It’s time.? A project that I’ve been working on extensively for weeks, probably to the great annoyance of the lady of the house, is finally at a point where I’m ready to share my work and begin testing the idea in a semi-public forum.? The working title for this project is the Uberman’s Portfolio, inspired by the infamous Uberman’s Sleep Schedule in that it never sleeps and because of the many late nights spent building the?gears and levers that make it all?possible.? Also, the acronym is UP, which is where I hope my equity will be when all is said and done.

Before describing the portfolio, let me first talk about currency carry trades.? Here is what Investopedia says:

A strategy in which an investor sells a certain currency?with a relatively?low interest rate and uses the funds to purchase?a different?currency yielding a higher interest rate.?A trader using this strategy attempts to capture the difference between the rates – which can often be substantial, depending on the amount of leverage the?investor chooses to use.

Here’s an example of a “yen carry trade”: let’s say a trader borrows 1,000 yen from a Japanese bank,?converts the funds into U.S. dollars and buys a bond for the equivalent amount. Let’s assume that the bond pays 4.5% and the Japanese?interest rate is set at 0%. The trader stands to make a profit of 4.5% (4.5%?- 0%), as long as the exchange rate between the countries?does not change.?Many professional traders?use this trade because the gains can become very large when leverage is taken into consideration. If the trader in our example uses a common leverage factor of 10:1, then she can stand to make a profit of 45%.

The big risk in a carry trade is the uncertainty of exchange rates.?Using the example above, if the U.S. dollar?were to?fall in value relative to the Japanese yen, then the trader would run the risk of losing money. Also, these transactions are generally done with a lot of leverage, so a small movement in exchange rates can result in?huge losses unless hedged appropriately.

While this explanation is sufficient, it’s a little removed from the actual mechanics of the everyday currency trader so let me put an example in terms of a spot forex trade.? When you buy or sell at a forex dealer, you select a currency pair (which represents an exchange rate), such as USD/JPY in the example above, and go long or short that pair.? If you go long, this means that you are buying USD and selling JPY, or to put it in terms of the other example, borrowing yen to buy dollars.? Each currency has an interest rate and you earn the rate on any currency that you?buy and?pay the rate on any currency that you are borrowing.? Remember?my previous post about the 4.7% interest?earned on Oanda US$ accounts?? Well,?that is because the cash in your trading account is technically a long position in US dollars held at 1:1 leverage, earning you the current rate on the US dollar.? Back to our example, if you are long the currency with the higher rate and short the one with the lower rate, then you will have a net positive earn or “carry”.? You will, of course, have a net negative carry if you hold the opposite position.? And, as in all trading, there is a bid/ask spread on the interest rates so the negative is always slightly?larger than the positive.

Following our example of a long position in USD/JPY, the going rate difference at Oanda is +4.48%, which is currently the #2 yielding pair available.? I know by now you are asking yourself why in the world would you accept the risk of adverse exchange rate movement (aptly marked in bold above) to get a rate that is less than the risk-free rate (RFR) of 4.7%?? Well, did you read that part about leverage?? Through the power of leverage, you can ratchet up the return as much as 50x, which in the case of USD/JPY is a 224% return.? You can’t do that with?the RFR.? ?”AH HA”, you say, “gotcha!? Leverage is dangerous and no one in their right mind would go full tilt to get that 224%.? The swings would make a sailor sick.”? Exactly right.? Unless you have a death wish or can tolerate 75%?swings in your equity?every time the Bank of Japan gives a press release, then no you wouldn’t.

So what now?? Well, there are currently 33 currency pairs offered at Oanda and 30 of them have positive carry of some degree if held in the right direction.? After removing the exotic and potentially unstable ones like USD/HUF (that’s the Hungarian Forint, if you didn’t know), there are 20, consisting of various combinations of dollars, euros, pounds, yen, francs, aussies, kiwis, kronas and loonies.? The average fully-leveraged return is around 75% per year in interest.? And yes, Virginia, you earn the RFR on your whole equity at all times, even the equity being used for margin, so add that on top.? Couldn’t you form a basket of currencies to hedge your risk and then maybe scale back that leverage a little to something more sustainable, lowering your return but still beating the RFR?? Well, yes you could and, in fact, that idea is not new or lost on the trading world.? You will find “cash & carry” trading ideas all across the internet in forums, blogs and articles.? They range from “hold a position in the highest yielding pair”?to “form a basket of the 10 highest” etc.? But most fail after the high yielding long position starts to lose money when the exchange rate heads south so fast that it wipes out all money earned from interest and then some.? Attempts at hedging are typically random or fail to adjust to changing markets.? Often times, the initial returns are because pairs with very?lucrative yields tend to go up initially as funds flow into them to take advantage of those yields.? In other words, the great equity curves for these portfolios?bear a striking resemblence to the underlying pairs.? But the party never lasts forever.? The goal should be to have a portfolio that, if it were not earning interest, would breakeven and, with interest, should follow the line of the average interest rate held, not the underlying market.

In other words, no one it seems has yet combined all the right elements into one plan:

  • Proper?hedging
  • Improved volatility measurement
  • Out-of-sample validation
  • Risk management
  • Optimal?pair selection
  • Adaptation to change
  • Reasonable expectations

And this is what I’ve been working to achieve: A complete, logical, non-discretionary?plan to trade a portfolio of positive-carry pairs hedged for minimal directional exposure using careful risk management and controlled leverage that automatically adapts to changing markets and earns enough above the RFR to make it a worthwhile endeavor.? BREATH…whew.? And I think I’ve done just that.? Will it work?? I sure as hell hope so.

Great!? So how!?? Well the total mechanics are quite involved but here is the gist:

  • Each weekend, while the markets are resting and time is more serene, check for any interest rate changes and calculate the optimally hedged portfolio using the past few months of data, the one that tries to erase as much volatility and market directional exposure as possible while still returning a worthwhile yield. (Proper hedging)
  • Once this is done, validate the result on some recent out-of-sample data.? Man, is this ever an important but hardly ever performed step. (OOS Validation)
  • If the OOS validation says that the hedge seems to be holding up, trade it over the next week.? If not, drop to cash and earn that nice RFR. (Adaptation)
  • But before you trade, run the past and some randomly generated data that resembles the past through a position-sizing model to find out what leverage you should trade given the amount of drawdown you are willing to accept.? Another oft ignored step. (Risk management)
  • Adjust your holdings to reflect the proper percentages and leverages and sit back until next weekend.
  • Roll the data window forward and repeat.

So I didn’t reinvent the wheel but I like to think I made it a little more like an all-season radial with run-flat than a wagon wheel.? The result could be a combination of some of the?most sought after?characteristics?such as low activity (i.e. low costs), limited market monitoring, unambiguous rules etc.? What can one expect?? Well, the power of compounding is amazing but a few tests have would lead me to believe that a typical Uberman’s Portfolio with a drawdown tolerance of 10% would probably equate to holding 5-7 pairs at 50:1 leverage on about 10% of equity and result in expected returns of 10% annually without compounding.? But only time and live testing will tell.? Something like this could end up being all the trading I’d ever need to do.? In fact, it’s more investing than trading.? It’s a self-made money market for the hardcore.? But it’s also probably more risk management and planning than most “conservative” investors fool themselves into thinking they’ve done, making it suitable for the long-term steady-rate type.? This is intelligent asset allocation at its purest.

Ok, ok!? So what does an Uberman’s Portfolio look like, dammit!?? Well currently, like this:

Leverage of 5:1 or 10% of equity will be used for margining positions with an account margin set at 50:1.? There is no good reason not to set the account margin at the highest level then simply use a % of equity to control the overall leverage.? Margin calls are not a good?method of?risk management.? For example, a $10,000 account would allocate $1000 to margin, which at 50:1, would control $50,000 worth of currency or 5:1 ($50,000:$10,000) overall leverage.

55.9% Long EUR/CHF (49.96% yield)

24.8% Short EUR/GBP (72.58% yield)

15% Long CHF/JPY (46.51% yield)

3.9% Long EUR/SEK (17.43% yield)

0.4% Long NZD/USD (87.50% yield)

Average yield = 54% at full 50:1 leverage. Positions are precents in US$ terms not the base currency of the pair.? (EUR = Euro, CHF = Swiss Franc, GBP = British Pound, JPY?= Yen, SEK = Swedish Krona, NZD = New Zealand Dollar)

Expected Yield = 9.63% (10% at 54% yield including RFR?+ 90% at 4.7% RFR only)

Expected Volatility = 3.65%

I’ve opened a demo account with Oanda and will begin to test the results in real time.? What I hope to learn from this experience are things like “Is updating once per week too much or too little?”, “How much cost is incurred in updating (which should hopefully be low since the?portfolio shouldn’t change that much from week to week)?”, “How close to the expected return and volatility am I getting?”, “How quickly does it adapt to changing conditions?”, etc.