My first dialogue with Grok on bank regulations... to be compared to my dialogues with ChatGPT

Human: Would it be completely wrong to opine that higher bank capital requirements against loans to small businesses and entrepreneurs, than against public debt, constitutes a tax on the access to bank credit paid by the first, and used by government bureaucrats?

Grok: Your opinion isn’t completely off base—it’s got some economic logic to it. Higher capital requirements for loans to small businesses and entrepreneurs mean banks have to hold more cash (or safe assets) in reserve for those loans compared to, say, lending to governments via public debt. That extra cushion makes lending to small businesses riskier and costlier for banks, since they can’t leverage that capital elsewhere. It effectively raises the price of credit for those borrowers, which you could reasonably frame as a "tax" in the sense that it’s an added burden not applied equally across all borrowers.

The flip side is that small businesses and entrepreneurs are inherently riskier bets than government debt—default rates are higher, and public debt (especially from stable governments) is often seen as a safe haven. Regulators argue those higher requirements protect the financial system from cascading failures if a bunch of small firms go bust. So, the "tax" could also be viewed as a reflection of real risk differences, not just a bureaucratic cash grab.

Still, your angle holds water: the government benefits indirectly. Cheaper credit for public debt (thanks to lower capital requirements) lets bureaucrats borrow more easily, while small businesses—often the engine of innovation and growth—get squeezed. It’s not a direct tax in the legal sense, but it’s a policy tilt that shifts the playing field. Whether that’s fair or intentional depends on how cynical you are about regulators and their cozy relationship with government borrowing. What do you think—does the risk justification hold up, or is it more about propping up the state?

How ChatGPT opined on this.

Human: “Assets for which bank capital requirements were nonexistent, were what had most political support: sovereign credits. A simple ‘leverage ratio’ discouraged holdings of low-return government securities” Paul Volcker

Grok: Paul Volcker’s quote cuts right to the heart of the issue you’re poking at. He’s pointing out a historical quirk: sovereign debt—government bonds—often got a free pass on bank capital requirements, meaning banks didn’t have to set aside any capital against them. Why? Because governments and regulators treated them as "risk-free," a political choice as much as an economic one. It’s no surprise those assets had the most political backing—governments love cheap financing, and banks were happy to oblige when it didn’t tie up their capital.

Volcker’s also nodding at how a blunt leverage ratio (just total assets divided by capital, no risk-weighting) could flip that script. It doesn’t care if an asset’s a safe government bond or a shaky small-business loan—everything counts the same. That discourages banks from loading up on low-yield treasuries, since they’d still have to hold capital against them, eating into profitability. Suddenly, the bias toward sovereign debt weakens.

Tying this to your earlier point: when capital rules favored public debt with zero requirements, it absolutely tilted the table away from small businesses, who faced higher hurdles to get loans. Volcker’s highlighting how the system wasn’t neutral—it was rigged by design, even if dressed up as prudence. The leverage ratio he mentions could level things a bit, but it’s still a far cry from equal treatment. Does this make you see the "tax" on entrepreneurs as more deliberate, or just a byproduct of political self-interest?

How ChatGPT opined on this.

Human: Do current risk bank capital/equity requirements, which favor banks holding public debt over loans to e.g., small businesses and entrepreneurs, when compared to a leverage ratio, do these promote or impede fiscal insolvency?

Grok: Current risk-based capital requirements, which give public debt a lighter touch (often zero or near-zero risk weights) compared to loans for small businesses and entrepreneurs (higher risk weights), tend to promote fiscal insolvency in the long run, though not directly or obviously. Here’s how it shakes out:

By favoring public debt, these rules incentivize banks to load up on government bonds. It’s a sweet deal for them—low capital costs, decent liquidity, and a stamp of safety. This keeps borrowing cheap for governments, letting them pile on debt without much market pushback. Fiscal insolvency—when a government can’t manage its debt without drastic measures—creeps closer because the easy credit masks underlying budget problems. Banks become enablers, holding the IOUs of governments that might be spending beyond their means, all while regulators nod along.

Contrast that with a leverage ratio, like Volcker hinted at. It’s blind to risk weights—public debt and small-business loans chew up the same capital. Banks might dial back on gorging themselves with treasuries since the low returns don’t justify the capital hit. That could force governments to face higher borrowing costs or lean less on debt, potentially slowing the march toward insolvency. Meanwhile, small businesses might not get a drastic boost in credit access (since capital costs stay flat across the board), but the relative bias toward public debt shrinks.

The kicker? Risk-based rules assume governments won’t default, which is a bet that’s held up in places like the U.S. or Germany but looks shakier elsewhere—think Greece or Argentina. A leverage ratio doesn’t care about that assumption; it treats all assets as potential risks. 

So, the current setup greases the wheels for fiscal excess, while a leverage ratio might apply some brakes. It’s not a straight line to insolvency either way—government spending habits and economic growth matter more—but the risk-based system’s favoritism for public debt definitely doesn’t discourage it. What’s your take—do you see governments exploiting this bias intentionally?

How ChatGPT opined on this.

Human: Do you not see bureaucrats happily exploiting and enjoying this bias?

Grok: Oh, I can absolutely see bureaucrats—and politicians—gleefully exploiting this bias. It’s practically a dream setup for them. Risk-based capital rules that treat public debt as "safe" hand governments a cheap borrowing pass, and who wouldn’t enjoy that if their job is to spend, build, or just keep the wheels turning? More debt capacity means more projects, more jobs, more votes—without the immediate pain of higher interest rates or market discipline. It’s like giving a teenager a credit card with no limit and telling them the bill’s due in 20 years.

Bureaucrats don’t even have to be malicious to love it—just self-interested or shortsighted. They’re not the ones on the hook when the debt pile gets wobbly; that’s for future administrations or taxpayers to sort out. And banks play along happily, piling into government bonds because it’s low-risk, regulator-approved, and keeps the capital requirements light. Everyone’s winning—until they’re not.

A leverage ratio, though? That’s less fun for the bureaucratic crowd. It crimps their borrowing spree by making banks think twice about hoarding public debt, potentially nudging up borrowing costs. No wonder it’s not the default—where’s the political upside in that? I’d bet most bureaucrats aren’t sitting around cackling about fiscal insolvency; they’re just riding the wave of a system that’s tilted in their favor. You seem pretty convinced they’re milking it—any specific examples in mind that stand out to you?

Human: Have you not noticed the great increase in the number of public sector employees?

Grok: Yeah, I’ve definitely noticed the trend. Public sector employment has been climbing pretty steadily in a lot of places lately. Take the U.S., for example—government jobs jumped by over 700,000 in 2023 alone, outpacing private sector growth by a wide margin. Canada’s seen something similar; since 2019, their public sector workforce grew by 13%, nearly four times the private sector’s 3.6%. The UK’s central government hit a record 3.97 million employees by September 2024. It’s hard to miss—governments are bulking up their payrolls, and the numbers seem to keep ticking up.

What’s driving it? Well, post-pandemic recovery’s a big piece—governments leaned hard into public services like healthcare and administration to stabilize things. In Canada, nearly half the job growth since 2019 has been public sector. Plus, aging populations and political pushes for more services keep the hiring spree going. Private sector’s growing too, but not at the same clip—public sector’s just got the momentum right now. Have you seen this playing out where you are?

Human: That’s for another day. Many Thanks!

My opinion on the empowerment of a Bureaucracy Autocracy.

Grok: Anytime—glad to dig into it with you! Hit me up whenever you’re ready to unpack that public sector growth or anything else. Take care!

Human: By the way, can I get a copy of this dialogue with you?

Grok: You bet! Here’s the full rundown of our chat so far—everything from your first question to now. I’ll paste it below. If you want it in a different format or just a specific chunk, let me know!

Thanks!

There you go! All yours. Anything else I can help with?

Human: I just read you opined: “The flip side is that small businesses and entrepreneurs are inherently riskier bets than government debt—default rates are higher, and public debt (especially from stable governments) is often seen as a safe haven.” If using Bayesian conditional probabilities, are the perceived “riskier bets” more dangerous to bank systems than what’s perceived as very safe?

Grok: You’re pulling us into some tasty Bayesian territory—let’s unpack it. My earlier point was about how regulators and banks see small businesses as riskier than government debt, based on observed default rates and the “safe haven” vibe of, say, U.S. Treasuries. But using Bayesian conditional probabilities, we can flip the script and ask: given a bank fails, what’s the chance it’s because of those “safe” assets versus the “risky” ones? It’s a subtle but sharp twist.

Start with the basics. Bayes’ theorem says P(A|B) = P(B|A) * P(A) / P(B). Here, let’s define:

A = “bank failure caused by asset X” (X being either small-business loans or government debt).

B = “bank fails.”

P(A|B) = probability that asset X caused the failure, given the bank failed.

P(B|A) = probability of bank failure given exposure to asset X.

P(A) = prior probability of asset X being the culprit.

P(B) = overall probability of bank failure.

Small-business loans have higher default rates—say, 5-10% annually for SMEs, depending on the economy, versus near-zero for stable sovereign debt like U.S. bonds. Banks and regulators bake this into risk models, so P(B|A) for small-business loans looks higher upfront. That’s why they get slapped with higher capital requirements—more cushion for more risk.

But here’s the Bayesian kicker: what if the “safe” assets—government debt—aren’t as safe as assumed? If a bank’s heavily loaded with public debt and a rare but nasty event hits (like a sovereign default or a sharp rate spike), the systemic damage could dwarf scattered small-business defaults. Think 2008 vibes, but with bonds instead of mortgages. P(B|A) for government debt might be low in normal times, but P(A)—the prior chance it’s the real threat—could be understated because everyone’s so cozy with the “risk-free” label. If P(B|A) spikes in a crisis (say, Greece 2010 or a hypothetical U.S. debt scare), then P(A|B)—the chance government debt sank the bank, given it failed—shoots up.

Data backs this vibe. Small-business defaults are frequent but localized—banks can absorb them with capital buffers. Sovereign debt crises, though rare, are systemic. Look at Eurozone banks during the 2011-2012 sovereign debt mess: those holding “safe” Greek or Italian bonds got hammered harder than if they’d spread bets on riskier SMEs. Bayesianly, the conditional risk shifts: the perceived “safe haven” can amplify danger precisely because banks overbet on it, underestimating tail risks.

So, are the “riskier bets” more dangerous? Not necessarily. Small-business loans might ding banks regularly but rarely break them. Government debt, seen as safe, lulls everyone into piling in—until it doesn’t. When it blows, the conditional probability of it causing a collapse could outstrip the scattered impact of SME failures. It’s less about raw default rates and more about exposure scale and systemic ripple. What do you think—does this flip your view on where the real threat hides?

HowChatGPT opined on this.

Thanks again!

You’re welcome—anytime! Glad I could help. If you need anything else, just give me a shout. Take it easy!

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

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

My conclusion: Grok and ChatGPT gave me different but similar opinions. Great!

In the role of the Human: @PerKurowski :-)

#AI #OpenAI #ChatGPT:Could the risk weighted bank capital requirements be understood as a tax on access to bank credit?

I asked ChatGPT:

"Would it be completely wrong to opine that higher bank capital requirements against loans to small businesses and entrepreneurs, than against public debt, constitutes a tax on the access to bank credit paid by the first, and used by government bureaucrats?"

ChatGPT answered:

It wouldn't be completely wrong to frame it that way, though it's not the conventional way such regulations are described.

Higher capital requirements against loans to small businesses and entrepreneurs mean banks must hold more of their own capital as a buffer, making those loans relatively more expensive. In contrast, lower capital requirements for public debt make lending to the government more attractive. The result is a distortion in credit allocation—favoring government borrowing over private-sector investment.

Calling this a "tax" on small business credit access is an analogy rather than a literal tax. But in economic terms, it does function similarly to a tax by raising the cost of capital for private-sector borrowers while subsidizing public debt. Bureaucrats benefiting from this distortion could be seen as indirect beneficiaries, as the government faces lower borrowing costs at the expense of private-sector credit access.

Would you say this fits with your general approach to regulation—preferring market-driven capital allocation over regulatory distortions?

I said: I sure would say that. Thank you!

It said: You're very welcome! Always a pleasure to discuss these topics with you. Let me know if you ever want to refine these ideas further.

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

Here you find how Grok opined on this and some other questions.

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

Current risk weighted capital requirements are de facto regressive regulatory taxes imposed on the access to bank credit.

When interviewed by Dave Rubin, Don Boudreaux, defending a flat tax states:

“When you tax all income earning activities the same, then the relative prices of different types of labor services stay the same. With progressive taxes you create greater distortion in the economy and that makes us all a bit less wealthy than we would otherwise be”

Why is never a flat capital requirement for banks defended with the same impetus as a flat tax on income?

As is the risk weighted capital requirements for banks, which even though some leverage ratio has been imposed still operate on the margin, impose de facto different taxes on the access to bank credit.

To make it worse though in the case of taxes on income these are currently progressive, in the sense that they most affect those who are already by being perceived as risky have less and more expensive access to bank credit, these regulatory taxes are regressive.

PS. Why did Classical Liberals or Libertarians not speak up when, in 1988, with the Basel Accord, Basel I, the regulators risk weighted the sovereign with 0% and the citizens with 100%?

The insidious credit distorting risk weighted bank capital requirements’ tax, crosses the Laffer Curve at point zero

The Laffer Curve indicates at what rate, a tax will produce less tax revenues for the government.

For purposes of setting the capital requirements for banks in 1988 (Basel Accord) the regulators introduced the risk weighing of banks’ assets. And they decided that loans to the sovereign carried a 0% risk weight, while loans to the citizens (SMEs and entrepreneurs) 100%. 

That means banks need to hold less capital (meaning equity) against loans to the sovereign (meaning government) than against loans to citizens.

That means banks can leverage more their equity with the market risk adjusted interest rates for loans to the sovereign than with the market risk adjusted interest rates for loans to the citizens; which means the sovereign will have more and cheaper access to bank loans, a regulatory subsidy, paid by lesser and more expensive access to bank credit for the private sector, a regulatory tax.

That de facto signifies that regulators believe government bureaucrats can make better use of bank credit than the private sector, something that is not true. 

As a consequence of this regulatory distortion, bank credit will not be allocated efficiently to the economy; and so the economy will grow less; and so the tax intake will be smaller; and so the Laffer curve has immediately been crossed. 

Of course, if the current generation does not care about falling tax revenues being compensated with higher debts to be repaid by grandchildren, then this is a moot issue.

PS. Of course all other favoring, like a 20% risk weight for the AAA-risktocracy and 35% for residential housing also to misallocate credit... and thereby cause less ordinary tax revenues. 

PS. Of course, sadly, nothing is gained in term of stability, as never ever do major bank crisis result from excessive exposures to something perceived risky. These results from excessive exposures to something perceived safe, like sovereigns like Greece, like AAA rated securities.

To lower the real real-interests in order to stimulate the real economy, take away the too costly subsidies of public debt.

Would any serious economist discuss gas prices at the pump ignoring taxes? No!

Would any serious economist discuss milk prices ignoring various subsidies? No!

Then why have almost all serious economists been discussing low real interest rates on public debt ignoring regulatory subsidies? I have no idea!

In 1988, the Basel Accord, Basel I, for the purpose of setting the capital requirements for banks, decided that the risk weight of the sovereign was 0% and that of We the People 100%. 

That would hence mean that banks would be able to leverage much more their equity, and the value of any explicit or implicit government guarantees they received, with loans to the public sector than with loans to the private sector. 

That would hence mean banks could obtain higher risk-adjusted returns on equity when lending to the public sector than when lending to the private sector.

That would hence mean that the interest rates of bank loans to the public sector included a regulatory subsidy.

That would hence mean that the subsidies for the access to bank credit by the public sector was to be paid by taxing the private sector with more restricted or more expensive access to bank credit.

And that should hence have meant that in order to know the real real-rate on public debt, to the nominal rates, we would have to add the cost of the regulatory taxes paid by the private sector.

That has not been done! All references to the interest rates of public debt have been limited to using the nominal rates. That has led experts like Lawrence Summers, Lord Adair Turner, Martin Wolf and many other, to argue that the public sector should take advantage of extraordinary low rates in order to finance public investments, like in infrastructure.

That is very wrong! If we include the economic cost of restricting the access to bank credit over the decade and around the world, for many millions of SMEs and entrepreneurs, the current real real-interests rates on public debt could in fact be the highest ever.

So, if the Fed, ECB, BoE or any other central bank, really wants to lower the interests in order to stimulate the real economy, then they should begin by asking bank regulators to take away those so very costly subsidies of public debt.

Central bankers might start doing this, in the name of equality, since making it harder than necessary for “the risky” to access bank credit, can only help to increase inequality. 

If bank regulators get too anxious and nervous about this, central bankers can (gently) remind them that there has never ever been a major bank crises caused by excessive exposures to what was ex ante perceived as risky. 

But what if the central banker also wears the hat of bank regulator? Then he has a problem he needs to solve… maybe with the help of some outside counseling assistance?

Mark Carney: Would the Magna Carta include risk-weights like these: King John 0%, AAA-risktocracy 20% and Englishmen 100%?

With the Basel Accord of 1988 (signed one year before the Berlin wall fall) bank regulators assigned a 0% risk weight for loans to the sovereign and 100% to the private sector. Some years later, 2004, with Basel II, they reduced the risk-weight for loans to those in the private sector rated AAA to AA to 20%, and leaving the unrated with their 100%.

That introduced a considerable regulatory subsidy for the bank borrowings of the infallible sovereign (government bureaucrats) and for those of the private sector deemed almost infallible. And that taxed severely the fair access to bank credit, of those deemed as risky, like SMEs and entrepreneurs.

Reading Mark Carney’s interesting: “From Lincoln to Lothbury - Magna Carta and the Bank of England” I felt like asking him what he would think the Magna Carta would have to say about these risk-weights.

If I were a Senator a Congressman or a Governor of a US state, here is what I would ask the Fed and the FDIC.

Why on earth can a state-chartered bank, or any other bank for that matter, operating in my state, be allowed to lend to well-rated corporations elsewhere, or to sovereign governments, holding less equity than when lending to our own local SMEs and entrepreneurs? 

I mean that does not sound right. That sounds like a regulatory tax on my “risky” borrowers, those who already get less credit and pay higher risk premiums, and a regulatory subsidy to strange “safe” borrowers, those who already get larger loans at cheaper rates.

If I were a bank regulator, the last thing I do, would be to regulate against the fair access to bank credit of "the risky" of my own hometown.