Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Discuss all general (i.e. non-personal) investing questions and issues, investing news, and theory.
Post Reply
User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Mon Oct 02, 2017 10:24 am

I've seen a graph showing that the yield on 10-year US Treasuries (2.33%) is now higher than the yield on European Junk Bonds (2.32%) (unfortunately I am not savvy enough to be able to insert the graph here - showing the evolution of the respective yields).
Anyway, how can this be consistent with the idea that assets are rationally priced? European Junk Bonds are much more risky; yet they yield less! And if currency exchange rates are efficient, this cannot be due to people thinking that the Euro will appreciate versus the dollar.
So how do you explain this? If I find out how to insert images, I'll post the graph later. :happy
When everyone is thinking the same, no one is thinking at all

Valuethinker
Posts: 33428
Joined: Fri May 11, 2007 11:07 am

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Valuethinker » Mon Oct 02, 2017 10:55 am

Lauretta wrote:
Mon Oct 02, 2017 10:24 am
I've seen a graph showing that the yield on 10-year US Treasuries (2.33%) is now higher than the yield on European Junk Bonds (2.32%) (unfortunately I am not savvy enough to be able to insert the graph here - showing the evolution of the respective yields).
Anyway, how can this be consistent with the idea that assets are rationally priced? European Junk Bonds are much more risky; yet they yield less! And if currency exchange rates are efficient, this cannot be due to people thinking that the Euro will appreciate versus the dollar.
Yes. It can. Currency forward rates reflect the differential in interest rates.

Within a currency, risky bonds are priced against the risk free bond in that currency (same maturity). You don't compare a US Treasury to an EUR junk bond.

Risk free bond in Eurozone is nearly at negative yield- -some of the shorter dated German bunds are trading at negative nominal yields.

If you buy the 10 year UST and a forward contract selling USD and buying EUR in 10 years time, the return will be very close to buying the German 10 year bund (they should be identical, in practice there are small variations but generally Covered Interest Parity holds between the main currencies)*.
So how do you explain this? If I find out how to insert images, I'll post the graph later. :happy
As above.

* there are differences because the yield curves have different shapes and because the market has slightly different views of the default probability of UST v. bunds. If you use 0 coupon bonds they should be essentially identical.

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Mon Oct 02, 2017 11:10 am

Valuethinker wrote:
Mon Oct 02, 2017 10:55 am


Risk free bond in Eurozone is nearly at negative yield- -some of the shorter dated German bunds are trading at negative nominal yields.

Yes, I know, French bonds too. That's why buying an All maturity Eurozone bonds ETF doesn't make any sense to me: all the shorter duration bonds from France and Germany (and even Italy below 2,5 yrs) within the ETF are by definition losing money!...
Perhaps the only bonds that might make sense to EU investors are longer duration Governement bonds, as a hedge in case stocks go down.
Anyway thanks for the good explanation.
When everyone is thinking the same, no one is thinking at all

garlandwhizzer
Posts: 1672
Joined: Fri Aug 06, 2010 3:42 pm

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by garlandwhizzer » Mon Oct 02, 2017 1:12 pm

I think Lauretta makes a good point. Pricing of bonds between US sovereigns and EUR sovereigns doesn't make any sense at current interest rate differentials IMO. Part of this is that the central banks in EUR and the US are in different points of the interest rate cycle: decreasing interest rates in the EUR, a tailwind for bond principal value, and increasing interest rates in US, a headwind. That doesn't answer the question completely however. The next question is why don't those who buy German of Swiss bonds at 0% yield, buy Treasuries instead? Some do, but not as many as efficient market theory would suggest which would be near 100%.

I believe the answer is that markets and asset pricing in general is not required to make sense to us at all times. EMH is usually a pretty accurate hypothesis but periodically there are large departures from what seems to us rational behavior. Bubbles are an example of this as are IMO current bond spreads as above. Markets are driven by humans and humans often fall well short of rationality. Lots of things about markets don't make sense--the low volatility premium for example which supposedly produces higher return with lower risk. How does that make sense?

Garland Whizzer

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Mon Oct 02, 2017 1:44 pm

garlandwhizzer wrote:
Mon Oct 02, 2017 1:12 pm
I think Lauretta makes a good point. Pricing of bonds between US sovereigns and EUR sovereigns doesn't make any sense at current interest rate differentials IMO. Part of this is that the central banks in EUR and the US are in different points of the interest rate cycle: decreasing interest rates in the EUR, a tailwind for bond principal value, and increasing interest rates in US, a headwind. That doesn't answer the question completely however. The next question is why don't those who buy German of Swiss bonds at 0% yield, buy Treasuries instead? Some do, but not as many as efficient market theory would suggest which would be near 100%.

I believe the answer is that markets and asset pricing in general is not required to make sense to us at all times. EMH is usually a pretty accurate hypothesis but periodically there are large departures from what seems to us rational behavior. Bubbles are an example of this as are IMO current bond spreads as above. Markets are driven by humans and humans often fall well short of rationality. Lots of things about markets don't make sense--the low volatility premium for example which supposedly produces higher return with lower risk. How does that make sense?

Garland Whizzer
Thanks for your input. And yes, I am also puzzled by the low volatility premium.
When everyone is thinking the same, no one is thinking at all

jalbert
Posts: 2297
Joined: Fri Apr 10, 2015 12:29 am

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by jalbert » Mon Oct 02, 2017 2:13 pm

Comparing the expected real returns normalized to a common currency will reflect market efficiency. Different currencies have different expected inflation levels.

If you hedge a Euro-denominated bond back to USD with rolling 30 or 90-day contracts you will (as of this writing) get about a 95 bp hedge return to supplement the yield based on the spread between USD and Euro risk-free rates. You can thus add about 95 bp to the yield of a Euro-denominated bond to calculate its risk premium relative to the USD risk-free rate.

European term premia have been very low, reflecting low inflation or deflation risk, hence the low rates. Lower or negative expected inflation corresponds to a positive expected return for the Euro relative to currencies for which expected inflation is higher, such as the USD.
Risk is not a guarantor of return.

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Mon Oct 02, 2017 3:04 pm

jalbert wrote:
Mon Oct 02, 2017 2:13 pm
Comparing the expected real returns normalized to a common currency will reflect market efficiency. Different currencies have different expected inflation levels.

If you hedge a Euro-denominated bond back to USD with rolling 30 or 90-day contracts you will (as of this writing) get about a 95 bp hedge return to supplement the yield based on the spread between USD and Euro risk-free rates. You can thus add about 95 bp to the yield of a Euro-denominated bond to calculate its risk premium relative to the USD risk-free rate.

European term premia have been very low, reflecting low inflation or deflation risk, hence the low rates. Lower or negative expected inflation corresponds to a positive expected return for the Euro relative to currencies for which expected inflation is higher, such as the USD.
ok but what I don't understand is this: a European investor who buys e.g. a 1 yr US treasury bill (currency unhedged) will have a yield of 1.3% (I think) in dollar terms. At the end of the year he will have 101.3 $ for each 100$ invested. Since a German investor would have actually lost money if she had bought a 1yr German bond offering comparable guarantees and for each 100 Euros initially invested would now have aprrox. 99,3E, does that mean that the market is actually predicting that the Euro will have appreciated against the dollar by roughly 1.5-2% over that year? Otherwise, if the market is agnostic as to which way exchange rates will go (and there is a 50%-50% chance that they'll go each way), it seems that a EU investor should buy US bills instead of German bonds since she'll make a profit in dollar terms, and then she additionally would have a 50% chance that the dollar will have appreciated against the Euro.
When everyone is thinking the same, no one is thinking at all

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Mon Oct 02, 2017 3:22 pm

Valuethinker wrote:
Mon Oct 02, 2017 10:55 am
Lauretta wrote:
Mon Oct 02, 2017 10:24 am
I've seen a graph showing that the yield on 10-year US Treasuries (2.33%) is now higher than the yield on European Junk Bonds (2.32%) (unfortunately I am not savvy enough to be able to insert the graph here - showing the evolution of the respective yields).
Anyway, how can this be consistent with the idea that assets are rationally priced? European Junk Bonds are much more risky; yet they yield less! And if currency exchange rates are efficient, this cannot be due to people thinking that the Euro will appreciate versus the dollar.
Yes. It can. Currency forward rates reflect the differential in interest rates.

Within a currency, risky bonds are priced against the risk free bond in that currency (same maturity). You don't compare a US Treasury to an EUR junk bond.

Risk free bond in Eurozone is nearly at negative yield- -some of the shorter dated German bunds are trading at negative nominal yields.

If you buy the 10 year UST and a forward contract selling USD and buying EUR in 10 years time, the return will be very close to buying the German 10 year bund (they should be identical, in practice there are small variations but generally Covered Interest Parity holds between the main currencies)*.

Am I right in deducing that this implies that the market is actually predicting that the Euro will appreaciate against the Dollar (since I understand that what you are saying is that a EU investing in USD has to pay to secure today's exchange rates in the future) - please also see the post I've just written above, where I give a concrete example illustrating my reasoning.
When everyone is thinking the same, no one is thinking at all

Valuethinker
Posts: 33428
Joined: Fri May 11, 2007 11:07 am

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Valuethinker » Mon Oct 02, 2017 4:31 pm

(1+ us yield) ( fwd currency rate USD to eur) = spot rate USD to Eur (1+ bund yield)

Covered interest parity means this relationship is true in equilibrium where period of interest rate and forward currency rate iscthe same.

If it is not true then an arbitrage opportunity opens up. Free money.

All evidence indicates it is true empirically.

The currency forward rate is not a forecast of future exchange rates. It is the cost of locking in a future exchange rate now. It is a poor predictor of future exchange rates (a guess that they will not change performs better).

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Mon Oct 02, 2017 5:04 pm

Valuethinker wrote:
Mon Oct 02, 2017 4:31 pm

The currency forward rate is not a forecast of future exchange rates. It is the cost of locking in a future exchange rate now. It is a poor predictor of future exchange rates (a guess that they will not change performs better).
Then, it's probably rational for a EU investor to buy US bonds (without hedging the currency): they yield more than EU bonds and from what you say exchange rates are equally likely to go either way during the time she holds the bond. So there's a 50% chance that she'll have the same (or better) rate at the end of hte holding period, when she converts the dollars back to Euro.
When everyone is thinking the same, no one is thinking at all

VaR
Posts: 388
Joined: Sat Dec 05, 2015 11:27 pm

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by VaR » Mon Oct 02, 2017 5:08 pm

Lauretta wrote:
Mon Oct 02, 2017 3:22 pm
Am I right in deducing that this implies that the market is actually predicting that the Euro will appreaciate against the Dollar (since I understand that what you are saying is that a EU investing in USD has to pay to secure today's exchange rates in the future) - please also see the post I've just written above, where I give a concrete example illustrating my reasoning.
Yes, to the extent that you equate forward FX rates with "predicted" FX rates. Note that the two aren't actually the same.

You seem somewhat amazed at this. What real-world impact does this have for your situation or world-view? Are you going to do a Euro carry trade? http://www.investopedia.com/terms/c/cur ... ytrade.asp

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Mon Oct 02, 2017 5:22 pm

VaR wrote:
Mon Oct 02, 2017 5:08 pm
Lauretta wrote:
Mon Oct 02, 2017 3:22 pm
Am I right in deducing that this implies that the market is actually predicting that the Euro will appreaciate against the Dollar (since I understand that what you are saying is that a EU investing in USD has to pay to secure today's exchange rates in the future) - please also see the post I've just written above, where I give a concrete example illustrating my reasoning.
Yes, to the extent that you equate forward FX rates with "predicted" FX rates. Note that the two aren't actually the same.

You seem somewhat amazed at this. What real-world impact does this have for your situation or world-view? Are you going to do a Euro carry trade? http://www.investopedia.com/terms/c/cur ... ytrade.asp
Thanks for the link! I had heard about carry previously; now I understand what it's about. An implication for me is: should I buy US bonds (without hedging for the currency) since EU high quality bonds have negative yields for terms up to 6,5yrs? (now I know that this would be called a carry trade ;-) ) Buying US bonds hedged to euro would bring down the yields because of the cost of hedging, so it only makes sense for the diversification and to have better credit quality, but does not solve the problem of negative yields.
Another implication: I suggested to my father, who is elederly and doesn't like volatility, to currency hedge his (small) investment in the S&P500: was that rational for me to do?
When everyone is thinking the same, no one is thinking at all

VaR
Posts: 388
Joined: Sat Dec 05, 2015 11:27 pm

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by VaR » Mon Oct 02, 2017 5:40 pm

Lauretta wrote:
Mon Oct 02, 2017 5:22 pm
Another implication: I suggested to my father, who is elederly and doesn't like volatility, to currency hedge his (small) investment in the S&P500: was that rational for me to do?
Conventional wisdom is to not currency-hedge your foreign equity investments, assuming you aren't invested internationally more than market weight.

That said, I did read that currency-hedged international equity is becoming a "thing" in asset management. ETFs in particular.

Personally, I wouldn't bother to currency hedge my investments, but I'd consider buying a currency-hedged fund. In this case it's more about cost of the hedge than whether the hedge was worthwhile.

jalbert
Posts: 2297
Joined: Fri Apr 10, 2015 12:29 am

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by jalbert » Mon Oct 02, 2017 5:57 pm

Otherwise, if the market is agnostic as to which way exchange rates will go (and there is a 50%-50% chance that they'll go each way),
The market is not generally agnostic regarding exchange rates.

An unconventional way to look at it is that if expected inflation in the US is higher than expected inflation in the Eurozone, then the market expects the Euro to appreciate relative to the USD (currencies are negatively correlated to inflation), so USD investors get a yield premium as compensation.

The more conventional way to look at it is that investors in Euro-denominated fixed-income investments are getting a smaller inflation risk premium in their yield because Eurozone expected inflation is very low.

The differing yields and differing currency expectations precisely tradeoff in a completely efficient market so that you cannot exploit the differences in real terms.
Risk is not a guarantor of return.

AlphaLess
Posts: 46
Joined: Fri Sep 29, 2017 11:38 pm

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by AlphaLess » Mon Oct 02, 2017 8:51 pm

Valuethinker wrote:
Mon Oct 02, 2017 10:55 am
Lauretta wrote:
Mon Oct 02, 2017 10:24 am
I've seen a graph showing that the yield on 10-year US Treasuries (2.33%) is now higher than the yield on European Junk Bonds (2.32%) (unfortunately I am not savvy enough to be able to insert the graph here - showing the evolution of the respective yields).
Anyway, how can this be consistent with the idea that assets are rationally priced? European Junk Bonds are much more risky; yet they yield less! And if currency exchange rates are efficient, this cannot be due to people thinking that the Euro will appreciate versus the dollar.
Yes. It can. Currency forward rates reflect the differential in interest rates.

Within a currency, risky bonds are priced against the risk free bond in that currency (same maturity). You don't compare a US Treasury to an EUR junk bond.

Risk free bond in Eurozone is nearly at negative yield- -some of the shorter dated German bunds are trading at negative nominal yields.

If you buy the 10 year UST and a forward contract selling USD and buying EUR in 10 years time, the return will be very close to buying the German 10 year bund (they should be identical, in practice there are small variations but generally Covered Interest Parity holds between the main currencies)*.
So how do you explain this? If I find out how to insert images, I'll post the graph later. :happy
As above.

* there are differences because the yield curves have different shapes and because the market has slightly different views of the default probability of UST v. bunds. If you use 0 coupon bonds they should be essentially identical.
Valueinvestor, you are 100% correct. However, you are simply explaining the mechanics of the following no-arbitrate condition:
- FX spot,
- interest rate differential.
- FX future,

However, in my opinion, you did not explain the *FUNDAMENTAL* reasons for why there is interest rate differential.

User avatar
whodidntante
Posts: 2335
Joined: Thu Jan 21, 2016 11:11 pm

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by whodidntante » Mon Oct 02, 2017 11:05 pm

The USA is further along in the business cycle than the Eurozone.
https://www.fidelity.com/viewpoints/mar ... ember-2017
Additionally, there is monetary tightening in the USA, while the ECB is still creating money to buy bonds.

Valuethinker
Posts: 33428
Joined: Fri May 11, 2007 11:07 am

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Valuethinker » Tue Oct 03, 2017 2:44 am

AlphaLess wrote:
Mon Oct 02, 2017 8:51 pm
Valuethinker wrote:
Mon Oct 02, 2017 10:55 am
Lauretta wrote:
Mon Oct 02, 2017 10:24 am
I've seen a graph showing that the yield on 10-year US Treasuries (2.33%) is now higher than the yield on European Junk Bonds (2.32%) (unfortunately I am not savvy enough to be able to insert the graph here - showing the evolution of the respective yields).
Anyway, how can this be consistent with the idea that assets are rationally priced? European Junk Bonds are much more risky; yet they yield less! And if currency exchange rates are efficient, this cannot be due to people thinking that the Euro will appreciate versus the dollar.
Yes. It can. Currency forward rates reflect the differential in interest rates.

Within a currency, risky bonds are priced against the risk free bond in that currency (same maturity). You don't compare a US Treasury to an EUR junk bond.

Risk free bond in Eurozone is nearly at negative yield- -some of the shorter dated German bunds are trading at negative nominal yields.

If you buy the 10 year UST and a forward contract selling USD and buying EUR in 10 years time, the return will be very close to buying the German 10 year bund (they should be identical, in practice there are small variations but generally Covered Interest Parity holds between the main currencies)*.
So how do you explain this? If I find out how to insert images, I'll post the graph later. :happy
As above.

* there are differences because the yield curves have different shapes and because the market has slightly different views of the default probability of UST v. bunds. If you use 0 coupon bonds they should be essentially identical.
Valueinvestor, you are 100% correct. However, you are simply explaining the mechanics of the following no-arbitrate condition:
- FX spot,
- interest rate differential.
- FX future,

However, in my opinion, you did not explain the *FUNDAMENTAL* reasons for why there is interest rate differential.
I believe I answered the OP's question.

FUNDAMENTAL reasons? These are just stories we tell each other.

Arbitrage is real, markets are real. Why? That's stories.

So I'll tell you a story:

- the ECB is still buying bonds through QE, that drives down yields on Eurozone bonds
- the Fed is looking to tighten, markets are anticipating that in the yield curve

The possible surprises:

- given the Eurozone's strong economic growth, QE stops sooner than we think
- given a likely US Fed Chairman in the future who is more populist, US does not tighten as fast as we think-- tends not to tighten in election years (i.e. every other year)

You will note I AVOIDED saying anything about inflation expectations, etc.

Valuethinker
Posts: 33428
Joined: Fri May 11, 2007 11:07 am

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Valuethinker » Tue Oct 03, 2017 2:51 am

Lauretta wrote:
Mon Oct 02, 2017 5:04 pm
Valuethinker wrote:
Mon Oct 02, 2017 4:31 pm

The currency forward rate is not a forecast of future exchange rates. It is the cost of locking in a future exchange rate now. It is a poor predictor of future exchange rates (a guess that they will not change performs better).
Then, it's probably rational for a EU investor to buy US bonds (without hedging the currency): they yield more than EU bonds and from what you say exchange rates are equally likely to go either way during the time she holds the bond.
Reread what I said. A naive forecast outperforms the forward curve. Doesn't mean either is a good forecast or that probabilities are equally likely.
So there's a 50% chance that she'll have the same (or better) rate at the end of hte holding period, when she converts the dollars back to Euro.
I didn't say that.

You like taking an idea and pushing it to some theoretical limit? Somebody says something and you try to push the meaning out right onto the frontier? Maybe that's what an engineer does? That's not really how investing works, for practical purposes. Too many uncertainties. You have to drive down the centre of the lane not run at 120 mph on the mountain road**.

What you are describing is the "carry trade" or "Uncovered Interest Parity". Empirically UIP holds *but* there is big event risk*. Yes you can invest in Icelandic government bonds (2007) at 15% and make money against investing in Japanese GBs at 0.5% -- currency position uncovered.

However every so often that strategy blows up-- for example Iceland defaulted. The old description of such strategies as "picking up nickels in front of bulldozers".

In picking up 1-2% p.a., say on US govt bonds, you risk taking a much bigger loss. However it is a strategy that will probably work for you, most years. There is likely to be a correlation between the strategy's failure and a bear market in equities which will hurt your portfolio diversification.


* a lot of currency overlay funds were launched exploiting UIP. From what I read, the profitability of that strategy then declined-- as you'd expect with efficient markets which respond to new information (research that UIP worked). The noisiness of the UIP data was always quite large.

** you have read When Genius Failed? about Long Term Capital Management? By Roger Lowenstein? There's also a book by Dunbar (which is better on the actual strategies, weaker on the narrative). If you have not read this, and you are interested in bonds, currencies, speculation and what happens when you are in a "crowded trade" that goes wrong, then you should.

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Tue Oct 03, 2017 3:32 am

Valuethinker wrote:
Tue Oct 03, 2017 2:51 am
Lauretta wrote:
Mon Oct 02, 2017 5:04 pm
Valuethinker wrote:
Mon Oct 02, 2017 4:31 pm

The currency forward rate is not a forecast of future exchange rates. It is the cost of locking in a future exchange rate now. It is a poor predictor of future exchange rates (a guess that they will not change performs better).
Then, it's probably rational for a EU investor to buy US bonds (without hedging the currency): they yield more than EU bonds and from what you say exchange rates are equally likely to go either way during the time she holds the bond.
Reread what I said. A naive forecast outperforms the forward curve. Doesn't mean either is a good forecast or that probabilities are equally likely.
So there's a 50% chance that she'll have the same (or better) rate at the end of hte holding period, when she converts the dollars back to Euro.
I didn't say that.

You like taking an idea and pushing it to some theoretical limit? Somebody says something and you try to push the meaning out right onto the frontier? Maybe that's what an engineer does? That's not really how investing works, for practical purposes. Too many uncertainties. You have to drive down the centre of the lane not run at 120 mph on the mountain road**.

What you are describing is the "carry trade" or "Uncovered Interest Parity". Empirically UIP holds *but* there is big event risk*. Yes you can invest in Icelandic government bonds (2007) at 15% and make money against investing in Japanese GBs at 0.5% -- currency position uncovered.

However every so often that strategy blows up-- for example Iceland defaulted. The old description of such strategies as "picking up nickels in front of bulldozers".

In picking up 1-2% p.a., say on US govt bonds, you risk taking a much bigger loss. However it is a strategy that will probably work for you, most years. There is likely to be a correlation between the strategy's failure and a bear market in equities which will hurt your portfolio diversification.


* a lot of currency overlay funds were launched exploiting UIP. From what I read, the profitability of that strategy then declined-- as you'd expect with efficient markets which respond to new information (research that UIP worked). The noisiness of the UIP data was always quite large.

** you have read When Genius Failed? about Long Term Capital Management? By Roger Lowenstein? There's also a book by Dunbar (which is better on the actual strategies, weaker on the narrative). If you have not read this, and you are interested in bonds, currencies, speculation and what happens when you are in a "crowded trade" that goes wrong, then you should.
thanks for the answer. Now I remember that a colleague from Argentina some time ago told me that people were doing something of this sort in his country, and a money manager last month told me about Brazil being good for the carry trade - I didn't really understand then what this is about but I do now. I agree that in these cases it may be like "picking up nickels in front of bulldozers" - because of default risk and probably risk of fast devaluation of these currencies (dont know enough about the latter). But I was thinking in terms of investing in USD Treasuries, which are actually safer than EU ones as far as default risk is concerned. The only risk is the devaluation of USD vs Euro.
When everyone is thinking the same, no one is thinking at all

alex_686
Posts: 2620
Joined: Mon Feb 09, 2015 2:39 pm

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by alex_686 » Tue Oct 03, 2017 8:59 am

VaR wrote:
Mon Oct 02, 2017 5:08 pm
Lauretta wrote:
Mon Oct 02, 2017 3:22 pm
Am I right in deducing that this implies that the market is actually predicting that the Euro will appreaciate against the Dollar (since I understand that what you are saying is that a EU investing in USD has to pay to secure today's exchange rates in the future) - please also see the post I've just written above, where I give a concrete example illustrating my reasoning.
Yes, to the extent that you equate forward FX rates with "predicted" FX rates. Note that the two aren't actually the same.
Let me throw on another level, inflation adjusted currency movements. The US is expected to have higher inflation than the EU region. I believe that I have mentioned before that inflation adjusted currency movements have a volatility of about 1%. For stocks that return on average 10% that is nothing, for bonds that have real returns of under 1% that is huge.

To tie back to your Warren Buffet thread, bonds are not offering the best return for the level of risk taken based on historical trends. Sigh. You have a couple of options, none of them good.

You could go down Buffet's road of a very high percentage of equities. However he has a high ability and willingness to take risk, and his time horizon for required cash flow is over 20 years (He says 100 years but I think he is exaggerating for effect.)

I would not focus too much on TIPS. The theory behind them is powerful, the actual practice is not. They tend to be expensive for the protection offered. Maybe if inflation cracked 5% I would change my tune.

I would think long bonds are the way to go. Maybe even individual ones.

One last question - I think the answer to your question is the differential between inflation rates between the EU and US. Nominal rates would be different but real rates should be closer. However, are junkie are EU junk bonds? US junk bonds tend to be very junkie. We love our debt and leverage. I think the EU might be a bit more conservative.

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Tue Oct 03, 2017 9:33 am

alex_686 wrote:
Tue Oct 03, 2017 8:59 am


I would think long bonds are the way to go. Maybe even individual ones.

One last question - I think the answer to your question is the differential between inflation rates between the EU and US. Nominal rates would be different but real rates should be closer. However, are junkie are EU junk bonds? US junk bonds tend to be very junkie. We love our debt and leverage. I think the EU might be a bit more conservative.
Thanks alex_686 Yes I had come to the idea of long term bonds too, because 1. they would fulfill the role of being a hedge in case stocks fall (if they are good quality (as in the arguments behind risk parity portfolios) - I guess for me they would have to be German bunds(?) and 2. I would avoid the short term ones with negative yields! Seems to me one has to be a masochist to buy them!... For example the yield for German bonds for terms up to 6,5 yrs are currently negative...
Question is: how long is long term? (i.e. roughly, what term did you have in mind?)
Concerning the question on EU junk bonds, I really have no idea... a lot of bonds from banks seem to be risky though, but I'm just saying this based on what I've heard.
When everyone is thinking the same, no one is thinking at all

Valuethinker
Posts: 33428
Joined: Fri May 11, 2007 11:07 am

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Valuethinker » Tue Oct 03, 2017 11:32 am

Lauretta wrote:
Tue Oct 03, 2017 3:32 am
thanks for the answer. Now I remember that a colleague from Argentina some time ago told me that people were doing something of this sort in his country, and a money manager last month told me about Brazil being good for the carry trade - I didn't really understand then what this is about but I do now. I agree that in these cases it may be like "picking up nickels in front of bulldozers" - because of default risk and probably risk of fast devaluation of these currencies (dont know enough about the latter). But I was thinking in terms of investing in USD Treasuries, which are actually safer than EU ones as far as default risk is concerned. The only risk is the devaluation of USD vs Euro.
Default risk is not part of a carry trade strategy AFAIK - one assumes the same default risk (in practice, that may be a dangerous assumption). Note your comparison of HY bonds v. US Treasuries muddies thinking about this issue.

The big currencies don't move a lot against each other - most of the time. Once, during the Asia Crisis, the Yen moved up against the USD 14% *in a single day*. Once, the GBP devalued against the DM by 20% in a single day ("Black Wednesday").

You see the problem? Big moves on the currency can be correlated with other big macroeconomic events-- same underlying causes.

And there's systemic risk. LTCM is your classic case of a "crowded trade"-- the position size on their "nearly risk free" arbitrages can become so large the positions are illiquid (the Crash of 1987 was also illiquidity risk - Portfolio Insurance was dependent on an assumption of continuous trading to work).

Nowadays you are far more likely to get many speculators pursuing the same strategy than one big hedge fund. But the risk remains.

Brazil has been a fantastic investment. Shortly this should lead some investors to make really big bets on it, and lost a lot of money ;-).
The only risk is the devaluation of USD vs Euro.
So US risks:

- 1. US goes into technical default due to budget politics, and US bond market sells off, badly. 2. Or simply higher US inflation. 3. Or tighter Fed policy in anticipation of same

This raises a general point that you haven't considered, that US yield gap over EUR bonds could go even higher

As I say, probably you make money on this trade. But that could be the worst thing that ever happens to you, but it will convince you to risk more on another risky strategy (or the same one) ;-).

You have to trade off your upside to your downside. Consider how we got here:

- you didn't (in an earlier thread) want to own bonds (much or perhaps even any?)

- you were (sort of) persuaded to own bonds for the usual reasons we suggest it here

- now you don't think bonds yield enough. So you want to increase the returns on your bond portfolio by using new "low risk" strategies. You want to identify an investing anomaly that all of those trillions of dollars and highly paid researchers out there have failed so far to identify

I would suggest you think about the chain of emotional logic that has led to this point.

The reality is you want higher returns than currency and default risk free bonds can give you. So why not just own more equities? Why own bonds at all?

alex_686
Posts: 2620
Joined: Mon Feb 09, 2015 2:39 pm

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by alex_686 » Tue Oct 03, 2017 12:10 pm

Lauretta wrote:
Tue Oct 03, 2017 9:33 am
Thanks alex_686 Yes I had come to the idea of long term bonds too, because 1. they would fulfill the role of being a hedge in case stocks fall (if they are good quality (as in the arguments behind risk parity portfolios) - I guess for me they would have to be German bunds(?) and 2. I would avoid the short term ones with negative yields! Seems to me one has to be a masochist to buy them!... For example the yield for German bonds for terms up to 6,5 yrs are currently negative...
Question is: how long is long term? (i.e. roughly, what term did you have in mind?)
Concerning the question on EU junk bonds, I really have no idea... a lot of bonds from banks seem to be risky though, but I'm just saying this based on what I've heard.
Multiple rambling overlapping points.

Technically you are not hedging and this is an important point. Hedges are things that move in the opposite direction of the market. You are diversifying by picking up bonds. You are picking up exposures to duration (real interest rates and inflation) and credit (None for government, lots for junk).

Why are you doing this? This is going to tie back to your Warren Buffet post. I would suggest risk.

One reason is a more efficient portfolio. Different assets should have exposure to different factors (duration, inflation, equity, etc.). Assets with low correlations to each other offer a better risk / return profile, a better Sharpe ratio. More return for unit of risk taken. Warren Buffet - a man of infinite ability and time to take risks - a man who is a market timer - is eschewing bonds. Not enough return per unit of risk.

One reason is to ensure that you hit your goals. It might help to divide your goals into required, desired, and aspirational. Take a look at insurance companies and pension funds. They must meet their promised cash flow needs. They can't quite making payments for a few years for the market to recover. Their required goals are so important that they will accept negative real returns. So large slugs of very safe government bonds.

How you define risk and the level of risk is probably your next step. This is echoing what Valuethinker is saying.

Personally I invest in intermediate corporate bonds. Sweet spot in terms risk adjusted returns for myself. I have a reasonable amount of preferred stock, junk bonds, and EM debt. Once again, I am aiming for a efficient portfolio, not one that minimize downside risks. This is based on my personal level of risks and the risks I am willing to take. You will probably get a different answer.

One last point, my knowledge of the US bond market is strong but of the EU market is weak. From a theoretical standpoint I would think a decent slug of your AA should be in bonds and that currency hedge bond fund would be your best choice. I don't know what practical choices you have.

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Tue Oct 03, 2017 12:23 pm

Valuethinker wrote:
Tue Oct 03, 2017 11:32 am


You have to trade off your upside to your downside. Consider how we got here:

- you didn't (in an earlier thread) want to own bonds (much or perhaps even any?)

- you were (sort of) persuaded to own bonds for the usual reasons we suggest it here

- now you don't think bonds yield enough. So you want to increase the returns on your bond portfolio by using new "low risk" strategies. You want to identify an investing anomaly that all of those trillions of dollars and highly paid researchers out there have failed so far to identify

I would suggest you think about the chain of emotional logic that has led to this point.

The reality is you want higher returns than currency and default risk free bonds can give you. So why not just own more equities? Why own bonds at all?
ok, I'll try to analyse the emotions that got me here (this post however started more as a theoretical question on something that puzzled me - that's why I posted it in this General section).
With due respect I must say once again that the problem I have with bonds in the EU is not that they don't yield enough, it is that short-mid term bonds have negative yields. I am new to finance and perhaps I am simple minded, but I am totally unable to understand why anyone would want to lend money to EU governments - and pay them for doing this. I have made this remark several times in this Forum, whenever people were adviced to buy an EU all Maturity ETF, and have not yet received an answer giving a good reason why it make sense to own an ETF containing short bonds that by definition will make you lose money.
I don't have any problem with cash at all - it might lose to inflation but at least it does not have a negative nominal yield.
It's not so much that 'I was sort of persuaded to own bonds', it's that I wanted to inquire further whether I had not made some mistake in my reasoning, because people on this Forum are smart and know a lot more than me - and they own bonds. However, neither in the US nor in the UK yields are negative, so I think this explains why other investors don't have my misgivings.
When everyone is thinking the same, no one is thinking at all

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Tue Oct 03, 2017 12:42 pm

alex_686 wrote:
Tue Oct 03, 2017 12:10 pm
[

Technically you are not hedging and this is an important point. Hedges are things that move in the opposite direction of the market.
I understood that high quality bonds could be considered as a hedge to stocks, as they tend to appreciate when stocks fall, particularly the long term ones, so that you have an inversion of the yield curve. I thought that this was the reasoning behind e.g; the risk parity strategy.
When everyone is thinking the same, no one is thinking at all

lack_ey
Posts: 5822
Joined: Wed Nov 19, 2014 11:55 pm

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by lack_ey » Tue Oct 03, 2017 12:48 pm

Lauretta wrote:
Tue Oct 03, 2017 12:23 pm
ok, I'll try to analyse the emotions that got me here (this post however started more as a theoretical question on something that puzzled me - that's why I posted it in this General section).
With due respect I must say once again that the problem I have with bonds in the EU is not that they don't yield enough, it is that short-mid term bonds have negative yields. I am new to finance and perhaps I am simple minded, but I am totally unable to understand why anyone would want to lend money to EU governments - and pay them for doing this. I have made this remark several times in this Forum, whenever people were adviced to buy an EU all Maturity ETF, and have not yet received an answer giving a good reason why it make sense to own an ETF containing short bonds that by definition will make you lose money.
I don't have any problem with cash at all - it might lose to inflation but at least it does not have a negative nominal yield.
It's not so much that 'I was sort of persuaded to own bonds', it's that I wanted to inquire further whether I had not made some mistake in my reasoning, because people on this Forum are smart and know a lot more than me - and they own bonds. However, neither in the US nor in the UK yields are negative, so I think this explains why other investors don't have my misgivings.
As a retail investor with other options, you don't want to own negative-yielding bonds. If you want to optimize, you don't want bond funds that include negative-yielding bonds. At least as much as you can help it, and considering what the alternatives are. That includes broad maturity funds. Those advising otherwise are either confused, prizing simplicity over returns, or think that the bonds are safer than bank accounts and other alternatives and think the negative yields justified for safety.

There are other market participants like large institutions who can't really use savings accounts or other retail products, other governments buying bonds to affect monetary policy, short-term traders, etc. that have use for negative-yielding bonds.


In fact, there's similar logic that could be applied to US investors for a good number of years. For a while, the Fed funds rate was about 0-0.25%, and short-term Treasuries yielded less than 0.1%, while simultaneously it was readily possible to put money in FDIC-insured savings accounts with rates around 1% with no special conditions or fees. Even though the Treasuries were positively yielding, they were less than alternatives that were no more risky, and as such it didn't really make sense for people to own broad maturity bond funds including all these bonds that were worse than retail products (online savings accounts). Many people went with the lower-returning bonds because this meant staying the course and sticking with the low-cost, broadly diversified fund out of principle, or they thought the hassle with micromanaging and switching wouldn't be worth it (especially for small account sizes).

alex_686
Posts: 2620
Joined: Mon Feb 09, 2015 2:39 pm

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by alex_686 » Tue Oct 03, 2017 1:07 pm

Lauretta wrote:
Tue Oct 03, 2017 12:42 pm
I understood that high quality bonds could be considered as a hedge to stocks, as they tend to appreciate when stocks fall, particularly the long term ones, so that you have an inversion of the yield curve. I thought that this was the reasoning behind e.g; the risk parity strategy.
In the US the correlation between stocks and bonds is between .4 to .6. I can't imagine that it is that different in the EU. Hedges would be closer to -1. And don't be too frustrated or hard on yourself. Professionals are struggling in this environment.

Life in the US is not that great either. After expected inflation our government bonds are yielding something close to zero. If you noticed I am going with intermediate corporate bonds. Medium duration risk, some credit risk. If I am going to beat inflation it will be because of the credit risk I am taking, not because of the duration risk.

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Tue Oct 03, 2017 1:34 pm

lack_ey wrote:
Tue Oct 03, 2017 12:48 pm
Lauretta wrote:
Tue Oct 03, 2017 12:23 pm
ok, I'll try to analyse the emotions that got me here (this post however started more as a theoretical question on something that puzzled me - that's why I posted it in this General section).
With due respect I must say once again that the problem I have with bonds in the EU is not that they don't yield enough, it is that short-mid term bonds have negative yields. I am new to finance and perhaps I am simple minded, but I am totally unable to understand why anyone would want to lend money to EU governments - and pay them for doing this. I have made this remark several times in this Forum, whenever people were adviced to buy an EU all Maturity ETF, and have not yet received an answer giving a good reason why it make sense to own an ETF containing short bonds that by definition will make you lose money.
I don't have any problem with cash at all - it might lose to inflation but at least it does not have a negative nominal yield.
It's not so much that 'I was sort of persuaded to own bonds', it's that I wanted to inquire further whether I had not made some mistake in my reasoning, because people on this Forum are smart and know a lot more than me - and they own bonds. However, neither in the US nor in the UK yields are negative, so I think this explains why other investors don't have my misgivings.
As a retail investor with other options, you don't want to own negative-yielding bonds. If you want to optimize, you don't want bond funds that include negative-yielding bonds. At least as much as you can help it, and considering what the alternatives are. That includes broad maturity funds. Those advising otherwise are either confused, prizing simplicity over returns, or think that the bonds are safer than bank accounts and other alternatives and think the negative yields justified for safety.

There are other market participants like large institutions who can't really use savings accounts or other retail products, other governments buying bonds to affect monetary policy, short-term traders, etc. that have use for negative-yielding bonds.


In fact, there's similar logic that could be applied to US investors for a good number of years. For a while, the Fed funds rate was about 0-0.25%, and short-term Treasuries yielded less than 0.1%, while simultaneously it was readily possible to put money in FDIC-insured savings accounts with rates around 1% with no special conditions or fees. Even though the Treasuries were positively yielding, they were less than alternatives that were no more risky, and as such it didn't really make sense for people to own broad maturity bond funds including all these bonds that were worse than retail products (online savings accounts). Many people went with the lower-returning bonds because this meant staying the course and sticking with the low-cost, broadly diversified fund out of principle, or they thought the hassle with micromanaging and switching wouldn't be worth it (especially for small account sizes).
Wow, thanks for this explanation! 100% of what you write makes sense to me! :happy thanks again.
When everyone is thinking the same, no one is thinking at all

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Tue Oct 03, 2017 1:45 pm

alex_686 wrote:
Tue Oct 03, 2017 1:07 pm


Life in the US is not that great either. After expected inflation our government bonds are yielding something close to zero.
Ah ok, I thought that it was a lot easier for you over there... Though yes, lik eyou say one has to consider inflation (and taxes). I saw a Tweet by Jim O'Shaughnessy the other day, in which he wrote: 'They say there are no guarantees in investing. I think anyone buying a 10-yr Tbond almost guaranteed to lose money'...
When everyone is thinking the same, no one is thinking at all

Valuethinker
Posts: 33428
Joined: Fri May 11, 2007 11:07 am

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Valuethinker » Wed Oct 04, 2017 11:53 am

Lauretta wrote:
Tue Oct 03, 2017 12:23 pm
Valuethinker wrote:
Tue Oct 03, 2017 11:32 am


You have to trade off your upside to your downside. Consider how we got here:

- you didn't (in an earlier thread) want to own bonds (much or perhaps even any?)

- you were (sort of) persuaded to own bonds for the usual reasons we suggest it here

- now you don't think bonds yield enough. So you want to increase the returns on your bond portfolio by using new "low risk" strategies. You want to identify an investing anomaly that all of those trillions of dollars and highly paid researchers out there have failed so far to identify

I would suggest you think about the chain of emotional logic that has led to this point.

The reality is you want higher returns than currency and default risk free bonds can give you. So why not just own more equities? Why own bonds at all?
ok, I'll try to analyse the emotions that got me here (this post however started more as a theoretical question on something that puzzled me - that's why I posted it in this General section).
With due respect I must say once again that the problem I have with bonds in the EU is not that they don't yield enough, it is that short-mid term bonds have negative yields. I am new to finance and perhaps I am simple minded, but I am totally unable to understand why anyone would want to lend money to EU governments - and pay them for doing this. I have made this remark several times in this Forum, whenever people were adviced to buy an EU all Maturity ETF, and have not yet received an answer giving a good reason why it make sense to own an ETF containing short bonds that by definition will make you lose money.
I don't have any problem with cash at all - it might lose to inflation but at least it does not have a negative nominal yield.
It's not so much that 'I was sort of persuaded to own bonds', it's that I wanted to inquire further whether I had not made some mistake in my reasoning, because people on this Forum are smart and know a lot more than me - and they own bonds. However, neither in the US nor in the UK yields are negative, so I think this explains why other investors don't have my misgivings.
1. if you can find an EU deposit account or term deposit that pays more than 0% then you can exploit an arbitrage opportunity available to small investors but not large ones-- investment in government guaranteed bank accounts (up to 100k Euros).

There is then the risk that the country fails to secure the bank deposit guarantee. That was in prospect with Cyprus, but then the IMF & the Eurozone backed off that and did make good on the 100k Euro, at the cost of 60% dilution of deposits above 100k EUR. But remember they started with a proposal to haircut everyone by 10%. Did they mean that? Was it a trial balloon? I do not know.

Assuming you are in Italy:

- risk of Italy having a crash like that is small
- however yield on Italian government bonds is not the same as German, so therefore there is a risk
- in such a scenario there is no Eurozone banking bailout fund - it's up to the government
- Italy's banking system is now probably the weakest in the Eurozone (one of, anyways)
- Italy's political system and the rise of the Northern League (name?) is now a significant source of Eurozone instability (there is the argument that Italy is always in a political crisis, and that's true, but I do think the threats now (see Brexit, NF in France etc.) are qualitatively different)
- Italy would need to default/ reschedule its debt AND/ OR exit the Euro (probably both; a period of exchange controls to prevent money flowing out of the country -- as Iceland did. Deposit holders with new Lira, even if they have their deposits guaranteed, will probably get a very large fall in buying power- -50% devaluation at first?). This is all bog standard IMF bailout stuff which we inflict on Emerging Market countries all the time. The new wrinkle is that developed countries can endure this sort of thing).

2. remember a negative nominal yield is not a negative real yield. If there is price deflation you will still have a positive real return.

3. it is a fair argument that QE by the ECB has driven government bond yields to very low level

4. if you buy US Treasury Bonds you have currency risk. If you buy a global bond fund hedged back into Euros you diversify away from credit risk (you take on Japan though which is the most indebted country of the G8-- 20% of a global IG govt bond fund) but your yield will still be close to zero (nominal) -- the negative yields on the strong Eurozone governments offset by the positive yields on the 2 largest markets-- Italy and Spain.

Therefore:

- you could buy a Eurozone corporate bond fund. Should pay around 1% (investment grade)

- you could take currency risk e.g. US IG fund. Should pay around 2.0-3.0%. Currency will either work for you, or against you-- there's no way of predicting it

- you should investigate ways of keeping savings (in less than 100k EUR amts per institution) in financial institutions (small depositor interest rate arbitrage, above)

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Sat Oct 14, 2017 6:52 am

Valuethinker wrote:
Mon Oct 02, 2017 4:31 pm


The currency forward rate is not a forecast of future exchange rates. It is the cost of locking in a future exchange rate now. It is a poor predictor of future exchange rates (a guess that they will not change performs better).
I've just come across this article, in which the things that puzzled me are indeed presented as evidence that markets aren't efficient: see e.g.
Now, (1) is basically efficient markets theory, which we know is wrong in detail – there are lots of anomalies. In international finance, for example, there is the well-known uncovered interest parity puzzle: differences in national interest rates should be unbiased predictors of future changes in exchange rates, but in fact turn out to have no predictive power at all.
see https://krugman.blogs.nytimes.com/2017/ ... d=tw-share
When everyone is thinking the same, no one is thinking at all

Valuethinker
Posts: 33428
Joined: Fri May 11, 2007 11:07 am

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Valuethinker » Sat Oct 14, 2017 11:35 am

Lauretta wrote:
Sat Oct 14, 2017 6:52 am
Valuethinker wrote:
Mon Oct 02, 2017 4:31 pm


The currency forward rate is not a forecast of future exchange rates. It is the cost of locking in a future exchange rate now. It is a poor predictor of future exchange rates (a guess that they will not change performs better).
I've just come across this article, in which the things that puzzled me are indeed presented as evidence that markets aren't efficient: see e.g.
Now, (1) is basically efficient markets theory, which we know is wrong in detail – there are lots of anomalies. In international finance, for example, there is the well-known uncovered interest parity puzzle: differences in national interest rates should be unbiased predictors of future changes in exchange rates, but in fact turn out to have no predictive power at all.
see https://krugman.blogs.nytimes.com/2017/ ... d=tw-share
Covered Interest Parity - currency forward is bought at the same time as fixed income investment is made. HOLDs pretty much all the time, with small anomalies reflected market maker spreads, time of day of quote etc. Arbitrageurs find it easy to quickly close the anomalies.

Uncovered Interest Parity - no currency forward or future is bought. noisy data. Does not seem to hold. If you read Krugman's post this week about the Nobel Prize and behavioural economics, there's some discussion as to why economic "laws" don't hold (and which do).

Re CIP the forward exchange rate is not a particularly good predictor of future exchange rates. Study I saw (Canada, 1980s) was a naive forecast (doesn't change) does about as well.

There are a number of online economics courses you can take, or just watch the lectures (Coursera). This might help you in your quest for knowledge-- also see introductory finance and financial markets courses.

User avatar
Lauretta
Posts: 236
Joined: Wed Jul 05, 2017 6:27 am
Location: Italy

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Lauretta » Sun Oct 15, 2017 9:04 am

Valuethinker wrote:
Sat Oct 14, 2017 11:35 am
Lauretta wrote:
Sat Oct 14, 2017 6:52 am
Valuethinker wrote:
Mon Oct 02, 2017 4:31 pm


The currency forward rate is not a forecast of future exchange rates. It is the cost of locking in a future exchange rate now. It is a poor predictor of future exchange rates (a guess that they will not change performs better).
I've just come across this article, in which the things that puzzled me are indeed presented as evidence that markets aren't efficient: see e.g.
Now, (1) is basically efficient markets theory, which we know is wrong in detail – there are lots of anomalies. In international finance, for example, there is the well-known uncovered interest parity puzzle: differences in national interest rates should be unbiased predictors of future changes in exchange rates, but in fact turn out to have no predictive power at all.
see https://krugman.blogs.nytimes.com/2017/ ... d=tw-share
Covered Interest Parity - currency forward is bought at the same time as fixed income investment is made. HOLDs pretty much all the time, with small anomalies reflected market maker spreads, time of day of quote etc. Arbitrageurs find it easy to quickly close the anomalies.

Uncovered Interest Parity - no currency forward or future is bought. noisy data. Does not seem to hold. If you read Krugman's post this week about the Nobel Prize and behavioural economics, there's some discussion as to why economic "laws" don't hold (and which do).

Re CIP the forward exchange rate is not a particularly good predictor of future exchange rates. Study I saw (Canada, 1980s) was a naive forecast (doesn't change) does about as well.

There are a number of online economics courses you can take, or just watch the lectures (Coursera). This might help you in your quest for knowledge-- also see introductory finance and financial markets courses.
Thanks. Well from a thesis I found the following on uncovered interest parity: ' This theory stipulates that the currencies of the countries where risk-free interest rates are high should depreciate relative to the currencies associated with low interest rates. However, empirical findings of last three decades report the opposite effect, and none of the proposed theoretical models have been yet able to explain why'. It seems to me that this corresponds to the example we were discussing in theis thread on dollars and euros. If I can get more yield on dollars, and it's not likely that they will depreciate relative to euros - rather, the opposite seems more likely based on history - then there's something strange.
When everyone is thinking the same, no one is thinking at all

Valuethinker
Posts: 33428
Joined: Fri May 11, 2007 11:07 am

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by Valuethinker » Mon Oct 16, 2017 8:04 am

Lauretta wrote:
Sun Oct 15, 2017 9:04 am
Valuethinker wrote:
Sat Oct 14, 2017 11:35 am
Lauretta wrote:
Sat Oct 14, 2017 6:52 am
Valuethinker wrote:
Mon Oct 02, 2017 4:31 pm


The currency forward rate is not a forecast of future exchange rates. It is the cost of locking in a future exchange rate now. It is a poor predictor of future exchange rates (a guess that they will not change performs better).
I've just come across this article, in which the things that puzzled me are indeed presented as evidence that markets aren't efficient: see e.g.
Now, (1) is basically efficient markets theory, which we know is wrong in detail – there are lots of anomalies. In international finance, for example, there is the well-known uncovered interest parity puzzle: differences in national interest rates should be unbiased predictors of future changes in exchange rates, but in fact turn out to have no predictive power at all.
see https://krugman.blogs.nytimes.com/2017/ ... d=tw-share
Covered Interest Parity - currency forward is bought at the same time as fixed income investment is made. HOLDs pretty much all the time, with small anomalies reflected market maker spreads, time of day of quote etc. Arbitrageurs find it easy to quickly close the anomalies.

Uncovered Interest Parity - no currency forward or future is bought. noisy data. Does not seem to hold. If you read Krugman's post this week about the Nobel Prize and behavioural economics, there's some discussion as to why economic "laws" don't hold (and which do).

Re CIP the forward exchange rate is not a particularly good predictor of future exchange rates. Study I saw (Canada, 1980s) was a naive forecast (doesn't change) does about as well.

There are a number of online economics courses you can take, or just watch the lectures (Coursera). This might help you in your quest for knowledge-- also see introductory finance and financial markets courses.
Thanks. Well from a thesis I found the following on uncovered interest parity: ' This theory stipulates that the currencies of the countries where risk-free interest rates are high should depreciate relative to the currencies associated with low interest rates. However, empirical findings of last three decades report the opposite effect, and none of the proposed theoretical models have been yet able to explain why'. It seems to me that this corresponds to the example we were discussing in theis thread on dollars and euros. If I can get more yield on dollars, and it's not likely that they will depreciate relative to euros - rather, the opposite seems more likely based on history - then there's something strange.
1. UIP doesn't hold but beware noisy data

2. mid 2000s a lot of currency overlay funds were founded to exploit this anomaly. So the anomaly then became smaller and the strategy became unprofitable. I don't know if that is still the case

3. one part of this is risk. Every so often you get explosive devaluations. During the Asia Crisis, the Yen moved up 14% against the Dollar *in one day*. UK devalued against the DM on Black Wednesday, leading to a 20% immediate fall in the exchange rate. Iceland went broke and imposed exchange controls. So in other words this is not "free money" this is risky money where most of the time the risk doesn't show up-- better to be a fund manager exploiting that with Other Peoples Money than with your own aka "picking up nickels in front of bulldozers strategy".

A high interest rate (Iceland was at 15%) attracts capital inflows which drive the currency up. But those inflows can turn to outflows quite suddenly. Think the Thai crash in 1997 and the chain of disasters which followed.

In the end, UIP doesn't hold because speculative demand for currencies is far greater than commercial and personal transactions demand. And because you have Central Banks that are not profit maximizing-- they deal in currency markets for reasons other than profit.

So I would bet, as a personal investment policy, that most of the time it will work for you. Periodically it will very much not work for you.

User avatar
sgr000
Posts: 347
Joined: Wed Feb 28, 2007 12:21 pm
Location: Boston
Contact:

Re: Why do 10-yrs US Treasuries have a higher yield than European Junk Bonds?

Post by sgr000 » Tue Oct 17, 2017 9:50 am

Valuethinker wrote:
Sat Oct 14, 2017 11:35 am
Covered Interest Parity - currency forward is bought at the same time as fixed income investment is made. HOLDs pretty much all the time, with small anomalies reflected market maker spreads, time of day of quote etc. Arbitrageurs find it easy to quickly close the anomalies.

Uncovered Interest Parity - no currency forward or future is bought. noisy data. Does not seem to hold. If you read Krugman's post this week about the Nobel Prize and behavioural economics, there's some discussion as to why economic "laws" don't hold (and which do).
Here's a recent MarketWatch article discussing the apparent anomaly in Euro-junk vs Treasury yields, and how it relates to covered interest parity and currency hedging:
MarketWatch wrote:Sunny Oh, "How the ‘scariest chart in financial markets’ ignores this economic law", MarketWatch, 2017-Oct-17.
In essence, hedging away the currency risk erases the yield spread. Lots of other niceties like the difference between the spot rate and the forward rate on currencies.

Post Reply