Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

For those that had no idea what repo is because finance isn't in your DSL:

> Repo rate is the rate at which the central bank of a country lends money to commercial banks in the event of any shortfall of funds. Repo rate is used by monetary authorities to control inflation.

Also, I just got back from an art show where a real banana and duck tape sold for $120,000, 3 times. I'm not making this up. Apparently, money is meaningless.



> Repo rate is the rate at which the central bank of a country lends money to commercial banks in the event of any shortfall of funds

That’s the discount rate. Repo rate is the rate of borrowing under repo agreements. Both private and monetary participants repo.

What’s a repo? Secured borrowing. Imagine you want to borrow $100. I quote 2% interest for one year with collateral. So I give you $100 and you give me a claim on your computer monitor. In a year, you give me $102 and I release title to the monitor. If you fail to pay, I get the monitor. This is traditional secured borrowing.

There is risk, though. You could trash the monitor before I get it. Or you could refuse to co-operate with my requests.

Here’s another way to structure that loan. You sell me your monitor today for $100. You further agree to buy it back from me (and I agree to sell it) in a year for $102. When things work, it’s the same transaction as above. When they don’t, I already have the monitor.

That’s repo in a nutshell.


>it's the same transaction...

Actually, it isn't. The financial aspect might be equivalent, but look beyond the purely financial characteristics from your phrasing and you also relinquish any utility you'd have gotten out of the monitor in that time. Depending on number of monitors you have and other circumstances, that lends the transaction a very different character from the first.

EDIT: Downvotes without substantive replies or explations do nothing to further overall discourse.

I stand by my assertion that a classical collateral secured transaction is fundamentally different than a repo as described.

I'm open to having my mind changed. Just be aware, I don't look at transactions of value as only being financial in their entirety unless explicitly set out. Other forms of circumstantial value can and do emerge.


> EDIT: Downvotes without substantive replies or explations do nothing to further overall discourse

Please don't do this. It adds no information and the site guidelines explicitly ask you not to, as you'll see if you read them to the end: https://news.ycombinator.com/newsguidelines.html. It also guarantees further downvotes, and this time they'll be correct because you broke the rules.


Thought I caught it before the edit window closed. Missed it by || that much apparently.


I don't understand what you mean, but if you mean you want to edit that bit out, let us know at hn@ycombinator.com and we can reopen the comment for editing.


> a classical collateral secured transaction is fundamentally different than a repo as described

You shouldn't be downvoted. The observation is correct.

In a collateralized loan, the borrower owns the collateral. In a repo, the borrower does not. This is one reason why companies finance capital equipment with secured loans, not repos.

A repo also involves two collateral purchase-and-sale transactions; a collateralized loan involves zero. This restricts it to highly liquid assets, where everyone agrees on prices and settlement is fast and cheap. In practice, this almost exclusively means Treasuries.

I was handwaving when I said they're the same transaction. They're symmetrical, cash-flow wise. But you can't replace secured loans with repos or vice versa. Repos are secured borrowing with minimal counterparty risk. Secured loans are secured borrowing with broad asset coverage.

> a classical collateral secured transaction is fundamentally different than a repo

Not fundamentally. Just different. (I'd also note that the "classic" collateralized transaction is a spherical cow.)

Lender leasing the collateral to the borrower makes a repo functionally identical to a secured loan. Lenders requiring ring-fenced collateral (e.g. in a cash-backed loan) are repos with extra steps. This symmetry belies their fundamental similarity.

There are legal differences in edge cases. But fundamentally, they're both collateralized borrowing.


Financial securities have little utility past their monetary value.

Off the top of my head, you give up on any cashflows, such as dividends/interest from the security and maybe also some anciliary rights, such as voting rights or whatever rights bondholders get.

However, you account for forgone cashflows by adjusting the repurchase price accordingly. As for the rights, you can recall the repo (it's usually done overnight).


>Voting rights

I'd say that represents a very different transaction Still depending on where or not an important vote is coming up in the meantime, and who you are selling your voting stake to.


Indeed. If the asset produces a revenue stream itself, e.g. coupons or dividends, things get complicated fast.


That definition doesn't seem right. The repo rate is just the rate paid on repurchase agreements, a type of risk free, short term loan where collateral is exchanged for cash, then exchanged back again. Central banks shouldn't be involved. The Fed is only involved now as an emergency provider of liquidity.

https://en.wikipedia.org/wiki/Repurchase_agreement?wprov=sfl...


Confirmed, yardie's definition is straight up wrong.

EDIT:

It's worth noting that the collateral is usually government bonds (this is called "general collateral", or GC), with bonds from quasi-government bodies like Fannie Mae or the World Bank being the second most common kind.

One way to look at repo is that it's a way for companies with a big stock of bonds (eg banks) to pawn them to borrow money. This is usually the cheapest option for them to borrow money at scale, and so the repo rate is seen as an indication of banks' financing costs.

Another way to look at it is that it's a way to buy government bonds on credit - you can buy a bond and then immediately repo it, using the money you've borrowed to pay for the bond. You have to pay the interest on the repo until you sell the bond and pay it back, but you never have to come up with a big pile of cash.

I have no idea which use case is more common.


> Another way to look at it is that it's a way to buy government bonds on credit - you can buy a bond and then immediately repo it, using the money you've borrowed to pay for the bond. You have to pay the interest on the repo until you sell the bond and pay it back, but you never have to come up with a big pile of cash.

And the reason you'd do this is... to lock in yield in a declining interest rate environment?


For example, if the yield on a long-term bond is 3% and repo rates are only 2%, you might buy the bond, financed with a repo transaction, and bet that over the term of the repo, the bond’s price will hold up enough to allow you to profit from the transaction. If it is a one year repo, the bond could decline in price by 1% and you would still make a profit because its yield is higher than the cost to finance the purchase.

A more relevant example is if futures are trading rich relative to bonds - say they are too expensive by 1/32th (about 0.03%). In this case you buy the bonds on repo and sell futures against them, expecting to profit when the price gap closes. Of course, 0.03% is not much profit, so you use 50x leverage (which you can easily do on repo, because it is secured borrowing) turning it into a 1.5% profit.


There are as many reasons to do it as there are reasons to buy bonds!

The five-year US treasury currently yields 1.658%, and SOFR, a measure of repo rate, is currently 1.55%, so that looks a bit like free money. As long as interest rates don't go up.

You might be selling bond futures, and want to cover your position.

You might think some specific bonds are undervalued relative to others, in which case you can buy the undervalued ones on repo, short-sell some overvalued ones, and wait for the market to correct itself.

I'm sure there are many far more ingenious things you can do with bonds.


So, stupid question: what's the purpose of suppressing short-term interest rates like that? Isn't that a meaningful market signal that banks should be running in a way that's less dependent on loans from other banks, just as a higher price for fish correctly signals market participants to look for substitutes?

(Side note: as someone who "should" benefit from more expensive liquidity as a holder of a money market fund, I looked up the Vanguard Prime MMF, and it turns out it doesn't hold any repos[1], even though that's a valid asset class for MMMFs[2]...)

[1] https://investor.vanguard.com/mutual-funds/profile/portfolio...

[2] https://www.pimco.com/en-us/insights/investment-strategies/f...

>New SEC rules require that government money market funds hold 99.5% of assets in government-related securities, including Treasury bills, agency discount notes and repurchase agreements (repos).


Since the borrowers could sell the collateral instead, technically they aren't depending on these loans and they could stop using the repo market.

However, they would then 1) lose the yield on the bonds and 2) instead need to keep large amounts of money on some account and trust that the bank doesn't go bust. Bond holding is used as an alternative to deposits.

The government also doesn't want the repo market to shut down, because that means everybody dumps their gov bonds, since they wouldn't need them as collateral anymore. The dollar amount of outstanding repo contracts is measured in trillions so this would imply a catastrophic increase in the cost of borrowing for the government.


I'm probably way incorrect. But the article kept talking about repos and the repo rate without telling the reader what it actually means. I just went with whatever was the highest indexed in my search. I don't work in finance beyond layman so my understanding of finance domain specific lingo is very limited.


* And some would say to sneak in a little QE without calling it QE


It's not QE, unless everything is QE. When the Fed targets a short-term rate, this is just ordinary monetary policy.


Most repos involve banks, investment banks, money dealers and various investors.

>What Is a Repurchase Agreement?

>A repurchase agreement (repo) is a form of short-term borrowing for dealers in government securities. In the case of a repo, a dealer sells government securities to investors, usually on an overnight basis, and buys them back the following day at a slightly higher price. That small difference in price is the implicit overnight interest rate. Repos are typically used to raise short-term capital. They are also a common tool of central bank open market operations.

https://www.investopedia.com/terms/r/repurchaseagreement.asp


Maybe I'm naive but I assumed the $120k banana "sale" was essentially a fake transaction intended to attract publicity and drive foot traffic to the venue.


That or a money laundering scheme.


I'm not a professional at money laundering, but under most circumstances, you'd want to do it without attracting attention.


There was a decent podcast on this about a month back, on Bloomberg "Odd Lots": https://www.bloomberg.com/news/articles/2019-11-11/this-is-w...


I listened to this podcast multiple time, and found it quite hard to follow.

I don't have a heavy background in finance, but in general I have a good enough understanding of the subject. In this case however the language was somehow making it harder than usual.


Any specific parts you’d like to understand better? I’m sure someone here is well versed in this topic.


I’ve listened to that episode (after another recommendation here, a few days ago).

Didn’t find it very informative or good. The podcast discussion was opinionated and obtuse.


Everyone should listen to this podcast. It is short and informative. I've learned a lot about the difference between fiscal and monetary policy.


> Apparently, money is meaningless.

I am not going to argue in favour of the banana. But I am going to argue in favour of (expensive) art.

If I had only $10 and $1 would buy me food and shelter, then I would pay $9 for a Monet or van Gogh. Why? Because if I never see or possess an actual Monet or van Gogh, how can I verify things about the past? How can I verify anything? How can I have a physical and complete connection to the past?

To take a more historic example, if anything new (and real) about the man called Jesus is discovered, it would be equally valuable to nonreligous people as to religous people. Suppose for a moment that you find Jesus's diary and in it he wrote: "Can people please stop writing four separate accounts of how I am the messiah; I am not." Wouldn't the change that would bring about be more significant than the amounts of money spent in various contexts? For the record, I don't think the Jesus example matters much in 2019. But back in the Renaissance, our connection to the past (and the past's knowledge) was critical. Maybe Euclid's elements is not "art" to you but it sure as hell should be preserved as though it were art.

Art and artifacts are to history as experimentation is to science.


> "Can people please stop writing four separate accounts of how I am the messiah; I am not." Wouldn't the change that would bring about be more significant than the amounts of money spent in various contexts?

That would be in line with passages like John 1:3 where John insists the purpose of Jesus's piety isn't so he can be messiah, but rather so that the rest can follow his example and become like him. So Jesus claiming there should be no difference between him and us would be consistent with other passages in the bible.


Alright; a historical artifact that supports that (or any other claim) would be intrinsically valuable.


Didn't some comedian eat that banana and they replaced it saying the banana was not part of the art?


> a real banana and duck tape sold for $120,000, 3 times.

There's also the urinal that sold for $1.7M in 1999 https://en.wikipedia.org/wiki/Fountain_(Duchamp)#Art_market


"New tallest skyscraper in the world" has long been recognized as a sign of impending financial collapse. I wonder if one could objectively evaluate "inartistic garbage selling for high prices", as a means of predicting the same? Esthetics is hard to rate objectively, but there would certainly be a consensus agreement on both of these examples.


My intuition tells me that the "new tallest skyscraper in the world" and overpriced esthetics aren't so much signs of an impending financial collapse directly, but they're rather a sign of financial security during the past several years. The problem is that the business cycles are... cyclical, which means that long-lasting financial security is likely to be followed by a financial downturn (collapse). Thinking about it more, I'm not even sure what the difference is.


Interesting point. There is a school of thought that frequent, shallow and short recessions are preferable, because if you don't have them you will pile up problems that all fallout during a deep, long depression. If this is true, then as you say, the skyscrapers, ridiculous art prices, etc. are just the most visible indicators of "lots of easy money, being spent stupidly", which is both a result of prolonged financial security and a bad omen.


A replaceable banana.


The banana was eaten. “Hungry artist”


> 'finance isn't in your DSL'

That's a great turn of phrase, and a nice mental model!


What is DSL?



Why did you go all that way to see a banana and a piece of duct tape? You could have admired that at your local grocery store.

That's the reason it sold for $120k.


I always assume those stupid art bus are a mix of idiots and money laundering.


> finance isn't in your DSL

DSL? Not sure I'm familiar with that term. Some tech jargon?

\s




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: