The Amateur Investor Manifesto, Part 2

17 responses so far ↓

• 1 brad // Dec 11, 2008 at 11:35 pm

  I’m curious why you’re not interested in the TD e-Funds (which include a US currency neutral index fund as well as Canadian indexes.

  There’s a list of low MER index funds here that’s pretty useful, although it hasn’t been updated since April:

  http://www.bylo.org/idxfunds.html

  Speaking of TD’s eFunds, can anyone explain to me how (or whether) I can track these funds’ performance automatically in Quicken? I haven’t figured out any way to do it.

• 2 Canadian Capitalist // Dec 12, 2008 at 12:13 am

  In Unconventional Success, David Swensen says that portfolio construction is part art and part science. It is an apt metaphor because constructing an “optimal” portfolio based on *past* return, SD, correlation data tells us very little about the future. In other words, portfolios that were optimal may not be in the future. Beyond thinking carefully about the split between fixed-income and stocks, I wouldn’t worry too much about the exact allocations as long as it is “reasonable”. Yes, “reasonable” is a weasly word but I would consider a 50% allocation to emerging markets unreasonable but 10% reasonable.

• 3 Jordan // Dec 12, 2008 at 2:20 am

  @ Brad

  The reason we aren’t going to use them for our main Amateur porfolio is for our level of assets the ETFs are less expensive, and there is a much wider range of asset classes and providers available. For example some of Vanguard’s funds have only a 0.07% MER.

  But I do like TD’s eSeries funds, we currently use them for my children’s RESP account. Which is pretty much an exact copy of CC’s “Mini Sleepy Portfolio”. But in the new year I will be moving them over to ETFs as well, just for the lower fees.

  I am able to track TD eSeries funds with Microsoft Money which gets data from the MSN Money site. They are also available on Google Finance. Quicken must use a different data provider which doesn’t have them. It sure can be a pain to track manually.

• 4 NN // Dec 12, 2008 at 2:32 am

  Wow, it took a long time to read through all the comments on the first post – congrats Reader J on getting everyone involved.

  The mean-variance model is not of much use, as the variance-covariance structure of asset class retuns will change for significant periods of time from the long term structure. (I track my own portfolio, and am amused how the ‘optimum’ mix of assets have changed over time).

• 5 Jordan // Dec 12, 2008 at 2:43 am

  I actually have something to add to the last question in this original post. Since writing it I discovered an interesting excel toolkit called Quantext or QPP (www.quantext.com)

  It’s sort of like an MVO, but this is a gigantic spreadsheet that will automatically do a back test on the historical behavior of your portfolio. Based on your allocation and the data it extracts from Yahoo Finance it is able to provide you with the actual projected annualized rate of return, standard deviation, beta, yield and it calculates the assets correlations to tell you the level of diversification.

  It’s fun to play around with the free trial but I think CC is right that it’s impossible to pick the “best” allocation this way, it would only help to understand how the assets have behaved together in the past.

• 6 Jordan // Dec 12, 2008 at 4:21 am

  @NN

  Thanks, I’m blown away by the response so far and eager for more discussion

  I’m curious, how do you track your portfolio’s variance-covariance to find the optimal mix? Are there any “normal” spreadsheet functions to derive it?

• 7 Four Pillars // Dec 12, 2008 at 9:51 am

  I’ll agree with CC – having been pretty keen on asset allocation for a couple of years, at this point I don’t worry so much about the details.

  It’s kind of like retirement planning (or predicting long-range weather forecasts) – you can use good estimates but I’m not sure how much value you get from being too detailed.

• 8 Smart Alloc // Dec 12, 2008 at 1:25 pm

  Jordon

  You are asking questions that are near and dear to my heart. Here’s some food for thought:

  1) Think of asset allocation at four levels – asset class, geography, business sector and company size.

  2) Allocation depends on your ability to take advantage of different tax benefits. For example, with a define benefit pension plan, there is less room to contribute to a RRSP – my wife is a good example. With a larger portion of savings in a taxable account, Canadian dividends are preferred over foreign dividends, and it makes sense to have a larger Canadian allocation.

  3) Factor your job into allocation. I work in the oil and gas industry with bonuses, options and job security dependent on sector performance and as a result significantly reduce my other investments in this sector.

  4) I have never seen the problem of Canadian versus foreign asset allocation argued in a quantitative way. It is just rules of thumb – not very satisfying! Has anyone seen a logic, quantitative argument?

  5) Do you own a house or other real estate? Will you inherit your parent’s house? This should be considered in your allocation in REIT or other like investments.

  6) If you have a spouse, the allocation between partners needs to be considered. And, what happens if you split?

• 9 Sampson // Dec 12, 2008 at 1:45 pm

  Is there evidence (I suppose using the analysis tools you mention) that portfolios with geographical allocations mirroring the size of the respective markets IS in fact a lower risk strategy than a portfolio showing bias towards your home nation?

  If so, then would it be good to simply design a portfolio consisting of say 10 full market ETFs representing the 10 largest markets of the world?

• 10 Jordan // Dec 12, 2008 at 1:51 pm

  @ Smart Alloc

  Those are interesting points and good questions.

  2) Tax considerations are high on my list as well, and a number of them will be discussed in the next post I think. Briefly we don’t have any pension, but we have maxed out RRSPs (and soon TFSAs), a private corporation with invested retained earnings and our children’s trust account (taxed in our children’s hands)

  3) This is a very interesting question I hadn’t considered. I’m a self employed web developer but my main client for the past 5 years is in the US but get paid in CND. I guess if I invested in sectors this would be a good reason to avoid investing in high tech. Can you think of any other effects?

  5) We don’t own any property, we’re waiting for the Vancouver housing market to crash and burn. Then we can swoop in and buy something that’s affordable.

  On a related note I actually just discovered an alternative form on real estate investing that is more like a mortgage called a Mortgage Investment Corporation (MIC) on the CanadianFinancialDIY blog. They might behave differently as an asset class compared to a REIT which is based on renting. I’ve never seen this discussed anywhere so will be researching into it.

  6) Do you mean the allocation between our investment accounts? Generally I think it’s best to just treat all of our accounts as one big asset allocation and just try to stick the assets into the most tax efficient accounts based on how much it distributes.

  But I’m dealing with 7 different investment accounts already (9 when the TFSAs are open) so I am already finding it a struggle to plan out which assets go into which account. As a programmer I’m feeling pretty close to writing my own stock database app to optimize the most efficient account/asset plan. I just wish I knew of an existing solution instead.

• 11 Jordan // Dec 12, 2008 at 2:22 pm

  @Sampson

  I think the argument is if you cap weight the individual securities by buying the index then you should do the same for the countries. I tend to believe this is a better allocation because it is much broader and more diversified, but there are reasons for home biases. Canada is is where our spending power is used.

  Beyond that I don’t really have the tools to model portfolios myself to find how much better one is over the other. I have a hunch there probably is a little bit of benefit available in tweaking the allocation just right, so the best I can do is copy the IFA portfolio which does use all of those tools to analyze, compare and build their portfolios based on their very long historic performance.

  Does anyone else do the same? Is it a good idea or better to stick with the tried and true simple allocations?

• 12 Rita // Dec 13, 2008 at 12:04 pm

  I’m curious, how do you track your portfolio’s variance-covariance to find the optimal mix? Are there any “normal” spreadsheet functions to derive it?

  Jordan, you might want to take a look at the following book – “Microsoft Office Excel 2007: Data Analysis and Business Modeling” by Wayne L. Winston.

  Chapter 62, Simulating Stock Prices and Asset Allocation Modeling, provides two detailed examples how to set up Excel spreadsheets to address the following sample questions:

  I recently bought 100 shares of GE stock. What is the probability that during the next year this investment will return more than 10 percent?

  I’m trying to determine how to allocate my investment portfolio between stocks, T-Bills, and bonds. What asset allocation over a five-year planning horizon will yield an expected return of at least 10 percent and minimize risk?

  I’ve personally found the technique used here – it’s called bootstrapping – to be rather interesting.

• 13 NN // Dec 13, 2008 at 6:21 pm

  Jordan – I copy the unit prices of the different funds I invest in into an Excel spreadsheet, convert it to percentage change (or return), and so build a performance history over time. The variance-covariance and mean (expected) return is easily determined from the performance numbers, and Solver used to choose an ‘optimal’ portfolio.

  I actually own the book Rita referes too, but have not worked through the relevant chapter yet…

  The point is the ‘optimum’ portfolio changes with time as well, and quite significantly given the recent movement in markets. As such I am in CC’s camp that too much hairsplitting about the exact percentage that would be an optimum allocation is a waste of time. I think it would be better to choose an allocation that delivers the required return, and learn to deal with (ignore) the emotions invoked by portfolio volatility, rather than to determine the optimum allocation to the second decimal, only to find out it changes from month to month.

• 14 TJ // Dec 17, 2008 at 1:55 pm

  Hi there,
  Could somebody suggest to a neophyte Canadian investor, what potential disadvantages exist regarding investing in Vanguard? Jordan mentioned that he wished that Vanguard was CDN… why? Can I open an RRSP account with Vanguard, just as I would with any other CDN financial institution, and enjoy all the benefits? Are there problems associated with currency exchange, which would errode my earnings?

  It seems to me that Vanguard has the lowest MERs in the marketplace, for potential investors interested in ETFs. Assuming a CDN investor can use Vanguard to create an RRSP account, why would someone invest somewhere else?

  Thanks very much for any input/comments/advice you might have.

  Sincerely,
  TJ

• 15 TJ // Dec 18, 2008 at 10:58 am

  Hi there,
  Just as a follow up to my previous post, I just called Vanguard and they said that Canadians cannot set up RRSPs through their company. Am I missing something? It seems that throughout the CC website, that Vanguard is promoted as viable option to CDN investors…

  Thanks for any input you might have,
  Tim

• 16 Jordan // Dec 18, 2008 at 6:39 pm

  @ TJ

  Vanguard traditionally has been a mutual fund company with the unique advantage that the company itself is owned by its funds not other shareholders. I like this because it shows a strong incentive to keep costs low, which you saw as they generally have the lowest MERs.

  Since Vanguard is based in the US you would have to live there to hold your account with them. That’s why they don’t offer an RRSP account.

  They do offer some of their funds as ETFs which means they are traded on the stock market so anyone can buy and hold them at any brokerage internationally, including in a self directed RRSP or TFSA account at your bank or a cheaper online brokerage like Questrade.

  I wish Vanguard would come to Canada because of their low fees. They could offer the lowest cost Canadian index, and give our market some much needed competition to lower fees and give the average investor good advise like buy a low cost index fund and hold it. They could also offer local mutual fund accounts where people with less to invest could easily make regular monthly deposits without the fees or hassle associated with ETF purchases. Think TD e-series funds but with fees around 0.1% instead of 0.35%

  Lastly the downside to some of investing with Vanguard is all of their funds are USD based, so you have the cost of currency conversion and risk of it going up or down to be aware of. For me that is less of a problem, but should be considered.

  Good luck TJ.

• 17 The Amateur Investor Manifesto, Part 1 | Canadian Capitalist // Jul 12, 2009 at 6:38 pm

  [...] Are you a De-Value Investor? The Amateur Investor Manifesto, Part 2 [...]

Leave a Comment