Hacker News new | comments | ask | show | jobs | submit login
As German Bund Yields Head to Zero, They Still Beat U.S. Treasuries (bloomberg.com)
68 points by paulpauper 6 days ago | hide | past | web | favorite | 66 comments

What I didn't get from the article is a "so what". Why does this matter? Or maybe they did explain why it matters, and I just don't have enough knowledge to understand.

What makes this article a bit confusing is that they give an answer without making it clear what the question was.

Here's the missing question: Imagine you are a Eurozone pension fund or life insurer. You have to invest billions of euros very safely and eek out a return that is no less than inflation. How do you do that?

Inflation (HICP) was 1.73% in 2018. German Bunds, the safest euro denominated investment with sufficient volume, yield close to 0%. Now you look across the Atlantic and find that 10 year treasuries yield 2.63%.

The problem is that simply investing in treasuries wouldn't work because currency fluctuations would likely dominate any interest income. So you would need to hedge the currency.

The Bloomberg article is about why hedging the EUR/USD currency pair is currently too expensive for this idea to work.

It matters for European savers, including everyone who is paying into a defined contributions pension (e.g most workplace pensions). A significant chunk of those savings currently yields negative returns. The alternative is to take a lot more risk than you want to or are allowed to take.

So with EUR/USD hedge the assumption is that USD is going to drop or become more unstable in the future?

I don't think anyone assumes the opposite, that the EUR will rise as a result of strong economics performance.

What if you simply don't want to take a directional bet on that?

I don't find it exactly reassuring if everyone is already thinking the same thing.

Being long one currency, is an implicit short against the rest whether one engages in fx/swaps/cfds/forwards on a pair or not.

Absolutely, but if an institution is taking some form of deposit in euros and promises to make predictable payments in euros then being long euros is not a directional bet for that institution.

Of course the picture gets a little murkier once you start considering inflation caused by exchange rate fluctuations. But that is probably not the main concern in a large currency area with an independent central bank.

We must have different ideas of what fx fluctuations are, +/-500 bps over a decade sure… not +/-500bps on -30% decline line over a decade vs a major pair…

Yeah it has to be the assumption, but at the same time it makes one wonder why EUR/USD hedging is increasingly expensive in the first place… though if one pulled up a monthly chart going back 10 years and their head wasn't tucked between their legs, its easy to see why.

"The alternative is to take a lot more risk than you want to or are allowed to take"

What DC scheme wouldn't let you put say 100% in shares?

Or if not you could take the currency risk on yourself, and buy US shares.

Want to take? That's an interesting question. I would contend that if bonds are yielding 0.08%, they're probably overpriced and thus too risky to buy.

I say that as some one 100% in equity with a long time to retirement, so to each their own.

>What DC scheme wouldn't let you put say 100% in shares?

I'm looking at it from the point of view of fund managers. If the fund promises to invest 50% in bonds then someone's job will be to find suitable bonds.

If you're young and you are free to choose how you want to save for your pension then I agree that stocks are probably safer. Perhaps even if the whole world slowly turns into Japan.

Why bother investing it at the point? What’s the advantage over holding cash?

Because it's a non-negative return

Because the ECB had negative rates for a while, i.e. some parties had to pay to hold cash.

And lastly, because economic developments which don't affect cash, may affect bonds. i.e. if the interest rates drop, existing bonds will become more valuable. I don't expect this to happen, but the point is that you don't just buy bonds for their return today, but also in expectation for their relative return compared to the rest of the bond market in the future, i.e. their future value. That doesn't apply to cash in the same way.

> What’s the advantage over holding cash?

Where do you hold the cash? Physically? As deposits in a bank? “Cash” is short hand. This article debates different forms of cash. (OTR sovereign bonds are usually considered cash.)

It's deposits in a bank, yes. Many banks charge large deposit holders a negative interest rate.

The alternative of holding it in actual cash, comes with security costs, storage costs,... The negative interest rate is still cheaper then the 1% annual cost of securing banknotes. It also puts an implicit floor on how low the central bank can lower the interest.

> Many banks charge large deposit holders a negative interest rate

Counterparty risk dominates, for most treasurers. You don’t want your billions of corporate deposits in the bank that went bust. Treasuries—or their local equivalent—are a safe way to hold cash.

How is a bond the same thing as cash? I can't walk into a supermarket and buy a loaf of bread with a sovereign bond.

In the world of finance, "cash" is often shorthand for "cash or cash equivalents".

Instead of the usual layperson interpretation where cash is "a bunch of currency, either physically in my hand or in a bank account", the meaning here is "stuff with incredibly low risk" i.e. you can be very sure that you'll not lose money over time.

Sovereign debt, including bonds, is regarded as the safest type of debt and therefore falls into the 'risk-free' bucket in the finance world. Since cash is usually regarded as risk-free, the word "cash" has come to be a short-hand for "(almost entirely) risk-free assets".

Yeah, but using the word "cash" that way isn't a good idea. I understand why they do it but it's still asking for trouble.

For one, sovereign bonds are very obviously not risk free. Governments have a long history of going bankrupt or giving bond holders haircuts. They're treated as such because regulations passed by governments force them to be treated as such, which is clearly self-serving. The finance world is mixing up "we must pretend it's risk free" with "actually risk free".

"Cash" is also not risk free. Probably higher risk than US, Japanese, German, or British bonds. In the current fiat era a country is more likely to print money than default on its bonds. Countries that have recently gone bankrupt or gave bond holder haircut, rarely can borrow in their own currency anywhere near the rates on US dollar denominated loans.

Printing money to pay a debt is defaulting in any sense that would be understood by a layman. The loan has technically been paid, but in a way that voided the point of requesting payment (transfer of value).

Again, the fact that it's not considered that way says more about the finance industry than it does about ordinary terms like cash.

Credibility claimer: I built risk systems for 3 Tier-1 investment banks in a previous life.

I'm not a fixed-income (read: bonds and other interest-rate derived financial instruments) person, but I think the erosion of 'value' via inflation is missing the point.

Sovereign debt is the least risky form of debt there is, period. Everything else that's denominated in the same currency has more risk, not least due to the fact that the pricing of every financial instrument takes the risk of the sovereign debt into account aka the risk-free-rate.

Yes, the absolute amount of risk of a sovereign debt is different depending on who controls a particular fiat currency - but in relative terms, any financial model will start with that baseline. Obviously, cash in that same currency bears some risk due to inflation, which is why people talk about 'real interest rates' that take that into account.

If you lend someone money, you take risk, and you get interest as compensation. That risk includes your predictions of inflation, and the interest rate you are prepared to accept should include that too. If you don't like it, don't lend the money.

Arguing about a 'transfer of value' implies that you have another way of representing value that is better than the currency itself. I can't think of one, but I'm open to suggestions.

Currency is a fine way to represent value, but a currency that's being inflated is certainly a worse way than a currency that isn't. If a government plans to inflate its way out of debt, the markets respond to that by trying to charge higher interest, but ultimately governments (outside the eurozone) control their own currencies and their own inflation statistics, so the markets are always playing catchup.

A truly risk free currency would be a currency that couldn't be inflated by a government to escape from its debts, like a cryptocurrency (if they actually worked).

For most of history there was gold and/or silver. Fiat currencies have always been short lived. Our current fiat phase started in the 1970's so maybe it can survive a little longer?

Because cash yields -1.73% at best. Perhaps less if negative interest is charged on large deposits. And where do you store it safely?

And this is a big part of the reason the majority of pension systems in Europe are going to be bankrupt within 10 to 15 years. (The other part being the inverted population pyramid)

You shouldn't make statements about the future with such confidence, it is impossible to predict what will happen with any real degree of certainty.

That's from projections of population and liabilities. A huge amount of people will retire in 10-15 years.

[1] https://www.populationpyramid.net/western-europe/2017/

There aren't that many defined benefits workplace pension schemes left. Some have already gone bankrupt. Others were restructured into defined contribution schemes.

And state pension schemes are not usually investment based at all. They are usually funded by current social security contributions and taxes. If they go bankrupt it's for political reasons.

More likely than bankruptcy is that many people will receive lower pensions than they had hoped for.

It probably doesn’t to most people. But since Bloomberg is primarily a financial services company, they write articles for finance professionals (mostly).

Even so, I think it’s a little disingenuous to compare EUR denominated T-notes with Bundes (which are natively EUR denominated). Of course if they just compared regular T-notes, with Bundes, they really wouldn’t be a story since T-notes are paying a great rate compared to any other government bonds.

It keeps artificial support under the euro. If it were to reverse any time soon you could see a big shift in the dollar vs the Europe. There are some early signs this might occur with changes in LIBOR recently.

It answers the question: „I am a European investor and want to just park my money without any risk. But US bonds pay much better interest. Can I just buy them?“

Euro-hedged US treasuries

Vs unhedged bund. Pretty lame.

Yeah I’m not sure they are really comparable. Most foreign governments buying T-notes are not fully hedging their currency exposure. As the Bloomberg article demonstrates, doing so heavily eats into your returns. You are paying an expensive premium, it would probably just be a better idea to buy USD denominated T-notes and deal with the currency exposure down the line.

Can someone with a finance background explain why this is happening?

Is the issue that there is loads of money in the system with not enough productive and/or low risk opportunities to invest it, so people are basically wiling to accept close to a zero return?

I own German Bundesanleihe and I wonder what stops anyone buying units that were released 20 years ago. Some of them yield around 4%. Of course they expire within the next 4-8 years which is still enough time to profit. Maybe I'm missing something...

What you’re talking about is the coupon, not the yield. Say you have Bunde that was sold at 100 EUR. The coupon is 4 EUR. So the coupon is 4%. The yield however depends on the current market price. To calculate the yield you take the coupon and divide it by the current market price. Because bonds mature into each other, that is a 5-year issued 4 years ago should be the same as a 1-year issued today, both bonds will have the exact same yield.

So effectively all the old bonds turn into new bonds, all linked to the same yield. This is why yields can be negative, it’s not like a negative yield bonds means you are paying someone. It just means that the price is so high the effective return is negative (remember that you get the face value of bond back after it matures)

One exception to this rule is newly issued bonds. Because of regulations and rules, some institutional investors must buy bonds direct from the treasury. This means that they often trade a premium (they have a lower yield) compared to old bonds.

Bonds can be a little counterintuitive because when bond prices go up, yields go down. So negative yields actually mean that market is very bullish.

You might then ask, why would anyone purchase a negative yield bond? Well, there are a couple different reasons: a) they have to, due regulations or internal rules. b) they are speculating on interest rates or currency. Or c) they are speculating on the bond market (related to b).

Bonds can be complicated because they have exposure to a lot of factors: rates, inflation, equities, debt, etc. But they are useful instruments because they have intrinsic value: they simply pay a fixed coupon and then the principal after N-years.

Just to add on this... for most investors, they provide some backbone and counterweight to market cycles. In terms of safety, they are a safer investment choice because they represent debt, not equity.

This is a good example, graphed: https://www.google.com/search?tbm=fin&q=MUTF:+VBMFX&stick=H4...

Bonds Summarized:

a) are less volatile compared to equities

b) provide consistent and guaranteed returns, albeit at a lower rate compared to equities

c) often work counter to market cycles: when interest rates rise, bonds fall in value, and when interest rates are cut, bonds rise. Depending on the depth of the fund, you may not see any changes at all.

Generally speaking, stocks almost always perform significantly better than bonds over long periods. If you can handle the risk, in theory 100% stocks is a better investment. Bonds provide the "don't freak out and sell" part of your portfolio.

A couple on interesting points:

a) Bonds, as you said, were traditional seem as counter cyclical, when equities or rates went up, bonds went down, and vice-versa. Nowadays, it seems like this might not be true anymore. So much money is pouring into the US securities market that both often now move in tandem. In the past, in risk-off environments (when you want to lower your risk), investors would move out of equities and into bonds. But now, since so much money is foreign, you see outflows from both bonds and stocks is a risk-off environment. This makes bonds a worse hedge today against equities than in the past.

b) It's true that 100% stocks almost always outperforms 100% bonds. But things get really interesting when you consider a mix of equities and bonds. Take an 100% equities portfolio of say 10 stocks: do a simulation in which you vary the weights of each position (the percentage of each stock that makes up your portfolio). Now graph the historical returns on Y axis and the historical volatility of all these simulated portfolios. What you'll see is called the "Markowitz Bullet"[1] The efficient frontier (EF), is all the portfolios that have the best risk adjusted returns (also known as the Sharpe ratio). Basically what that means is that those portfolios on the EF (the top "edge" of the bullet) have better risk/reward characteristics than any of the portfolios below them. so with a 100% equities portfolio, the best you can do to maximize your Sharpe ratio will be on the top edge of the curve.

Now consider mixing in some bonds which are at the risk-free rate. 3 month US T-notes are traditionally considered to be the risk-free rate. That is, they have no volatility and a fixed return (say 2%).

By adjusting your allocation of risk-free bonds and equities you can achieve a better Sharpe ratio than anything you could do with just equities alone. So now, with adding bonds, you can now beat the EF. The EF is now the line drawn between the risk-free point (0 variance, 2% returns) and tangent portfolio (the highest point on the curve). Check out the picture below for an illustration.

But it gets even better! If you can borrow at the risk free rate, you can now your leverage returns (but also volatility) into outer space!

Modern portfolio theory is based on this concept that adding the risk-free rate to your portfolio can actually make you achieve better risk-adjusted returns than you otherwise could with just equities alone. Seems pretty magical when you think about it (though it's predicated on the original sin of finance: that the risk-free rate actually exists)


It's not about buying a return, it's about buying stability.

If you've got a billion dollars/euros to manage, and assuming you can't just stuff notes in a mattress and hope silverfish don't eat them, you also can't just dump it in consumer-grade investments (typical bank insurance wouldn't cover most of it if the bank collapsed). Virtually any combination of stock/bond/tangible/land market investments, no matter how diverse, has non-trivial risk of significant loss over some time horizon.

In contrast, the most premier government bonds (US, German, maybe Japanese and/or Canadian) have a huge "too big to fail" factor. If those suddenly lose 30% of their value, it probably means the world a a whole has far bigger problems than your investment portfolio.

If your goal is asset preservation above all else, it's worth eating a known, predictable zero or negative rates of return on those bonds, as a safe harbor from the much bigger risks of the broader market.

> You might then ask, why would anyone purchase a negative yield bond? Well, there are a couple different reasons: a) they have to, due regulations or internal rules. b) they are speculating on interest rates or currency. Or c) they are speculating on the bond market (related to b).

d) they expect them to go down even further

That’s what “speculating on the bond market” means.

I used to trade on a fixed income desk, which includes trading bunds.

What you are quoting is the coupon, not the yield. What this means is that back then if you bought 100 worth of the bond, you'd get a promise from the government to give you 4 back each year, plus the principal when the bond matured.

It is not quite the same as the yield.

Now consider if later a bond came out that paid 8 each year per 100 that you invested. The ones that pay 4 are still around. What would happen?

Well you would be able to buy the old 4 coupon bonds at 50 from the old buyers (the government will still give you 100 at the end). That is the price that makes the yield the same.

Why is this the case? Why don't the people who bought the 4s simply insist on getting 100 for their bonds? Because if they did, everyone would just spend their 100 EUR on the 8s. And that would continue to be the smartest thing to do until the 4s yielded the same as the 8s.

Simplified a bit, so no convexity, term premium, on-the-run premium, repo rate, futures, basis and so on. But worth googling if you're interested.

So you have a €100 bund that will yield €4 interest at the end of the year, plus the €100 of the original capital that will be sold back to you (it is a 19-year-old 20 years bund, so it will be paid back at the end of the year), so you will be paid €104 at the end of the year. At the same time, new 1 year bunds cost €100 and will be payed back only €99.9.

As a lucky owner of a luctrative bund, why would you sell it €100 on the market? In less than a millisecond, you could find people that would happily buy it €104.1 and get almost all their money back, guaranteed, at the end of the year.

As a client, you are as likely to find a fool that will sell a €104 check from the German treasury for less than €104 as to find a fool selling a €100 banknote for less than €100 (actually, the latter is more likely, for liquidity reasons).

Actually, in your example, the traders would be all over your 10 year bond.

The 1 year bond is yielding -0.1%. The 10 year bond is yielding more (-0.096%).

I don't get it. Why would anyone buy a 99,9 return for 100 or 104 return for 104,1. Are the bounds a token for large bills or what?

If 100'000 € bills existed, would founds etc. get those instead?

Is there a tax on cash or something.

Finance always seems like a big scam to me. It doesn't make any sense and the truth is protected by a guild of banksters under layers of complex hard to access rules and what not.

If you have a billion € in cash, even in the form of a 1 billion € banknote, how do you store it? It has a cost, for obvious security reasons. You cannot put it on a bank account either, because putting money on a bank account is equivalent to having a 0% bond with a non-risk-free bank: you don't own your money, you just lent it to the bank, if it fails, you lose your money.

Having a treasury bond nowadays is the same as having money on a bank account, it's just a line of credit written somewhere, the huge difference is that if the organism recording the line of credit (a broker, for instance) fails, you still own your bund.

Oh, that actually makes sense. It atleast explains why interest can be nagative on bunds. It's money with a back-up copy in e.g. the Central Bank if you lose it.

Why make the fuzz and having bunds in the first place and not just have the Central Bank offer accounts with a card connected? The bund system seems like an arcaic paper thing that made sense 40 years ago.

> Is there a tax on cash or something.

Something. It is far from trivial to store significant amounts of cash. If it gets stolen, lost or destroyed, you basically have lost your money. Also, significant amounts of cash are not that liquid anymore. It takes time to get that from your vault and deposit that to a bank. Cash is just a pain in the ass in many ways.

You don't get it so it it's a scam?

It seems like a scam.

I'm feeling scammed when I'm buying stock with a 30 min delay on pricing and orders becouse I'm not hooked up to the really expansive investor network.

I'm feeling scammed when I can't buy real estate obligations directly but have to indirectly buy it via some expensive fund.

I'm feeling scammed when Master Card charges x.x% of every transaction I do with my bank card and it doesn't even show up on the recipe.

But maybe it's just me. It's not really a scam since I know I'm being scammed, than it's just business I guess.

The answer is that no work in life is free.

Everything you pointed to is a transaction cost (or, equivalently, a limitation to reduce the number of actors to a manageable number). You're paying for the work to make that thing happen.

And, thank god, financial and medical industries are largely precluded from offering free services (a la Google) in exchange for selling every scrap of your data.

Ye, well agree I get that. But the price of the banks services is in no way justified by the value added?

A 2% card fee is 24 minutes off a 40h work week spending half your wage with card. It's like going to the physical bank and withdrawing money saves you time if you are near by anyways (pretending the card fee was added to the recipes as a cost).

If you make 1000 USD a week, that's 10 USD or having 20 letters sent all over the country being delivered by hand every week.

Card purchases are not being manually reviewed by the bank anymore like the old photo copy ones, but we still pay for it.

It's just insane how this arcaic system prevails and how much profit can be squeezed out off it.

I guess with card fee's the main problem is that the consumer is not paying it directly, and the companys don't have the guts to openly put the cost on the consumer.

To me, it sounds like getting fast food and being amazed at how much the french fries and soda are selling for compared to the cost. Businesses don't make equal profits on everything they do, and most businesses have a lot of overhead that isn't directly linked to transactions, but is just the basic necessity to even be in business.

So I think your intuition that the fees aren't linked directly and proportionally to services is correct, but that's true of all businesses and it doesn't mean the profits are unreasonable overall.

Credit cards, at least in the US, rebate up to 2% or so of fees back to some customers, so they aren't charging all that much for their services.

Credit unions and banks that I've dealt with provide basic services like checking for the price of the float on your deposits, which has to be extremely small these days.

Simple. You can only get old bonds on the market, where that surplus is priced in.

To buy (on the open market) a 100 EUR zero-coupon bond issued by Germany, with 4 years left, you would need to pay more than 100 EUR. i.e. your reward for waiting for 4 years would be that you'd get almost all your money back.

10 year bonds yield just over zero. 5 years and lower yield even less than zero:



A $100 bond is released paying 4% ($4 per year). If the prevailing interest rate drops than the bond price will increase to compensate. Say if the prevailing interest rate drops to 1%, then you will need to spend much more to buy that bond. Now it probably wont be as high as $400 because when you redeem the bond you'll only get back $100 thus suppressing your effective yield further , but that's the basic gist.

The higher yield is already priced in. You are not the first with this idea :-)

those are above par

Basically everyone is desperate to invest money somewhere.

Bonds are one of the safest instruments and often pension funds are legally obligated to invest at least a proportion of funds in safe, not risky instruments.

This will not end well IMO.

TIPS bonds are one of the safest instruments. A regular bond can be as risky as stocks, whether traded short term or held to maturity. What people get with bonds, depending on the rating, is something short of a guarantee they'll get principle back with the agreed upon interest - but can you really accurately assess risk over 10 years? 30 years? There's widespread agreement that corporate bonds aren't paying enough compared to the risk, and yet people continue to invest in it.

I would argue that a "regular" bond refers to government-backed securities, basically AA at a minimum. When people want to talk about corporate bonds they're careful to refer to them as junk bonds.

Literally none of this is correct. Bonds do not inherently or even typically refer to government securities. And not all government issued debt is AA, for example Mexico bonds are currently rated BBB+. Junk bonds are speculative rather than investment grade, quite a lot of corporate bonds are investment grade.

Junk bonds (high yield) are a small subset of corporate bonds. But it’s true that even investment grade debt is not as good as it used to be.


German Bund is clearly something totally different than German Bonds.

Applications are open for YC Summer 2019

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