With the first three months of an interesting 2018 behind us, it’s time for an update on my stock picks, as well as some reflections on my 2017 tax bill - including lessons learned about the tax consequences of cryptocurrency trading. I’ll also discuss how my first-quarter stock trades were shaped by the return of market volatility. All figures below are as of March 31, unless otherwise stated.
My after-tax return for 2017 was 25.1%. I recalculated the pre-tax figure after developing a new time-weighted return algorithm (and accompanying data set), which uses regular semi-monthly snapshots to eliminate the selection bias I noted in my last article. This method produced the same 25.8% result.
Stock Trades in the First Quarter
In my January portfolio strategy article, I mentioned two outstanding goals: to increase value stocks as a percentage of all single stocks, and to increase my exposure to the technology sector if valuations were to become attractive (spoiler: they became attractive). A new goal I’ve made since then is to reduce the number of my single stock holdings to 30 or less, in order to raise the bar of confidence I have in a name before buying it, and to curb the temptation to hang on to underperforming names.
As shown in the table below, I’ve met the value target, and lowered the holding count meaningfully from 36 at the end of the year. An explanation of other targets can be found in my January strategy article. Two new items since that article are a 40% minimum allocation to stocks in my Asset Quality strategy (strategies are explained in my April 2017 article), and a formalized version of the cost basis target for speculative assets I've referenced in the past ("Spec. Cost Basis / Asset Value"). The latter target is actually a dollar cost basis, but I report it here as a percentage of the present market value of all assets.
The buying opportunities presented by the recent market volatility lowered my cash position from 13% of assets to 8% in the first three months of the year; equities, meanwhile, are up from 73% of assets to 81%. I deployed cash mainly in the tech sector, which increased from 15% of equities to 20%. In particular, I doubled my positions in Alphabet (GOOG) (NASDAQ:GOOGL) and the T. Rowe Price Global Technology Fund (PRGTX).
I also initiated a Facebook (FB) position during the stock’s late-March sell-off. I plan to hold Facebook only as long as the company’s current scandals weight on its share price. Although the regulatory threat may justify part of the bearish move, I don’t expect a notable impact on user engagement for such a habit-forming product. Over the long term, I find the risk/reward of Facebook unattractive, given its high dependence on a single product which attracts controversy like a magnet.
Other new holdings this year are Enbridge (ENB), Omnicom (OMC) and PepsiCo (PEP), ideas lifted from Morningstar StockInvestor. PepsiCo has replaced Coca-Cola (KO), as my optimism about Coke’s superior share of the overall beverage market was finally outweighed by its large exposure to a soda market in secular decline.
I’ve exited seven other names this year: 3D Systems (DDD), Express Scripts (ESRX), IBM (IBM), Monsanto (MON), Phillips 66 (PSX), SCANA (SCG) and Twitter (TWTR). Selling IBM, Monsanto and 3D Systems was part of the effort to reduce my holding count (all were among my smallest holdings). The rest I exited due to valuation concerns or changes in their investment theses.
Below is the complete list of my single stock holdings, which represent 61% of all assets:
GE’s position size was as high as 5% in January. Given the unbroken upward market trend at the time, it looked like one of very few higher-quality stocks at a bargain price. This was no longer the case after the market correction, so I sold nearly a third of my shares in late February at an 11% loss.
A quarter of my current Wells Fargo shares were purchased this year. Despite the company’s recent mistakes, and the headwind it faces from Federal Reserve penalties, the commercial banking business offers exceptional downside protection, and the stock has some of the most attractive valuation metrics among all my holdings.
The Tax Headache of Cryptos
2017 taught me that it’s much harder to trade cryptocurrencies with the discipline I apply to stock trades, and with that difficulty comes a major tax disadvantage. Although cryptos produced 44% of my 2017 pre-tax investment returns, they generated 90% of my investment-related tax burden. My effective tax rate on crypto returns (Federal and state combined) was 30%, versus a modest 3% for the rest of my returns.
Much of that 30% can be attributed to heavy selling at extremely low-cost bases, without the benefit of a retirement account. A subtler but equally important factor was my lack of foresight about the tax records I would eventually need from the three exchanges I use (Coinbase, Poloniex and Bittrex). While I credit Coinbase with making an effort to accommodate the tax reporting needs of its users, most other crypto exchanges have made no such effort. A key deficiency on Poloniex, for example, is that its transaction history for crypto-to-crypto trades is denominated entirely in cryptocurrency exchange rates, not in the US dollar amounts required by the IRS.
Documenting the correct dollar amounts of a hundred crypto trades for the IRS is hard, especially at the end of the year. I erred on the safe side by overestimating sales proceeds and underestimating cost bases. Time will tell if I care enough to replay this mission on expert mode (i.e., amend my return) for the full reward.
Cryptos finished the quarter at 1.0% of my assets, down from 2.7% at year-end. On April 1, I deployed cash in this asset class for the first time since 2016, increasing it to 1.5% of assets with a Bitcoin purchase. I’ve made no crypto sales since early January. I’m skeptical of the odds that Bitcoin will return to $19,000 or higher, given the increasing regulatory drag on the market, but with recent prices at nearly a third of that peak, the risk/reward looks compelling again.
Estimating Taxes by Asset Class
I’ll attempt to summarize the convoluted process of estimating how much of my income tax came from each asset class, but you’re forgiven if you stop reading here. The reason I do this is to get a better idea of whether active trading pays off for me, given the aforementioned tax impacts I’ve seen.
For this example, I’ll use single stocks. I begin by adding up all the numerical inputs in my tax forms which aren’t related to single stocks (i.e., wages, deductions, exemptions, and Form 1099 entries for mutual funds). Next, I look up that sum in the IRS tax table, and subtract the resulting tax number from the actual tax reported on my 1040, in order to remove the progressive component from the ordinary tax rate (my single stock income alone is unlikely to move me to a higher bracket after all else is accounted for).
I then separate the single stock income subject to the ordinary rate from that subject to the long-term capital gains rate, and apply the appropriate rate to each amount. This gives me my single stock tax in dollars, which goes into my time-weighted annual return calculation. Since this calculation is done in steps (i.e., monthly or semi-monthly) to achieve the time-weighted effect, and the tax is paid the following year, I apply the tax only to the final step (i.e., December).
I’m positive there’s a margin of error in this kind of estimate, but it’s likely immaterial for my purposes.
Disclosure: I am/we are long ALL ASSETS, UNLESS OTHERWISE STATED.
I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it. I have no business relationship with any company whose stock is mentioned in this article.