Max Yield Backtest Results

[MachineGhost]

I've had a chance recently to implement the Max Yield strategy** in the cash allocation and backtest the 25x4 portfolio.  The strategy adds about 2% CAR.  Hmm!  The outliers compared to the vanilla PP are 1983 and 1984.

[img width=825 height=825]http://i59.tinypic.com/2h67i3p.png[/img]
Yearly Returns

[img width=825 height=825]http://i57.tinypic.com/2ry6f40.png[/img]
Drawdown

** It was first described in the book Getting Rich Outside the Dollar published in 1993.  A fictitious Max Yield buys the highest yielding major world currency every January.  I contest the selection of the Turkish Lira for years 2007 to 2013.

[Kshartle]

What constitutes major world currency MG and have you actually read the book? Do you recomend it?

[Reub]

MG, what do you think about holding, instead of cash, a fund such as the rock solid Vanguard Wellesley (VWINX), which has been recommended in another thread? This strategy has shown a total return of around 10% over the past 41 years.

http://gyroscopicinvesting.com/forum/pe ... ook-juicy/

[MachineGhost]

What constitutes major world currency MG and have you actually read the book? Do you recomend it?

That's what I've wondered over the years and I do have the book in my to do pile, but have yet to read it.  I'm hoping its objective criteria, but I kind of doubt it will meet our modern standards. 

I know you're not a MR believer, but I think we could qualify "major world currency" according to those terms since I can't imagine the post-EU currencies qualify anymore since they cannot monetize their debt, which is why I suspect the Turkish Lira was picked due to a diminished pool of choices.  But Turkey is a second world country at best.  It did not make a net positive return in the 6 years held.  Like that couldn't have been figured out ahead of time?

Off the top of my head, these would qualify today: US Dollar, Japanese Yen, British Pound, Australian Dollar, New Zealand Dollar, Norweigan Kroner, Singapore Dollar, Hong Kong Dollar, some of the other Nordic/Scandinavian currencies, not sure about the South African Rand, maybe the Malaysian Ringit and Indonesian Rupee.

I forgot to post the returns of the Max Yield itself, but its about 10% CAR with about -11% MaxDD.  Very decent enhancement for cash.  The USD was only held once in the 44 years from 1970.

[MachineGhost]

MG, what do you think about holding, instead of cash, a fund such as the rock solid Vanguard Wellesley (VWINX), which has been recommended in another thread? This strategy has shown a total return of around 10% over the past 41 years.

http://gyroscopicinvesting.com/forum/pe ... ook-juicy/

A really strange choice; its 60% bonds and 40% equity.  How does that replace the virtures of cash?  The last 30 years or so have been a raging bond bull market, so of course it will look good.

[Mark Leavy]

Off the top of my head, these would qualify today: US Dollar, Japanese Yen, British Pound, Australian Dollar, New Zealand Dollar, Norweigan Kroner, Singapore Dollar, Hong Kong Dollar, some of the other Nordic/Scandinavian currencies, not sure about the South African Rand, maybe the Malaysian Ringit and Indonesian Rupee.

First, thank you MG for introducing the idea of maximizing cash yield.  Considering yield as the inverse of price, then the annual cash swap idea falls broadly into the buy low, sell high maxim.  It seems like an idea worth exploring.

As far as Malaysian Ringit and Indonesian Rupee... I just came back from a couple of months in SE Asia and I would be very leery.  Both countries reeked of corruption.  Malaysia not quite as much as Indonesia - but both of them much much worse than Thailand - and that is a low bar...

I got the impression that both the Malaysian and the Indonesian governments are seriously sophisticated criminal organizations.  They are very good at what they do - but it is not for the benefit of their countries and certainly not for the benefit of any third party.

[Mark Leavy]

Malaysia not quite as much as Indonesia

A quick anecdote from Indonesia.  They sell a visa on arrival and sell is clearly the correct term.  It is priced at US 25 dollars regardless of your country of origin.  The immigration folks won't accept their own currency.  If you don't have USD, they will convert what you have to USD at extortionist rates.  I've never been to a country that wouldn't accept their own currency before.

[ns3]

I've had a chance recently to implement the Max Yield strategy** in the cash allocation and backtest the 25x4 portfolio.  The strategy adds about 2% CAR.  Hmm!  The outliers compared to the vanilla PP are 1983 and 1984.

[img width=825 height=825]http://i59.tinypic.com/2h67i3p.png[/img]
Yearly Returns

[img width=825 height=825]http://i57.tinypic.com/2ry6f40.png[/img]
Drawdown

** It was first described in the book Getting Rich Outside the Dollar published in 1993.  A fictitious Max Yield buys the highest yielding major world currency every January.  I contest the selection of the Turkish Lira for years 2007 to 2013.

An interesting idea but how do you implement it? Is it just a matter of buying ETF'S?

[MachineGhost]

An interesting idea but how do you implement it? Is it just a matter of buying ETF'S?

Good question.  Maybe if the yield is good.  Usually a 1-year CD is bought, so that would involve using a brokerage that has an international presence or opening a foreign bank account.

The currency for this year is the Australian Dollar at 5.25% yield.

[Kshartle]

I got the impression that both the Malaysian and the Indonesian governments are seriously sophisticated criminal organizations.  They are very good at what they do - but it is not for the benefit of their countries and certainly not for the benefit of any third party.

All governments are criminal organizations. If it was obvious to you that they are then maybe they aren't that sophisticated. 

[Kshartle]

An interesting idea but how do you implement it? Is it just a matter of buying ETF'S?

Good question.  Maybe if the yield is good.  Usually a 1-year CD is bought, so that would involve using a brokerage that has an international presence or opening a foreign bank account.

The currency for this year is the Australian Dollar at 5.25% yield.

Well....FXA has a 5.0% return YTD vs. 0.2% for SHY

[ns3]

Well, it was a tempting idea given the low current yields on cash but upon further review I think it is way too speculative for what is supposed to be the most stable part of your portfolio.

[MachineGhost]

Well, it was a tempting idea given the low current yields on cash but upon further review I think it is way too speculative for what is supposed to be the most stable part of your portfolio.

Yes, I think the concept is valid, but it needs further development to reduce the risk.  It's probably only appropriate for shallow cash, not deep.  One reason the strategy works is because returns are zero sum in forex as purchasing power nets out over time in different major world currencies, so there's no advantage of one vs the other in the long-term (10 years).

[Kshartle]

returns are zero sum in forex as purchasing power nets out over time in different major world currencies, so there's no advantage of one vs the other in the long-term (10 years).

What is your basis for this belief?

For example, FXA has returned 71% in the last 8 years (inception) and SHY only 24%. Do you think this will even out over the next two and why is that?

FXF is up 41% over that period.

The New Zealand dollar has returned more than the Aussie even and over the last 10 years as well. The Kiwi has beat the dollar by a very large margin over the last ten years, especially when you count interest. New Zealanders have gained a lot of purchasing power vs. the dollar.

[MachineGhost]

What is your basis for this belief?

It's standard economics dogma: https://en.wikipedia.org/wiki/Purchasing_power_parity

PPP gets problematic with smaller countries that have commodity or export-oriented economies such as New Zealand or Australia.  That's why Australia's interest rate is so high, they have more relative economic activity from exporting resources to China and incoming Chinese investments into the local real estate bubble, so inflation is higher.  New Zealand also exports a massive amount of food, especially to China.  With the Chinese slowdown, demand for AUD to buy their exports has decreased which is why it declined last year. 

Personally, I don't believe that just picking the highest yield in a single month out of the year is the optimal way to account for all of a currency's influences, but it seems to have worked remarkably well for over 40 years.

Maybe there is an infleection point where a currency has decreased in capital terms combined with an outdated, previously fixed rate set by lagging central banks.  In an uncontrolled market, yield would increase on a falling asset.  There could be a way to calculate the imputed yield.

[Kshartle]

What is your basis for this belief?

It's standard economics dogma: https://en.wikipedia.org/wiki/Purchasing_power_parity

I hear what your saying but I think your application of the PPP theory is not correct. PPP states that at any given time exchange rates should not move to a point where there is an arbitrage opportunity. An example would be the the Euro vs. the dollar. Since it costs 1.5 dollars to buy a Euro, a theoretical widget that has a price of $1 should have a price of .66 Euro. That way no one can gain just by buying in one currency and selling in another.

Of course this assumes essentially zero conversion and buying/selling costs so that true arbitrage could occur.

The PPP has nothing to do with exchange rates or one currency returning more (via exchange and interest rates) than another over time, including the majors.

Obviously when the Euro came out it was around 80 cents. Now it's at $1.40.  You can't tell me higher US rates have made up the difference. This is a 75% increase in a major currency over another.

This concept applies equally to little currencies. There is no reason that major currencies must equal out over time. It will depend on many factors although certainly the major ones (Euro, Swissy, USD, GBP, Yen) should be less volitile. Should be is the key...observe the Yen.

[Kshartle]

Personally, I don't believe that just picking the highest yield in a single month out of the year is the optimal way to account for all of a currency's influences, but it seems to have worked remarkably well for over 40 years.

I agree that there is no certainty but the historical evidence of picking the highest yielder is hard to ignore.

Maybe this is sort of like the dogs of the dow effect. The dow is 30 major multinationals that should in theory be profitable going concerns. If you buy the 10/30 with the highest dividend yield you are more likely to get the companies that are closer to the bottom of thier cycle, giving you a higher likelihood of outperformance of the index.

So maybe major currencies that are at the highest rates are also at the bottom of their cycles most likely.

That would provide an explanation for the historical record.

Perhaps not taking just the top currency but taking the top 2-3 would smooth it out but still show a positive return. I would guess it would give you a higher risk-adjusted return as well.

[MachineGhost]

The PPP has nothing to do with exchange rates or one currency returning more (via exchange and interest rates) than another over time, including the majors.

I literally have no idea what you're talking about!  PPP identifies whether a currency is over or undervalued and it will revert to the equilibirum PPP in the long-term.  So the short term movements against or towards to the PPP is what I believe the Max Yield strategy works upon.  So essentially any major world currency you hold will all reach the same PPP in the long-term and there won't be any net real gain.  The relative nominal values of the currency mean nothing to the Max Yield strategy, only the capital gains and yield payments.

[Kshartle]

The PPP has nothing to do with exchange rates or one currency returning more (via exchange and interest rates) than another over time, including the majors.

I literally have no idea what you're talking about!  PPP identifies whether a currency is over or undervalued and it will revert to the equilibirum PPP in the long-term.  So the short term movements against or towards to the PPP is what I believe the Max Yield strategy works upon.  So essentially any major world currency you hold will all reach the same PPP in the long-term and there won't be any net real gain.  The relative nominal values of the currency mean nothing to the Max Yield strategy, only the capital gains and yield payments.

When the Euro was $0.80 let's say a widget was $1.00 even.  Due to the PPP theory the widget should cost 1.25 Euros.

That is PPP.

Fast forward 20 years. The Euro is $1.40.  If the widget is still $1.00 even it will now cost only .71 Euro

The widget stayed $1  but went from 1.25 Euro down to .71 Euro

That's PPP. The Euro gained but the PPP between the two at any given time is equal or should be. Either the exchange rates or the widget prices should adjust over time if the PPP is not equal. This does not mean the exchange rates will change however if they should not for overwhelming fundamental reasons (greater inflation or inflation fears or deflation or productivity changes etc.)

Why do you think the Chinese must constantly intervene to maintain the nominal exchange rate? The RMB should be rising in nominal terms against the dollar but for the "pegging". If they didn't have the peg on someone who bought RMB 10-15 years ago would likely have a significant nominal gain (resulting in capital gains). The chinese would have gained PP over that time but at any given time there should be relative PPP between the USD and RMB.

It has nothing whatsoever to do with the theory that a major currency can't provide long-term gains against another due to exchange rate changes and interest rates. That a completely false theory and easily disproven with ample evidence. Heck the Yen used to be 360 to the dollar, two years ago it hit 80 right?!?!?

Does anyone else understand this?

[Kshartle]

The relative nominal values of the currency mean nothing to the Max Yield strategy, only the capital gains and yield payments.

You realize capital gains come from changes in nominal values right?

[MachineGhost]

It has nothing whatsoever to do with the theory that a major currency can't provide long-term gains against another due to exchange rate changes and interest rates. That a completely false theory and easily disproven with ample evidence. Heck the Yen used to be 360 to the dollar, two years ago it hit 80 right?!?!?

Does anyone else understand this?

But thats not what I said?  I more or less said the PPP makes long-term real gains in forex a zero-zum game, because all currencies aspire to the same PPP fair value in the long-term.  It is arbitraged away.  The short term changes in yields and capital gains are due to other factors that move the currency for/against the PPP.  I think we are merely arguing over semantics of how short-term short-term is and how long-term long-term is.

[Kshartle]

I think we are merely arguing over semantics of how short-term short-term is and how long-term long-term is.

That might be the case but I think if you're saying that you can't have long term purchasing power gains in one currency over another, I just don't understand where that is coming from.

Some currencies gain in purchasing power over time and some lose purchasing power. Certainly you can compare two currencies over any time period and have a difference in the exchange rate and the interest rate returns. PPP is only about the present time...that at any given time there shouldn't be an arbitrage opportunity between two currencies whereby you could exchange one for another and buy goods cheaper then sell them in the other currency and reap an arbitrage gain. Either the rates will move or the prices of the goods will move to eliminate this.

PPP deals not only with exchange rates but the prices of goods.

The gains of one currency vs another are not related to the price of other goods though, just the exchange rate changes over time coupled with the interest returns. For example the Yen used to be 360 to the dollar. The Swiss franc used to be $0.23 now it's what....$0.90?  The loonie when I was a kid was $0.60, now it fluctuates around $1 even.

PPP is not realted to long term changes in the relative value between two currencies it's about how much stuff they can buy at the present time and why you shouldn't be able to just do a currency change and buy more stuff now.

If any readers of the thread who think they understand this can clear up the dissagreement I would appreciate the thoughts. If I'm mistaken I'd like to understand and correct it.

[MachineGhost]

I had a chance to read the relevant sections in the book and clearly the Max Yield is a gunslinger, devil take care strategy that they don't recommend, but is used for purpose of illustration of how diversifying outside the dollar was profitable.  So if you did want to pick a currency properly, then all the usual factors apply: rule of law, productivity, interest rates, inflation, etc. as the 7-year (2005-2013) losing streak in picking the Turkish Lira evidences (I do not know if the Lira was an official pick of the authors, but I would assume so).  I also think that the strong performance is similar to the bond bull market, because yields were a lot higher back in the 70's and 80's than they are now.  The EU formation also complicated things by reducing a lot of currencies to choose from.

Under Monetary Realism, I don't think it can be justified in switching from the USD because it is the world's reserve currency.  Nothing will top it for safety, just as nothing topped the Roman Denarius.

As with the PP in the other thread, I just ran an updated backtest and the CAR is about 1.40% less than the PP.  Case closed.

[MachineGhost]

Further on this subject, there is a "January Effect" for currencies.  Whatever currency performs the best in the month of January has a high probabilty (80%) of outperforming all other currencies throughout the rest of the year.  The Max Yield strategy could perhaps be improved quite simply by using this observation.  I've little interest in exploring this further.  The yen was the best currency this past January.

[Kshartle]

Further on this subject, there is a "January Effect" for currencies.  Whatever currency performs the best in the month of January has a high probabilty (80%) of outperforming all other currencies throughout the rest of the year.  The Max Yield strategy could perhaps be improved quite simply by using this observation.  I've little interest in exploring this further.  The yen was the best currency this past January.

Is January the best predictor? Is any other month better?

My point is January is a subset of the entire year and so it factors into the final results, just like any other month.

It's like when stocks are up in January there's an 80% the year ends up. Well....the year is up 65% or so anyway, maybe 70% so if you're up in January you're 1/12 of the way there and you've got a headstart built in!

That might be the same with the currency. Maybe there's a November effect that puts it at 85% over the next 12 months.