Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Do central banks’ mounting losses actually matter?

ft.com

Post reply on HN
Print view View original

171–180 of 241 posts

Re: Do central banks’ mounting losses actually matter?

#171

Post by simoncion »

pjc50 wrote: post #138
roenxi wrote: post #126

Earlier quoted context omitted.

"lay people": It looks a lot like you're printing money and handing it out to your wealthy buddies. "experts": This matter is too complex and jargon-filled for you to take part in the debate. "lay people": Explain again why we have an institution that alternates between handouts to asset owners and crashing the economy? I think prices going up exponentially with time might be bad for my welfare! "experts": Those craz…

> printing money and handing it out to your wealthy buddies. So much confusion is caused by the inability to distinguish between "gift" and "loan". The problem with allowing banks to hard-fail is that imposes real, huge costs on customers even if the deposits are 100% covered, because the failover is not instant and you end up with money locked up for a period and are unable to make your own payments. > prices going…

> [When a bank fails, depositors] end up with money locked up for a period and are unable to make your own payments.

From the FDIC's Deposit Insurance FAQ (https://www.fdic.gov/resources/deposit-insurance/faq/index.h...>):

> Q: What happens when a bank fails?

> ... First, as the insurer of the bank's deposits, the FDIC pays insurance to depositors up to the insurance limit. Historically, the FDIC pays insurance within a few days after a bank closing, usually the next business day, by either 1) providing each depositor with a new account at another insured bank in an amount equal to the insured balance of their account at the failed bank, or 2) issuing a check to each depositor for the insured balance of their account at the failed bank.

Honestly, I'd expect just about any judge in the country to be _very_, _VERY_ angry at a creditor that was unwilling to -free of charge- delay receipt of money owed by (typically) one or (sometimes) a few business days because the Federal Government needed that time to take over a failed bank's obligations.

This reason you've provided is just plain bogus.

Re: Do central banks’ mounting losses actually matter?

#172

Post by doubleunplussed »

concinds wrote: post #152
WastingMyTime89 wrote: post #136

Earlier quoted context omitted.

There is not that much jargon. It’s just an advanced topic so people use economic terms they expect people to know. Grab an introduction on macroeconomics for exemple Mankiw then any book on modern monetary theory and everything should be quite clear.

Strange to recommend Mankiw, and Modern Monetary Theory (a discredited, non-mainstream economic concept that Mankiw wrote against) in the same sentence. MMT is practically pseudoscience, makes unclear claims, claims a foundation in mainstream economics but makes completely wild logical leaps when it comes to policy recommendations, and it's promoted by politicians more than real economists.

Well, it seems half the people criticising MMT say it's a vacuous re-statement of how things already work, and the other half say it's wrong and discredited :p.

What does it actually advocate that has been discredited? I get the impression MMTers want to erase the distinction between central banks and the government, and basically run everything with fiscal policy, letting the government decide how big its deficits are instead of effectively the fed deciding. Is that right?

I mean, that would work in principle, right? It would just be a bad idea because the decision is best made by an independent entity rather than the government, which has short-term incentives that may lead it to make bad decisions about spending (usually: too much).

And maybe it would be a bad idea because interest rates are a more reliable lever for controlling the money supply than tax and expenditure rates? I mean, if that's the case, it's not super obvious.

And since these things haven't been tried, I don't suppose it's those things that have been discredited.

From what I see, people are saying MMT has been discredited because large government deficits have led to high inflation recently. But this seems off to me - MMT doesn't say infinite deficits are possible, they're saying the deficits are limited by inflation. Which is true and we're now seeing the consequences of exceeding that limit.

Re: Do central banks’ mounting losses actually matter?

#173

Post by cm2187 »

satellite2 wrote: post #159
cm2187 wrote: post #103

Earlier quoted context omitted.

They can absolutely make credit conditions worse, through capital requirement and liquidity requirement. Also if you withdraw QE, it drains banks deposits and their ability to fund new loans.

Capital and liquidity requirements are not decided by central banks generally. In most democracies it's a legislative branch responsibility. They cannot "withdraw" QE. What they have been doing is buying bonds at a higher price than the market thus lowering the interest rates they simple buy them a bit cheaper now.

Capital and liquidity requirements are negotiated by central banks in the basel committee. Then the central bank will add its own home made requirements, and that’s the legislation that gets proposed to parliament. The legislation includes all sort of discretionary buffers that the central bank can freely impose. Central banks are front and centre in banks regulatory requirements.

Withdraw QE = either sell bonds or let them mature, reducing the size of the fed balance sheet. Which drains liquidity from the market (new treasury issuances have to be absorbed by the market).

Re: Do central banks’ mounting losses actually matter?

#174

Post by kypro »

sedeki wrote: post #143
doubleunplussed wrote: post #110

This thread shows a phenomenon that I've noticed a lot of lately: that many smart people turn absolutely loopy when it comes to the topic of central banking. There's nothing quite like it, and I'm not sure how to explain it. The topic seems to make conspiracy theorists out of otherwise very reasonable people. Good to see the highest voted comments are sane, but the sanity ratio is pretty low compared to other topics.

So where can I read up on how modern central banking actually works? Like, what book(s) or curriculum do I need to study? Genuine question.

Central Banking 101 by Joseph J Wang is decent.

Re: Do central banks’ mounting losses actually matter?

#175

Post by doubleunplussed »

hydroreadsstuff wrote: post #122
doubleunplussed wrote: post #113

Earlier quoted context omitted.

I would hope they set that rate to what they think is optimal for the economy, regardless of their solvency. I only skimmed the article, but I don't see where it gives reasons why solvency matters, although it hints it might. I was under the impression it doesn't really matter if the fed is solvent, since I can't think of any concrete consequences.

Afaik the needed difference in interest payment from the fed comes from the treasury. So it’s indirectly connected to the budget. In theory though you can still raise more debt to pay the interest. But I’m not sure about the longterm consequences of this. Having inflation above the interest rate helps decreasing the debt/gdp ratio.

Why can't they just pay the interest with newly-created money?

I mean, that might not be how things are set up right now, and it would make their position even more negative since newly-created cash is a liability on their balance sheet and they didn't get an asset for it. So maybe at the moment it's on the treasury to make up the difference. But I don't see a fundamental reason why that needs to be the case, it seems incidental to the current setup.

And yeah, even under the current system the treasury could sell bonds to fund the shortfall, and then the Fed could just buy those bonds (via the market) and keep them on their balance sheet, rolling over forever and so effectively loaning the money to the treasury at zero interest forever. So that's just the same thing but with extra steps.

Re: Do central banks’ mounting losses actually matter?

#176

Post by satellite2 »

sedeki wrote: post #143
doubleunplussed wrote: post #110

This thread shows a phenomenon that I've noticed a lot of lately: that many smart people turn absolutely loopy when it comes to the topic of central banking. There's nothing quite like it, and I'm not sure how to explain it. The topic seems to make conspiracy theorists out of otherwise very reasonable people. Good to see the highest voted comments are sane, but the sanity ratio is pretty low compared to other topics.

So where can I read up on how modern central banking actually works? Like, what book(s) or curriculum do I need to study? Genuine question.

https://www.coursera.org/learn/money-banking is an extroardinary introduction on what is actually money and how banking works (including the role of central banks)

Re: Do central banks’ mounting losses actually matter?

#177

Post by mattbrewsbytes »

doubleunplussed wrote: post #110

This thread shows a phenomenon that I've noticed a lot of lately: that many smart people turn absolutely loopy when it comes to the topic of central banking. There's nothing quite like it, and I'm not sure how to explain it. The topic seems to make conspiracy theorists out of otherwise very reasonable people. Good to see the highest voted comments are sane, but the sanity ratio is pretty low compared to other topics.

As are a lot of discussions here on HN with topics people aren't particularly knowledgeable about. You have to be decently smart to program computers to do things and that doesn't necessarily translate to other fields but because people feel so smart in one area they feel like they can problem solve other areas. I think dunning-kruger effect is the term but I'm not a psychologist so that could be wrong (ha!).

Re: Do central banks’ mounting losses actually matter?

#178

Post by mjburgess »

nwienert wrote: post #165
mjburgess wrote: post #151

Earlier quoted context omitted.

It's not a dilution of value when the risk is deflation; and at zero-inflation, it's not a tax because it encourages growth by investment. So it's only a "tax" if it (1) causes inflation to rise; and (2) rise above the level which promotes growth. Neither of those conditions obviously obtains in the era of QE. Indeed, (1) never did.

You’ve made up three things, each without convincing rhetoric let alone evidence, and then put QED at the end.

I presume I'm speaking to a person who understands the evidence for each, and that the argument is about how one frames QE.

There are videos, no doubt, if you want to understand why a positive rate of inflation promotes growth, and hence why central banks are legally obligated to "print money" to ensure a 2% target.

The relevant disagreement here, I think, isn't the basic economics. Its under what conditions the "'deliberate' inflation = tax" view is correct. Since it is correct under some conditions.

I'd dispute that it accurately describes the last decade of central bank policy, which on the whole, seems fairly successful at avoiding a post-crash depresson, and a post-pand. depression.... Both seem highly implausible feats done via taxation.

Re: Do central banks’ mounting losses actually matter?

#179

Post by satellite2 »

cm2187 wrote: post #173
satellite2 wrote: post #159

Earlier quoted context omitted.

Capital and liquidity requirements are not decided by central banks generally. In most democracies it's a legislative branch responsibility. They cannot "withdraw" QE. What they have been doing is buying bonds at a higher price than the market thus lowering the interest rates they simple buy them a bit cheaper now.

Capital and liquidity requirements are negotiated by central banks in the basel committee. Then the central bank will add its own home made requirements, and that’s the legislation that gets proposed to parliament. The legislation includes all sort of discretionary buffers that the central bank can freely impose. Central banks are front and centre in banks regulatory requirements. Withdraw QE = either sell bonds or l…

Minimum capital requirements are laws. They are defined in the Dodd-Frank Act in the US for instance.

Re: Do central banks’ mounting losses actually matter?

#180

Post by LatteLazy »

So as far as I can tell there are 2 types of losses: losses on paper (the market price of the bond I own fell but I have not sold yet) and cash losses (the government defaulted and didn't pay).

Losses on paper are irrelevant because they don't effect the position of the bank or the amount of cash in circulation. The same applies for gains on paper. The bank just holds the bond until it expires at parity.

Defaults or selling the bond for less than the real value is much more serious. This is because it leaves money in circulation.

So if the central bank prints 100USD/Euro etc, holds the bond and then get's it's 100 back and burns it, then no currency has been created over all (just for a while between printing and burning). So there is no inflationary effect etc. If the price of the bond goes up or down, that's irrelevant as long as the bank never sells and just collects the 100 back.

But if the bank prints 100 and buys a bond. Then sells it for 70. Then it can only burn the 70 it has, and then other 30 remains in circulation and there is a lasting increase is the money supply. Similarly if they sold the bond for 130 and burnt that, they would be actively reducing the supply to less than it was before (deflation)

I don't think any central bank is SELLING bonds (other than at face value) right?

So there is no issue here.

Also, as the bonds come closer to their maturity date, their price will tend towards the face value. So this "issue" will disappear on paper too...

Post reply on HN