch · In Progress

Lorem ipsum dolor sit amet — willingness-to-pay (WTP) is, in most pricing literature, modelled as a property of a human: a distribution over reservation prices, shaped by income, preferences, framing, and the moment of decision. Consectetur adipiscing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua.

What changes when the buyer is an agent acting on behalf of a human? Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. The agent doesn't experience the loss-aversion sting of paying. It doesn't get embarrassed by a price that "feels too high". It doesn't anchor on round numbers. But it does have a budget, a utility function, and a posterior over whether this purchase will be approved post-hoc.

Three asymmetries I keep coming back to: (a) the agent's price elasticity should be much sharper around the human-defined budget cap and almost flat elsewhere; (b) the agent will comparison-shop with far more breadth than a human ever would, collapsing brand premia; (c) the agent's reservation price reflects expected human approval, not expected human utility — and those can diverge quite a lot.

Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. The visualisation I want to build: a side-by-side simulator of human WTP vs agentic WTP under varying levels of agent autonomy, with the equilibrium price drift shown over a few hundred transactions. Excepteur sint occaecat cupidatat non proident.

Pricing Behavioural Econ Agents Market Design