The value of deferral http://www.valuewalk.com/2016/08/how-warren-buffett-used-deferred-taxes-to-make-twice-as-much-money/ " “In economic terms, the liability resembles an interest-free loan from the U.S. Treasury that comes due only at our election… Imagine that Berkshire had only $1, which we put in a security that doubled by yearend and was then sold. Imagine further that we used the after-tax proceeds to repeat this process in each of the next 19 years, scoring a double each time. At the end of 20 years, the 34% capital gains tax that we would have paid on the profits from each sale would have delivered about $13,000 to the government and we would be left with about $25,250. Not bad. If, however, we made a single fantastic investment that itself doubled 20 times during 20 years, our dollar would grow to $1,048,576. Were we then to cash out, we would pay a 34% tax of roughly $356,500 and be left with about $692,000. The sole reason for this staggering difference in results would be the timing of tax payments. Interestingly, the government would gain from Scenario 2 in exactly the same 27:1 as we – taking in taxes of $356,500 vs. $13,000 - though admittedly, it would have to wait for its money." No wonder his favorite holding period is forever
http://www.zerohedge.com/news/2016-...esperate-attempt-project-image-yuan-stability China changes FX basket so they dont have to devalue more next year. Both Obama, China and some others are all making moves before Trump takes over
For future reference I'm saving a chart of BRL investment returns. Markets will be closed to day in Brazil so they already wrapped up 2016 market performance
For reference, I'm going to save the closing prices for 2016 for the main positions I have on BR USD 2024 bonds 122.7450 (IB mark) EWZ $33.34 EWZS $10.97 Bovespa 60.227,29 BRLUSD 3.250 PSH $14.54 BRKB $162.98 GLD $109.61 UST bonds 2045 STRIP 40.73, 2045 2.5 bond 88.625 TLT $119.13 OAK $37.50 VRX $14.52 GDX $20.92 GDXJ $31.55 GREK $7.79 FMCC $3.74 FNMAS (preferred) $8 Brazil REITs (too many to count, they dont trade in the US also) I'm short some EWZ, BRKB puts also, but it isnt big. I'm also long some HLF to hedge PSH but still net short Soon, I will count my returns and compare with Dalio's risk parity allocation. Then post my current allocation
I'm looking at significant LT allocations to specific country ETFs, for example Nigeria, Colombia, Vietnam. While a country's index can certainly go lower, It's not likely to hit 0(and if it does there's some SHTF issues), and in the LT as far as I know there's a rally at some point when macro turns around. No govt in the world ever just says, fuck it. Reforms will be made.
So I calculated some of my numbers and it came out at 2016 Return: 8.32% Key drivers EWZ EWZS +3% (not counting FX gains) BRKB +2.1% BR USD bonds with interest +1.85% PSH -1.8% (counting VRX hedge) FNMA FMCC FNMAS +1.55% UST bonds and futures +1.2% OAK -1% VRX -0.8% ZQ futures +0.6% vs Dalio "All-Seasons" portfolio Even though most of my investments are in USD, I redenominate a lot of them to BRL by using derivatives to control my FX risk. About 50% or so of my exposure is in BRL. If I count the interest income implied in the derivatives that i receive by rolling them over, then my gains go higher but I have mixed feelings about adding that. If I add only half of them (the 'real' return, my return goes to ~9.9%) I was able to beat the Dalio portfolio partially because I didnt bagheld GLD and TLT on the way down after the election. I still had some but I had cut my bond durations on the way up and lightened up in Gold after the election My biggest mistake was probably not selling more BR USD bonds when I had the chance, now the liquidity has gotten really thin there (the other BR USD bonds are fine, seems that the only one that dried up is the one I own). I could have sold some at nice gains at 129 bid and 131 offer when bonds were up a lot globally. Now the bonds are 116 bid and IB is marking them at 122.75 My best decision was to get in Brazilian stocks (late 2015) and add more as they went lower Lets see how 2017 plays out, I'm positioned to have small losses or huge gains in a lot of my positions. It can potentially be a fantastic year
My allocation for 2017: EWZ EWZS 16.9% PSH 14.4% BRKB 12.2% BR USD 2024 bonds 12.2% (5% YTM) TLT 8.7% GLD 5% GDX GDXJ 1% US 2045 Bonds and Strip 5.4% Brazil REITs 4.3% OAK 4.1% FMCC 1.4% FNMAS 1.5% (preferred stock) GREK 1.7% (stop at ~2016 lows) MDLZ 1.6% (short-term trade, will exit it soon if no merger news) VRX 1.4% HLF 0.9% (hedge to PSH) BRL bonds 0.2% EWZ and BRKB short puts 0.2% (will roll over when the VIX picks up) By class Stocks 56.1% Bonds 26.5% Gold + gold stocks 6% REITS 4.3% Other (cash and others) 7.1% Off this 50% is BRL denominated (outright or through derivatives), 5% is in gold and the rest (45%) is in USD. I got a lot of USD exposure but I'm forced to have a lot because: -USD annual fees and costs (software, data, imported products, others) that I have -I need margin to run short-term trading strategies so If I put on a huge BRL derivative position and the BRL tanks a lot, I will run out of margin and miss out on trading gains -International travel costs -Safety (its hedges out my country's risk) If you consider the leverage of PSH, BRKB and the beta of EWZ, one could argue my real stock exposure is 70%+. But those things are quite cheap, PSH is trading 20% bellow NAV, BRKB is ~20% above the Buffett buyback and EWZ has a CAPE ratio of 9-10. Limited downside but big upside. Same thing with FMCC, FNMAS, VRX and GREK So I'm overweight stocks, underweight bonds and gold/inflation hedges relative to my "benchmark". Its hard for me not to be underweight bonds, Dalio recommends a 40% allocation to long-term bonds (20+ year) plus 15% to 7 to 10 year bonds. That's a lot of duration to own in this enviroment, its hard for me to get comfortable with that (and if you notice recently, I added a fair amount of duration recently to counter the fact that I was super underweight). So better valuations got me a little more comfortable but its still hard to put 50%+ on UST bonds. Dalio recommended that in 2014, and he has been talking bearishly about bonds recently, maybe even him would tilt things in his portfolio if he had to recommend an allocation starting today. And on the inflation part, I got a fair amount of Brazil exposure, a commodity country, so there is a shadow long in commodities there. As a result I think 6% in gold and gold stocks provide me with enough protection due EWZ, REITs, exposure to commodity prosperity etc Best of luck to everyone in 2017
I decided to build a US database of investment returns to run some historical tests on. I used a Damodaran excel spreadsheet to get S&P500, 3m T-bills and 10y UST bonds returns since 1928. I then added Gold prices to get gold returns. I also put in the year over year CPI and then calculated the real returns since 1928. I reused my excel functions from my Brazil database to calculate how a portfolio over different assets would have done since 1928. I run different tests and it all looked it was supposed to (diversification works, limting stock exposure is important to mitigate downside volatility, etc) but then I thought "wouldn't it be cool if I could run an algorithm that automatically finds the best portfolio possible for any given piece of history?". It turns out that there is an way to do that, through Excel "Solver" tool. Solver changes a few variables (like the weightings of different asset classes) and then "checks" to see how that affects the portfolio (by checking things like the Sharpe Ratio, portfolio volatility, or anything that I choose). I'm pretty excited about the use of this tool, it was hard work but it has paid off very well. It might not find the ultimate most 'optimal' solution (the best possible asset mix) but it gets pretty close, especially if I run it multiple times with different starting inputs.
Buffett asks "why do people buy gold, it doesnt pay a dividend", I will tell you why mr Buffett, its because it sends your Sharpe Ratio or Sortino Ratio (which is an improved version of the Sharpe) through the roof! Both of the "best" (or close to the optimal best) portfolios, historically, needed quite a bit of gold to reach the peak in the Sharpe/Sortino Ratio as can be seen bellow Not everybody is like Buffett who is ok with putting 100% of their money in the stock market. Some people are not comfortable with risking a 50% drawdown every 10-20 years and a 80% drawdown every 50-100 years. For those people (99% of the population of the earth), adding gold looks like a necessity, it protects the investor against a deflation (depression, like in the 30's) and inflation (like in the 70s), plus it produces a small real return over long periods. I tried to mess around with Solver and I cant get a high Sharpe ratio allocation that doesn't rely on gold. Solver will quickly start putting money into it once it sees that it helps with the Sharp/Sortino (up to a point, which seems to be at 20%) That 20% point is interesting, because its close to the 15% of gold+commodities that Ray Dalio recommends. I'm trying to reverse engineer his portfolio a bit using this spreadsheet and Solver. So far, it looks like it explains why his inflation protection allocation is what it is. That 15-20% point is where you get the most bang for the buck in terms of Sharpe/Sortino What I dont understand is why he recommends 40% long-term bonds. Its tough because I do not have data on 30y bonds going back to 1928. The use of 30y bonds became widespread since 1977 so the data is very much limited I need to do more thinking about this
Here is another cool finding. I asked Solver to tell me what was the portfolio since 1928 that had the lowest downside volatility (standard deviation of negative return years). What I call the "widow portfolio". Solver came back with this Along the lines with my previous theory that the widow cannot just sit on short-term fixed income and think she is protected, she needs some duration and/or inflation protection The 100% T-Bill portfolio is close in results But whats interesting is that its possible to reach a similar result using a vastly different portfolio Not quite as good in terms of SD of Negative returns (3.03% vs 3.26% for the latter) but you get a big bump in real returns, which boosts both the Sharpe and Sortino ratios. % up years also improve, max DD as well The dumb widow sits on savings accounts and gets killed during inflations/financial repressions, the smart widow diversifies and carries less risk as a result