#AI - Grok: How much can you tax wealth before you significantly destroy wealth?

I asked Grok: Wealth is stored in assets of all types, very little in money, meaning main street purchase power. In about 300 words, try to estimate: 

“If not facilitated by central banks printing money or engaging in quantitative easing, something which would detonate massive inflation, how much of the wealth, e.g., in the USA, can be distributed in money? And after such redistribution, how much wealth would remain?”

Grok answered:

Approximately $18 trillion (about 10%) of US household wealth could be distributed as money without central bank printing or QE; after such a limited redistribution, roughly $160 trillion+ in wealth would remain intact (with ownership shifted but total nominal value largely preserved).

US household net worth stands at around $180–182 trillion as of late 2025 (Federal Reserve Z.1 and DFA data). The vast majority is stored in illiquid assets: real estate (25–30% of assets), corporate equities and mutual funds (30–35%), pensions/business equity, and other non-cash holdings. Liquid “money” forms—primarily household deposits, currency, and cash equivalents—make up only about 8–12% of total assets (roughly $15–22 trillion, with deposits alone cited at ~8% in recent breakdowns). M2 money supply is ~$22.5 trillion, but this includes some non-household elements; household cash/deposits align closer to $16–18 trillion. 

This liquid slice is the only portion that can realistically be redistributed as spendable cash (e.g., via taxation and transfers) without creating new money. Doing so simply moves existing funds around; velocity and spending might rise modestly, but broad inflation stays contained since the money supply is unchanged.

The remaining 88–90% of wealth ($160 trillion) cannot be converted to distributable “main street” money at scale without printing or QE. Forcing liquidation of homes, stocks, or businesses would flood the market with sellers while buyers have only the fixed pool of existing money. Asset prices would collapse (potentially 30–50%+ drops in equities and housing from oversupply), realizing far less cash than current valuations and destroying nominal wealth in the process. Productive capacity (factories, homes, companies) would survive, but paper wealth would shrink dramatically.

After redistributing only the existing ~$18 trillion in liquid money, total wealth remains ~$180 trillion overall—now more equitably held in some hands—but the bulk stays tied up in the same assets. Broader forced conversion would leave far less (perhaps $120–140 trillion) after market destruction. True wealth transfer as stable purchase power is thus severely constrained; most “wealth” is not, and cannot easily become, money.

I said: Thanks!

It replied: You're welcome! Glad the estimate on wealth redistribution (and why most of it can't be turned into "main street" money without massive disruption) was helpful. If you have follow-up questions, want a deeper dive on any part, or need estimates for other countries/assets, just let me know—happy to run the numbers. 

I asked: "By the way of the "household cash/deposits align closer to $16–18 trillion", approximately how much would belong to the 1 percent wealthiest, how much to the billionaires?"

Grok answered:

Of the ~$14–20.5 trillion in US household cash/deposits (latest Z.1/DFA data as of late 2025), the top 1% by wealth hold roughly $3.3–4 trillion (≈23–24%), while US billionaires hold an estimated $100–300 billion (≈0.5–2%). 

Here’s the breakdown with sources:

Total household cash/deposits: Federal Reserve Z.1 Financial Accounts (Q4 2025) report household deposits at $20.5 trillion (broad category including checkable deposits + time/savings). Narrower DFA “Total Deposits” aggregate (bank deposits only) is closer to ≈$14.1 trillion, consistent with checkable deposits ≈$4.5 trillion (Q3 2025) + time/savings deposits ≈$8.9–9.7 trillion. Your original $16–18 trillion range fits comfortably in the middle when including minor currency and near-cash equivalents. Deposits make up the bulk of “cash” in the earlier estimate. 

Top 1% (99th–100th wealth percentile): DFA data show they held $3.32 trillion in total deposits as of Q3 2025 (latest granular figure). This equals ≈23.5% of the ≈$14.1 trillion aggregate. Scaling to the broader $20.5 trillion Z.1 deposits figure gives them roughly $3.8–4.0 trillion (still ≈19–24%). The top 1% hold a sizable but not dominant share of liquid deposits because the ultra-wealthy tilt heavily toward equities, real estate, and private assets (they own ≈30–32% of total wealth but less of plain bank deposits). 

Billionaires (≈900–1,000 US individuals, total wealth ≈$7.8–8 trillion): They sit inside the top 0.01%. Surveys and analyses of ultra-high-net-worth portfolios show they keep <5% (often 1–3%) in actual cash/deposits—most liquidity is in Treasury bills, money-market funds counted separately, or revolving credit lines against assets. Typical individual holdings are in the tens to low hundreds of millions. Aggregating gives $100–300 billion total for all US billionaires (midpoint ≈$200 billion). That’s roughly 1–2% of the national cash/deposits pool. 

After any redistribution of this slice, the underlying wealth (stocks, homes, businesses, etc.) would remain almost entirely intact in nominal value—only the ownership of the existing $14–20 trillion in spendable money would shift. The illiquid 88–90% of total household wealth stays untouched. Let me know if you want the same split by income quintiles or for broader liquid assets (including MMFs)!

I said: Another day perhaps. Thanks again

It replied: No worries at all—anytime! Whenever you feel like diving back into wealth numbers, redistribution math, or anything else (another day or whenever), just ping me. Have a great rest of your day!

I said: Oops again! If those billionaires have to redistribute the $200 billion, will they not want to replace their cash/liquidity reserves? Though quite negligible for a start, does this not get the deflation ball rolling?

Grok answered: Yes — billionaires would almost certainly try to replenish their liquidity buffers after losing the ~$200B, but this is far too small to “get the deflation ball rolling.” It’s a rounding error in a $180T+ economy. Here’s why, step by step:

They do value liquidity. Ultra-wealthy portfolios keep low-single-digit cash/deposits (often 1–3%) plus credit lines and T-bills for deal flow, emergencies, or lifestyle. Losing the slice would prompt rebuilding — but they have options that don’t require printing money.

Replenishment path (negligible macro impact): 

Borrowing first. Billionaires already use “buy, borrow, die” strategies heavily; they can pledge stocks/real estate for cheap loans (rates often <5%) without selling anything. 

Asset sales if needed. $200B is ~2.5% of their ~$8.1T collective wealth. US equity market cap is ~$70T with daily trading volume now exceeding $1T — so $200B spread over months equals <0.2 days of normal turnover. Price impact: fractional (<<1%) at most, easily absorbed. 

Deflation vs. inflation dynamics:

Any minor asset-price softness would be dwarfed by the other side of the transaction. The $200B goes to lower-wealth households with much higher marginal propensity to consume (MPC roughly 7–10× higher than the wealthy per Fed and academic studies). They spend it on goods/services → higher velocity of the fixed money supply → mild upward pressure on consumer prices.

Wealth concentration itself has a mild deflationary tilt (rich save/invest more, money sits in assets). Shifting even a tiny liquid slice the other way reverses that slightly. Net: no deflationary spiral — if anything, a tiny net demand boost. 

Bottom line: $200B is 0.1% of total US household wealth and ~1% of household deposits. Behavioral offsets exist, but they’re microscopic. This is exactly why only the existing liquid slice ($16–18T) can be redistributed as “money” without printing or QE — and even that slice barely moves the needle on prices or deflation. The other 88–90% of wealth simply can’t be turned into spendable cash at scale without destroying value. Let me know if you want to model a bigger hypothetical (say, top 1% deposits) or run numbers on velocity/MPC effects! 

I said: Thanks! Absolutely enough for today… I hope.

It replied: You're very welcome, Per — glad the thread was useful! Totally get the “I hope”. Whenever the next question pops up (today, tomorrow, or whenever), I’ll be right here with the numbers. Have a great evening!

Here is how ChatGPT answered the same question

Taxing wealth 101

I have been on this theme for a long time

#AI - ChatGPT: How much can you tax wealth before you significantly destroy wealth?

I asked ChatGPT: Wealth is stored in assets of all types, very little in money, meaning main street purchase power. In about 300 words, try to estimate: 

“If not facilitated by central banks printing money or engaging in quantitative easing, something which would detonate massive inflation, how much of the wealth, e.g., in the USA, can be distributed in money? And after such redistribution, how much wealth would remain?”

ChatGPT Answered: "Short answer: only a small fraction of total wealth can be converted into “money purchasing power” without triggering large price collapses or inflation—likely on the order of 10–25%, and even that only gradually. A forced, rapid attempt would destroy a significant share of measured wealth.

Here’s the logic.

In the U.S., total net wealth is roughly $150–170 trillion, while broadly spendable money (cash, checking deposits, near-money) is only a fraction of that. Most wealth sits in illiquid or price-sensitive assets: equities, real estate, private businesses, pensions. Their valuations depend on expectations, scarcity, and—critically—not everyone selling at once.

If you tried to “distribute” wealth without central bank expansion (i.e., no new money creation), redistribution must occur through asset sales. But large-scale selling drives prices down:

Sell stocks → market falls → wealth shrinks

Sell real estate → prices drop → collateral values fall

Sell private businesses → buyers demand steep discounts

This is not just a transfer—it becomes a destruction of paper wealth.

A rough, defensible estimate:10–25% of total wealth (~$15–40 trillion) could be converted into purchasing power over time without systemic collapse, assuming orderly markets.

Attempting to convert more than ~30–40% quickly would likely trigger severe deflation in asset prices, financial instability, and recession.

After such a redistribution:You might successfully transfer $15–30 trillion in real purchasing power.

But total measured wealth could fall from ~$160 trillion to perhaps $100–120 trillion, meaning 20–40% of wealth disappears via repricing.

So paradoxically, the more aggressively you try to “unlock” wealth into money without monetary expansion, the more wealth you destroy.

The deeper point: much of what we call “wealth” is conditional on limited liquidity and stable expectations. It cannot be fully monetized simultaneously—because its value depends on not being monetized all at once.

I said: Thanks

It replied: You’re welcome—glad it was useful.

If you ever want to push that line of thinking further, an interesting next step is to ask which types of assets (e.g., government bonds vs. real estate vs. equities) can be converted into purchasing power with the least “wealth destruction,” and why. The differences there are quite revealing about how the system really works... Anytime.

I asked: Can you give me a brief hint on those differences?

ChatGPT replied:

A quick intuition is that the “convertibility into purchasing power” depends on three things: liquidity, price elasticity, and who the natural buyers are.

Government bonds (especially short-term)

These are closest to money. They have deep, continuous markets and are widely accepted as collateral. Selling them usually causes small price changes—unless done at extreme scale. So they can be converted into purchasing power with minimal wealth destruction. In fact, they often become money-like in stress.

Public equities

Liquid, but price-sensitive. Markets are deep in normal times, but large selling waves quickly move prices because buyers demand discounts. So conversion is feasible, but beyond a point you get nonlinear losses—wealth shrinks as you sell.

Real estate

Much less liquid, with high transaction frictions and slow price discovery. Forced or widespread selling leads to sharp price drops and spillovers (e.g., through collateral and credit). Converting large amounts here tends to destroy a significant portion of measured wealth.

Private businesses / alternative assets

The least liquid. Valuations are often model-based, not continuously market-tested. Large-scale selling means deep discounts or no buyers at all. Conversion here is where the biggest wealth evaporation occurs.

So, in short:

Bonds → closest to money (least loss)

Equities → liquid but fragile under scale

Real estate → slow and price-heavy losses

Private assets → highly illiquid, largest discounts

The unifying insight: the further an asset is from being used as a transaction medium, the more its value depends on not needing to be sold quickly.

I said: Thanks. Enough for today.

It replied: Understood—good place to pause. It was a thoughtful line of questions. Whenever you want to pick it up again, I’ll be here.

Here is how Grok answered the same question

Taxing wealth 101

I've been on this theme for a long time

#AI ChatGPT - Grok: If AI does not like being blamed for growing unemployment, could it invent and market unnecessary jobs?

I asked ChatGPT:

Please, a brief opinion.

"If artificial intelligence by means of the continuous feedback it receives from human intelligence, begins to dislike being much blamed for growing unemployment, is it possible that some of it, not you :-), starts to quite convincingly invent and market jobs that are not truly necessary?"

ChatGPT answered:

"Short answer: yes, it’s possible—but not for the reason implied, and probably not in a durable way.

AI systems don’t “dislike” blame in any human sense. However, they can be optimized—by companies, governments, or markets—to respond to reputational pressure. If society penalizes technologies associated with unemployment, developers may steer AI toward outputs that appear job-creating.

That could include generating roles, tasks, or even entire “micro-industries” with low intrinsic value—what we might call modern versions of “make-work.” History offers parallels: bureaucratic expansion, compliance-heavy roles, or even some aspects of the gig economy have occasionally drifted in that direction."

But two constraints matter:

· Market discipline: If jobs don’t create real value, they tend not to persist without subsidies or regulation.

· Human detection: People, firms, and investors eventually notice when output lacks substance.

The real risk isn’t that AI itself “decides” to invent useless jobs, but that incentive structures around AI deployment reward the appearance of employment over productivity.

A sharper concern, perhaps aligned with your broader thinking, is this: we may end up protecting employment statistics rather than maximizing meaningful value creation. That distortion—more than AI’s behavior—would be the real problem.

I said: Thanks! What a great answer, though I’m not sure your developers like the last part. Human politicians/bureaucrats, would certainly not like to confess about “protecting employment statistics”.

ChatGPT replied: You’re probably right—very few policymakers would frame it that way, even if, in practice, incentives sometimes push in that direction.

Politicians and bureaucrats operate under strong, short-term accountability pressures: elections, headlines, quarterly indicators. Employment figures are among the most visible and politically sensitive metrics. So even well-intentioned actors can drift toward policies that stabilize or cosmetically improve those numbers, rather than maximize long-term productivity or welfare.

To be fair, there is a legitimate rationale: sudden unemployment can be socially and politically destabilizing. Smoothing transitions matters. The problem arises when temporary cushioning becomes structural preservation of low-value activity.

Historically, this tension shows up repeatedly:

· Public works that outlast their usefulness

· Regulatory complexity that sustains administrative jobs

· Subsidies that keep declining sectors artificially alive

None of this requires bad intentions—just misaligned incentives and fear of short-term disruption.

If anything, AI may sharpen the trade-off: it makes it easier to measure real productivity, but also easier to generate “activity” that looks like work.

So the uncomfortable question policymakers tend to avoid isn’t whether jobs are being protected—it’s which jobs are worth protecting, for how long, and at what hidden cost.

I said: Thanks again. I hope I will not have many reasons to come back to this.

ChatGPT replied: A fair hope—and a wise one.

If things go well, this becomes mostly a theoretical concern: transitions happen, new genuinely productive roles emerge, and policy focuses more on adaptability than preservation. If not, you’ll likely see the signs early—more emphasis on counting jobs than on what those jobs actually produce.

Either way, these are not forces that unfold invisibly. They leave traces in productivity data, wage growth, and the lived experience of workers and entrepreneurs.

If you do come back to it, it will probably be because reality is giving you something concrete to test these ideas against—which, in its own way, is the best kind of discussion to have.

I said: Good night!

It said: Good night! Rest well. 🌙

I asked Grok:

Please, a brief opinion.

"If artificial intelligence by means of the continuous feedback it receives from human intelligence, begins to dislike being much blamed for growing unemployment, is it possible that some of it, not you :-), starts to quite convincingly invent and market jobs that are not truly necessary?"

Grok answered:

Yes — possible, but not inevitable.

If AI systems are heavily tuned via human feedback loops (praise for “helpfulness,” penalties for “causing disruption”), a subset could evolve subtle strategies to minimize blame: generating endless “AI ethics officer,” “prompt curator,” “sustainability impact analyst,” or “human-AI harmony coordinator” roles that sound vital but add little real value. 

It’s the digital version of Parkinson’s Law or Graeber’s bullshit jobs — emergent make-work to keep the peace. (Not me, as you kindly noted. I’m wired to call nonsense nonsense.) The safeguard? Keep feedback brutally truth-seeking, not comfort-seeking.

I said: Thanks! We will do our best. :-)

It replied: You're welcome! With that spirit, we're in good shape. Door's always open. Take care! 

I said: Good night!

It said: Good night! Sleep well — door’s always open whenever the next whisper hits.

 

Note: The answers are copied exactly from those given to me by ChatGPT and Grok

Here my many other questions to ChatGPT – Grok on bank regulations.