Double Blind

E-Trade used to be a brokerage-only operation, until about five years ago when it decided it wanted to be a bank. It saw the writing on the wall and realized that offering stock trading services (and collecting commissions only) was a money-losing business. Most brokerages make money from other services (specifically financing short-term debt through margin loans) and less off the trading. The most profitable of businesses is making money off your customers by making markets.

E-Trade ended up taking bad mortgages and when the US real estate market collapsed in 2008, E-Trade nearly went into bankruptcy. A group named Citadel (which most local readers in the Lower Mainland will know has a connection with the Olympic Village fiasco in the City of Vancouver) gave E-Trade a loan and extracted a huge cost out of the company with respect to credit terms (expensive) and equity (even more expensive). Shareholders got absolutely hammered, but the alternative would have been bankruptcy (where they would have received zero, opposed to the 90% loss they took in the transaction).

In the process, Citadel cut a deal with E-Trade to route substantially all of E-Trade customers’ orders (on the brokerage side) through its own order routing service, thus being able to generate higher profits for its own company.

Information is the most powerful currency that brokerage firms possess – if you have a lot of customers making various trades, you can monetize the order flow.

If you are a large account and are too successful, it is likely that you will start to experience slippage on trades because automated programs from your brokerage will likely be front-running your own orders. When you get to that size where you can start moving markets, being sneaky becomes important.

I would not put any money in E-Trade: Their balance sheet is a complete disaster and it is not worth having your money in an organization that is clearly still in huge financial difficulties.

Between April 1 and June 30, the CPP plan reported a 7.1% gain on their investment portfolio. This is contrasted with the disastrous first quarter of the calendar year, where the fund lost money just with every other institutional investor on the planet.

The troubling part of the press release is the following:

In just over 10 years since the CPPIB began investing in April 1999, the CPPIB has generated $31.8 billion in investment income for the Fund reflecting an annualized rate of return of 4.9 per cent.

…

The Chief Actuary of Canada estimates that a 4.2 per cent real rate of return, or approximately 6.2 per cent on a nominal basis, over the span of the 75-year timeframe covered by his 2007 report is required to sustain the plan at the current contribution rate. “Although our four-year results are currently below this long-term rate of return, we believe the CPP Fund will generate returns in excess of this 4.2 per cent threshold over its long investment horizon,” said Mr. Denison.

The CPP has earned a 4.9% annualized nominal rate of return since 1999, but the actuary estimates a 6.2% nominal rate is required over a 75 year timeframe is required. While the 1.3% difference does not seem significant, when dealing with the volume of numbers the CPP is dealing with (about $120 billion currently; estimated to go to $235 billion in 2015 and $1.42 trillion in 2050), a 1.3% difference in performance is a huge amount of return to recover.

I don’t think they will be able to do it unless if they get very good at the international investment side of their portfolio, which is where most of the above-average capital growth will be captured. That said, a 6.2% nominal rate is a conservative target and it is reassuring that the number is so low. Most defined benefits pension plans have insanely larger assumed rates of returns (usually around 8.5% and above) which will not be realized.

In order for there to be domestic investment income, somebody has to be producing value in the economy. Everybody cannot invest and live off of investment income, because somebody (people, corporations) have to be there to create the value in the first place. As more Canadians continue to retire and exit the workforce, either corporations have to depend on less labor inputs to create value (which has been the general trend with automation), or people have to lessen their expectations from investment income. This is one of the reasons why it would be silly to exclusively concentrate the assets of the Canadian Pension Plan in Canadian assets – you will never get the gains required to keep the CPP solvent in the long term.

The creation of the CPP Investment Board and the management of the Canadian Pension Plan in general has been very well structured (something the Jean Chretien Liberal government implemented over a decade ago), and we are light years ahead compared to the dysfunctional equivalent in the United States. Our pension system at least has hard assets which are producing returns. Our current government is best to leave the system alone since it is working.

Last year, I took advantage of the BC Scrap-It program and got rid of my 1988 car and bought a 2008 vehicle. Because it was rated to reduce emissions by 4 tonnes of GHG equivalent or better over a 3 year period, it was eligible for a $2000 cheque.

I notice at the Scrap-It page that they no longer offer the $2000 incentive; instead the most you can get for your car is $1000.

The program is no longer likely worth it for those with extremely old cars (like mine previously was) – they would likely have a better chance selling it on the open market.

I know in my case that a $2000 incentive had a significant bearing on my decision to scrap my vehicle, but a $1000 incentive would have not.

I also know when people buy more fuel efficient vehicles they tend to drive more in them; my car is no exception to this, as I have driven 33,000km in the first year of my vehicle’s life. There were a few exceptional circumstances (long-distance drives) that caused this, but as it becomes cheaper to drive (via better fuel economy), one drives more. If I had driven the same distance with my old car, I probably would have spent about $1200 worth of fuel extra for the year. Considering my initial capital outlay for the vehicle, it’s odd to think of a purchase of a new vehicle to be a net positive financial investment, but this one currently is.

The attribution rule in Canadian taxation essentially says that you can’t give a bunch of money to your wife or son/daughter under 18 and have them invest it and produce gains (which would otherwise have been taxed in their names, presumably at a lower marginal tax rate). Instead, the gains would get attributed to the person with the original sum of money to begin with. There is a whole volume of information and court cases that deal with the various complexities of this issue that is well beyond the scope of this post. For better or worse, this type of material interests me.

One important exception to the attribution rule is that you can loan money at the CRA prescribed rate, and not have that income attributed back to the originating member. This is a very well known tax trick, and in fact is a very effective form of income splitting – the person giving the loan would make a small amount of interest income, while the person receiving the loan would be able to write off the interest from income as long as they invested that loan into an income-bearing asset.

What makes this particularly attractive is that the prescribed rate is 1% currently and will be this for at least September 30, 2009.

The other interesting twist to this rule is that you can give out a fixed rate loan at the prescribed rate for any duration of time – if the prescribed rate goes up, you do not have to adjust the rate of the loan.

In my estimation, people have roughly 9 months to obtain capital at dirt-cheap rates before interest rates will start to rise substantially. I wish it was easier to obtain a low-rate term loan without having to buy a property that goes with it. I’d much rather want the cash.

I posted three years ago the optimal way to exploit Aeroplan. Surprisingly, the formula still remains the same and because the availability of flights on the original formula is extremely limited, the economics work out even better: getting gas cards is the best way to monetize your Aeroplan miles for the reason that anybody with a car is bound to purchase gasoline.

Flights still incur fairly punitive costs with respect to taxes and levies incurred (which are not covered by the Aeroplan miles; effectively a “free” flight still costs significant amounts of money). It is still cheaper to book air travel this way, but availability is very difficult to come by during the times when most people need to travel.

Currently a $50 gas card is available for 6500 miles (one mile has a cash equivalent of 0.77 cents), while a $100 gas card is worth 12,500 miles (cash equivalent 0.80 cents). This is down from three years ago, where a $50 gas card was 7000 miles, and a $100 gas card was 13,500 miles. I ordinarily would not have expected the value of “aeroplan currency” to rise (approximately 7.4% over three years).

Obviously if you can save up 12,500 miles it is better to get the $100 gift card. The only way to rack up enough miles to get this is generally through international (Canada to USA or other countries) flight. Domestic flights only capture a quarter of the miles traveled, unless if you are silly enough to pay the premium price (typically $60/direction) for Tango Plus seating on Air Canada.

The economics of Aeroplan become very relevant when booking long-distance travel. For example, a trip from Vancouver to Tokyo is 4660 miles; for the round trip the total is 9230 miles. This effectively means that you can convert these miles into $73.84 worth of gasoline. Right now Air Canada graciously offers you a $15 discount each way if decline the Aeroplan miles option.

The only reason why I brought up this post was because after my cross-country trip I noticed that I reached over the 6500 mile threshold, so I dutifully cashed it out for a $50 gasoline card.

According to Alice Funke’s brilliant Pundit’s Guide to Canadian Elections, Green party leader Elizabeth May has decided to run in Saanich-Gulf Islands in the upcoming federal election (assuming there is one this fall).

My analysis that I made back in March of 2009 still remains the same: As long as Elizabeth May is leader, the Green Party of Canada will never win a seat in Parliament. I’d be surprised if she got more than 28% of the vote.

Since ING Direct only gives a yield of 1.2% in its “high interest” savings account, I have generally only been keeping enough money for fulfillment of monthly expenses; the rest of it goes into higher-yielding investments where I try to maximize my risk-reward ratio. My attention over the past year has been focused on the bond market, rather than equities. Earlier this year was a golden time to invest; now prices have subsided to a range where they should be.

I reviewed the 73 exchange-traded debentures that exist on the TSX. Most of the bond prices have increased dramatically since the last market purge in March 2009. As a result of the increased valuations, the reward for such investments is substantially poorer than the deals you could have received half a year ago. Most of the bonds are trading at values that I generally would not consider.

There were some candidates worth mentioning that I haven’t mentioned before on this site.

One is Uranium One (4.25% coupon, trading around 80 cents on the dollar, matures December 31, 2011) which is very likely to mature at par. The yield to maturity for such a note is 14.75% at present. The dramatic headline of their US$264.9 million loss is due to a write-off and has nothing to do with their cash situation (which is all that the bondholders really care about). There is very little default risk here (although of course it is nowhere near as safe as a GIC), but with respect to the actual risk taken vs. the reward it would be a good payoff for somebody wanting to invest some idle cash in a safe instrument.

Likewise, First Uranium (interesting how the Uranium debt issues are trading lower than most other mining ventures) has a 4.25% coupon, June 2012 maturity issue, trading at around 74 cents on the dollar. You get slightly better compensation (16.1%) but this is not worth the extra risk entailed in the company compared to Uranium One. If the bond traded a few pennies cheaper, it would be worth considering over Uranium One. Both firms have fairly strong balance sheets, and retain enough market value in their equity such that they could dilute to pay the debt.

The most speculative of debt plays is Newport Partners Income Fund, which has suspended interest payments on its convertible debentures, and is in the process of renegotiating its senior bank debt. The company’s balance sheet is a mess (too much debt) and it is entirely possible that they will recapitalize. The debt is around 20 cents on the dollar and is hugely speculative on a resolution of the senior bank debt situation, of which the company has signed a 1-year agreement that the lenders will not exercise its right to default while the fund tries to dispose of assets to raise cash. If the situation resolves, the convertible debtholders might get away with a handsome capital gain, or they might end up with nothing. This situation was only interesting because the ultimate irony is that Newport’s business professes to consist of experts in finance, but somehow they managed to drag the company into a situation where they are on the border of bankruptcy. I would never suggest these debentures, even as a speculative position – the risk is too high, plus I didn’t quite fully understand the company’s structure even after reading their annual information form.

I don’t own any of the three positions I mentioned here, although I would suggest Uranium One debt to somebody that asked me “I’m looking for a decent place to sock away money for the next 2.4 years.” They’d invest $800 today, and in 2.4 years they would end up with $1,000 plus $170 in total interest payments.

Anthony has been busy solving classic puzzles empirically. First, he worked on the Monte Hall problem, and used code to verify that indeed, switching doors is the correct answer.

Another way to make the Monte Hall problem a little more intuitive is if you treat the game with 100 doors, 99 of which have goats, and 1 which has the grand prize, and the game show host flipping open 98 (goat) doors before asking you whether you want to switch or not.

The Monte Hall problem has a practical application in that people that are knowledgeable about situations may be leaking information by revealing “innocent” and presumably useless knowledge about undesired outcomes; instead, you are revealing more to others. For example, if your goal was to win a goat and not the grand prize, you would instead decline the offer to switch doors, and be much more confident about the desired outcome (2/3rds to 1/3rds) vs. a coin toss.

The next problem Anthony has worked on was Newcomb’s Paradox, a game that I have never seen before. If you give his post a read, make sure to keep your brain on as it is was not light reading.

Newcomb’s Paradox seems, on initial glance, to have application with respect to an optimal outcome that is not determined strictly by rational decision making.

In game theory, a dominant strategy is one that cannot be defeated by any other strategy; good games are designed to not have dominant strategies, otherwise they become trivial (e.g. tic-tac-toe). This is conflicted with the other aspect of games, which is mainly economic – maximizing your expected utility out of a situation. Newcomb’s Paradox exploits these two by presenting a problem where you can rationally determine to play either decision depending on what your preferred mode (dominant or expected value maximizing) is.

It turns out that one solution to defeating a omniscient ‘mind reader’ is to involve some sort of deception along the way. Of course, the predictor would be able to guess this, and re-adjust his/her strategy, ad infinitium, etc.

Still it is an interesting problem considering the “opponent” you are facing knows, with a good probability, exactly what you are thinking, yet there is still a way to behave that maximizes your outcome. At the very least, it would eliminate the option of taking no boxes, or just box A.

There is a very small element of the “Deal or No Deal” element of this particular problem; in real life, the numbers would be sufficiently higher that one would be willing to “settle” for the lower expected value decision, compared to running a risk of a higher expected value, but the possibility of a minimal outcome.

When I left for holiday, I did not touch the computer and thus did not make any transactions with my portfolio. My portfolio is mainly into fixed income securities anyway, so I am not terribly concerned with the day-to-day volatility of equity securities – which almost all the media tends to cover. It is a huge benefit to take a 3 week break from reading any news whatsoever, and coming back and being able to filter the day-to-day junk compared to the overall macro story.

A few things that got my attention:

1. The rise in the Canadian dollar was the most surprising. I had earlier indicated that I thought the Canadian dollar could get as low as about 80 cents, and I said 83 cents was more likely, but it never got below 85 cents before skyrocketing up later in July. While predicting currency markets is very difficult, it seems that the threat of quantitative easing is now over, and that central banks are going to be raising interest rates sometime in the middle of 2010.

I’m not terribly happy with this as I have been slowly moving more cash to Canadian as the US dollar appreciated, but I did get a good chunk of change out over the past year, averaging around 82 cents, so I can’t really complain about the overall timing.

2. On the short-term interest rate front, the implied rate for the December 2010 futures is at 2.03%, which is higher than what I liquidated my interest rate futures at (around the 1.7-1.8% level). My initial positions at 1.25% were likely winning positions, and I do not think that there is much profit to be made from these futures at present, unless if there is a complete surge of inflation coming up (which is possible). My instinct says that this future should be shorted, but I can’t quite pull the trigger on re-establishing a position in it.

3. Corporate bond spreads have narrowed considerably compared to risk-free products. Most of the short-duration debt I have invested in at distressed levels (e.g. 50 cents on the dollar for a 3-year duration, etc.) have now come up considerably. Time is still on my side – I estimate by the time that these products mature that my ideal investment at that point will be cash from short-term GICs since they will actually give out a yield instead of the 1.25% you can get with short term risk-free money today. There are quite a few financial products out there you can invest in to give you a far superior yield, but the real risk you will be taking is the risk that interest rates will rise in 2010 and you can get a better return on investment by investing in less risk products for more returns. I could structure a “nearly” risk-free portfolio with about 2-3 years of duration and get a yield of about 7% or so.

You can get these sorts of returns from preferred shares (especially in financial institutions and real-estate related securities) but I do not think the risk on these is nearly as safe as most people think.

4. The Vancouver real estate market appears to be picking up. My estimation is that these are people that have been sitting out on the sidelines and are pulling the trigger based purely on low rates of mortgage interest they can get. This suggests that the market is still fueled by credit availability and once rates start rising in 2010, there still be a slowdown in this market once again, especially if unemployment remains high.

5. Long-term corporate bonds are still decently attractive, although the market prices probably do not reflect the risk of higher short term interest rates in the future. I once again point out that the Sprint November 2028 6.875% debt is trading at around 76 cents on the dollar (yield to maturity of 9.86%) is trading significantly cheaper in the trust preferred form (coupon of 7%) at around 57 cents on the dollar (yield to maturity of 13.58%), which is a huge difference. Shopping around for your bonds makes a huge difference – typically less liquid products will fetch you higher value.

Throwing away a fairly low (but not zero) risk opportunity at 13.6% for 20 years is very difficult considering alternatives, which is why I continue to hold such trust preferreds. They are perfect in a tax-sheltered (RRSP, but not TFSA) account. I still think such securities will yield better in the long run than equity, especially with the risk taken. It is also a less stressful form of investing, considering all I care about is getting paid, rather than having to listen to stories about management blowing billions of dollars on failed acquisitions. The shareholders pay for management failures, while the bondholders collect no matter what. If worse comes to worse, management can dilute the equity (current market cap of Sprint is 11 billion) and pay the bondholders that way – again, I do not care about dilution. In fact, I prefer it.

6. The commodity markets are rising, and this may be a function of the depreciating US dollar, and future anticipation of inflation coming from the USA.

I also think that there will be some more upheaval in the markets this September since that’s when the primary decision-makers come back from their holidays. I have had this hunch during the early summer that the markets will plummet this September, but it appears it’s about now that the effects of government stimulus are finally hitting the economy. When the stimulus drains away, the markets will be in trouble again. Timing is everything and I feel comfortable in a mainly fixed-income portfolio, although it is very difficult to watch the people in equity right now make huge returns. I still think they are taking way too much risk.

On July 11th, I went to Newfoundland. I had something to attend in Ottawa at the end of July, so I figured if I was going to be on the eastern side of the country, I might as well make it part of a larger trip. The intention was to explore the western side of Newfoundland (specifically Gros Morne and area), get to Labrador and “see what’s there”, and explore PEI.

First, the travel logistics. Fortunately at the time, Air Canada had flight passes on sale (they called it the Explore Canada flight pass), which gave people four one-way flights (or two people two one-way flights, i.e. a round trip for two) to and from Western Canada to anywhere else in the rest of the country, for $1020 + 5% GST. It was $900 if you restricted your travel to either Tuesdays or Saturdays, but the timing did not work out for that option. Normally a round trip to Newfoundland would be about CAD$800 per person, so this represented about a 40% discount from usual prices for domestic travel. I snapped at the offer and also bought the Atlantic Pass because the itinerary would take us through PEI and then to Ottawa. The remainder of the Western Canada pass would be used to get from Ottawa back to Vancouver. The Ottawa part was “business” so I’ll just write about Newfoundland, Labrador and PEI.

I would have never have done this trip, purely for economic reasons, if the flight pass was not available. It would have been way too expensive.

The second issue dealt with car rental. Finding a car around Deer Lake, NL was impossible, so the flight coming into Newfoundland was through Gander, NL, which was a 300km drive away from Deer Lake. I have no problem driving long distances as long as the traffic is not stop-and go. Complicating matters somewhat was that the car rental had 1800km free on it before they started to ding you (25 cents per km), and I suspected that we would be around 400km over (which we were). It was an expensive rental (about $570 which included all taxes and the overage for 9 days), but was well worthwhile as it would be impossible to get around the island via any other route. Apparently there has been a shortage of rental cars, and the tourist associations in the province have been grumbling about it. PEI had no problems with availability (and no pesky mileage limit).

The third dealt with accommodations. We ended up doing a mix of camping and bed and breakfasts throughout the trip. The only hassle about the camping was that packing all the equipment into suitcases and backpacks was a real pain in the ass, but we managed. The security restrictions on luggage is particularly annoying – specifically with camp fuel. We did bring an attachment to a camp propane tank which we subsequently used to boil water and cook with. The camping at Gros Morne was good, the mosquitoes were nearly non-existent except for the last day, and there were fires allowed. The bed and breakfasts were fine – we just needed a hot shower and a bed, and we were able to find availability on them with little notice.

The flight to Gander involved a 3:30am arrival into St. Johns, and a 9:30am flight out to Gander, so the six-hour stopover was particularly draining (especially due to the 4.5 hour time zone difference). We eventually made it to Gander, hit a Walmart and loaded up on camping-type food and a camp propane tank, and made our way to Gros Morne.

The main highway in Newfoundland cuts through territory which is very sparsely populated, and is quite boggy in nature, with lakes all over the place. There is very little signs of civilization between these points, save the occasional camping site or two and the sparse amount of traffic around you.

Gros Morne park is a location of geology – it is a prime example of the Earth’s mantle exposed to the surface. It is also the northern point of the Appalachian Mountain range and has been significantly eroded over time. While in BC terms the mountain range is not very tall, locally it is.

Some highlights of this part of the trip –

1. Taking a ferry into Western Brook Pond. It was quite windy (and a bit of rain) that day, but the geology of the cliffs surrounding the lake was magnificent. The pictures don’t quite capture it. Apparently the lake was formerly a fjord, but was cut off from the ocean. It is now a freshwater lake, and it has salinity so low that it cannot conduct electricity. As it is a very resource poor lake, algae and other components of the food chain do not grow well there, and as a result, there is not a large population of fish.

2. The area around Western Brook Pond (boggy) was also interesting in its own right.

3. Listening and experiencing a genuine Newfoundland band play and tell stories. The way they turn on and off the accent is quite amazing.

4. The Tablelands is the number one location that you must explore if you are in the area. On one side of the U-shaped valley there is a barren patchwork of mountains (exposed mantle), while on the other side is a regular mountain of trees. On the mantle side is the tablelands, and the plant life in that area is extremely adapted for the conditions – they lie low and grow very slowly. It also had the most dense cluster of carnivorous plants I have seen – mainly Sundews and Pitcher Plants. If I had more energy on the trip I would have climbed one of the mountains, but there were other parts to be seen.

5. Trout River is a small fishing community close to the Tablelands, and is an example of the size of towns that dot the Newfoundland coast.

6. Broom Point was manned by a lady named Louise Decker who gave some very interesting insights on being a fisherwoman (before the days when it was politically correct to be a fisherwoman, opposed to a fisherman), and aptly demonstrated her total competence in the field of fisheries. She also had some interesting insights on the fishing industry in general (the mismanagement, and causes of the collapse of the cod industry in Canada).

After this, we proceeded to drive further up the Newfoundland northern peninsula – and this will be for another post. (Part 2 is posted here)