# IVY Portfolio Summary: returns, risk, & SWRs

Now that I’ve presented several variations and updates to the IVY portfolios, in this post I update all the relevant portfolio statistics and compare them to some commonly held diversified portfolios. I provide the definitions of the portfolio stats and the various portfolios. I conclude with some brief observations.

In the table below, you’ll find the various IVY Portfolios I’ve presented along side some commonly recommended portfolios and benchmarks. The statistics presented are all for the period from 1973 through 2012.

The statistics presented are as follows:

1. CAGR: Compound Annual Growth Rate. Basically, the annual return of the portfolio over the period from 1973 through 2012.
2. Standard Deviation: aka the Volatility of the portfolio. Most often used as a measure of the riskiness of the portfolio.
3. Best year: the highest return year during the 1973 to 2012 period.
4. Worst year: the lowest return year during the 1973 to 2012 period. This is the same as max drawdown if the investment decisions are only made on annual basis. More often max drawdown is measure on a monthly basis and is thus higher.
5. Sharpe ratio: risk adjusted return or return per unit of risk. The higher the sharpe ratio the better.
6. \$1 Becomes: What \$1 invested in the portfolio becomes by the end of 2012. This allow you to see how important even small differences in returns can make when compounded over long periods of time. For example, \$1 invested in a 60/40 portfolio at the beginning of 1973 became \$28.9 dollars by the end of 2012.
7. Safe Withdrawal Rate (SWR): the most important statistic for retirement portfolios. It represents the maximum annual percentage that a retiree can withdraw from their portfolio and have the portfolio last through their 30 year retirement period. Caveat: unfortunately there is no performance data for the 30 year period, starting 1966, that defines the worst SWR going back to before the great depression. That SWR is 4.33% which compares to the 5.3% SWR for the period starting in 1973. For more realistic figures I would use an adjusted SWR which is 20% less than the figures in the table.
8. Average End Wealth (AEW) Retired: My first of two stats purely for retirement portfolios. For all 30 yr retirement periods in the 1973 to 2012 period, this is the average ending portfolio value, in millions of dollars, for the retirees at the maximum safe withdrawal rate (SWR).

Below is a description of the portfolios presented in the table.

1. S&P500: the most popular measure of the US Stock Market. While almost no one would carry a 100% US stock portfolio this is the benchmark most often used when discussing portfolio performance especially in bull markets.
2. VBINX (60/40): the most common portfolio recommended for retirees or risk averse investors. The portfolio is composed of 60% US stocks and 40% US Bonds. I use the Vanguard 60/40 fund, VBINX, to represent this allocation.
3. IVY B&H 5: the original IVY buy and hold 5 asset class portfolio. It consists of equal weights of US stocks (VTI), foreign stocks (VEU), US real estate (IYR), US intermediate term gov’t bonds (IEF), and commodities (DBC).
4. GTAA 5: the timing version of the IVY buy and hold portfolio using the 200-day (10 month) simple moving average based on dividend adjusted closing prices.
5. IVY B&H 13: a broader more diversified version of the original IVY buy and hold portfolio using 13 assets classes. See this post for more information on the composition of this portfolio.
6. GTAA 13: the timing version of the IVY buy and hold 13 using the 200-day (10 month) simple moving average based on dividend adjusted closing prices.
7. GTAA AGG 6 and GTAA AGG 3: portfolios that combine timing using the 200-day moving average plus momentum, described in this post.
8. Permanent Portfolio. Another popular diversified portfolio made famous in the books by Harry Browne. The portfolio consists of equal weights of US stocks (SPY), US long term bonds (TLT), Gold (GLD), and Cash (SHY).
9. 10 yr bond. The bench mark 100% US intermediate gov’t bond portfolio. Used here to book end the 100% US stock portfolio (SPY). Only the most risk averse investors would carry a 100% bond portfolio.
10. Inflation. US inflation as measured by the CPI index published by the US gov’t.

All right. Enough definitions. There is a ton that can be said about the results in the table above since I’m running a bit long for today I’ll conclude with a few observations and have more to say about these results in the future.

At the most basic level, the biggest and most important take away from the results presented is that passive investing, diversification, and re-balancing work wonders. The more diversified portfolios of the IVY B&H 5, the IVY B&H 13, and the Permanent Portfolio provide better results across the board versus the most commonly recommended passive diversified portfolio, 60% US stocks, 40% US bonds. And not to mention the fact that even the passive 60/40 portfolio beats most mutual funds over time. The IVY portfolio’s results in particular, with their broad diversification, have impressive results across the board whether it be annual return, risk adjusted return, or safe withdrawal rates. Combined with a focus on low fees these are great portfolios for a great many investors who want to spend the minimum amount of time managing their investments whether they are in the wealth building or retirement phase of their lives. Also, the broadly diversified portfolios are best for investors that believe the more active portfolios presented below are a result of data mining or a statistical anomaly or that the out performance will go away in the future.

Unfortunately many investors cannot stick with the simple broadly diversified portfolios mostly due to behavioral biases that we all as humans have. Investors tend to chase markets, buy high and sell low, give up on stocks at exactly the wrong time, etc… So for many investors portfolios with automatic risk management rules are better choices. This is particularly true for investors in the the withdrawal phase (retirement) of their life. Negative returns during the withdrawal phase are particularly harmful for the survival rate of portfolios. For this class of investors the IVY timing portfolios are a better option. Investors get better returns with less risk as shown by the markedly higher sharpe ratios and most importantly much higher withdrawal rates (SWR) in retirement. These portfolios require more effort than the passive portfolios (buy and sell decisions made once per month) but come along with much better results. The GTAA 5, GTAA 13, or the GTAA AGG 6 or 3 portfolios are impressive all around. I’m particularly impressed with the performance of the aggressive IVY timing portfolios – this is the first time I’ve run the full portfolio stats with them. I’ll focus some more on these in future IVY updates. It seems well worth the effort.

There’s a ton that can be said about these results but that seems to be enough for now. What are your thoughts about the performance of these portfolios?

Full Disclaimer - Nothing on this site should ever be considered advice, research or the invitation to buy or sell securities. These are my personal opinions only.

This entry was posted in Portfolio, TAA Investing and tagged , , , , , . Bookmark the permalink.

### 33 Responses to IVY Portfolio Summary: returns, risk, & SWRs

1. kevin D says:

Paul:
What an out standing job! I’m sure I won’t be the only you to give you a big “thank you” for the time and effort you put into this post.
Ever since Mebane posted the expansion of the IVY a few months ago, I’ve become convinced it is the right strategy for me. Having said that, I’m moving slowly but deliberately in that direction. Thanks again. I’m sure glad we were able to coax you back into posting. If you ever head back east in the direction of Florida please let me know. Would love to buy you dinner.

Kevin D
01 CC Magna

• libertatemamo says:

Thanks Kevin. Much appreciated.

Paul

2. Scott says:

Am I reading correctly that the worst year (DD) for the GTAA AGG 3, 1973-2012 was 4.17%? That time frame is basically my working career, not looking for another job when this one is over in Dec, any plan with a DD of less than 20% looks good to this B&H guy. A little alpha goes a long way. Nice job presenting!!

• libertatemamo says:

Scott, yes, that’s correct. In 1994 GTAA 3 had a -4.17% return, its worst year during the period. Impressive, I know.

One comment on drawdowns. Its quite dependent on measurement period. So, the -4.17% worst year for GTAA 3 is its worst drawdown measured on a calendar year basis. The most popular measure of drawdown is monthly, peak to trough. Measured on this monthly basis the GTAA 3 had a worst case drawdown of -20.29%. Still damn awesome.

The reason I didn’t publish the monthly drawdowns is I don’t have that data for all the portfolios in the table. Also, I really don’t think investors should be making portfolio decisions on a monthly basis. If I get the data for the other portfolios, I’ll update the table accordingly.

Paul

• Scott Wharton says:

Thanks Paul, this is the first time in my life that I’ll have the extra time to look beyond the traditional retail investing mantra. And because of you I’ve read What Works On Wall Street, The Ivy Portfolio, Faber’s white papers etc. It a good thing the dogs like to snuggle while I read.

The hardest choice for me is the volatility I’m willing to accept. My portfolio will be approximately 60/40 taxable/non and have about 20% of the total in high dividend/MLP space with a basis yield of about 8%. I work for an MLP so when they crashed I was aware of some real bargains.

So, at year end I’ll end up with equal amounts taxable/non to rebalance. My mock Trending Value is up 6.8% vs. the S&P @ 2.65% and will want to see how it reacts during a negative market, but really like the underlying value slant. Think the IVY portfolio selection will be used to balance volatility.

Life is wonderful and amazing; from “Technomadia” I found “Wheeling It” and then “Investing For A Living”, although I’ll be installing a 12V solar system.

Thanks for the inspiration and look forward to more!

PS: And yes -20.29% is damn awesome.

• Eric says:

Hi Paul,

Just catching up on some of these IVY posts. Great posts and comments and I’ve already invested based on a lot of your insights over the past couple of years.

You said in the comment above that you “don’t think investors should be making portfolio decisions on a monthly basis.”

But isn’t the GTAA timing approach itself making cash/invested decisions at the beginning of each month?

Thanks,

Eric

• libertatemamo says:

Hey Eric, nice to hear from you. Hope things are going well.

What I meant was that I don’t think that investors should choose among overall asset allocation strategies or portfolios, like GTAA 5 vs GTAA 13, based on historical monthly statistics. I think its best to use the annual statistics. Once say GTAA 13 is chosen then yes, all the buy/sell decisions are automatic.

Makes sense?

Paul

• Eric says:

Hi Paul,

Doing well and retired from corporate work for about a year now myself.

Yes, your answer makes sense, thanks. Those AGG 3/6 approaches look great. I’ll be using those next.

Appreciate the quick response. If you’re back in the Bay Area and have time for a drink, let me know.

Eric

3. Dave & Doris Hoagland says:

Hi,

Our IRAs are now driving me insane and I am looking for a simple strategy for investing them, frankly the psychic toll has been very high for me, after initial great success it has been downhill ever since and I am ready to throw in the towel and turn the whole thing over to professionals. At age 58 it’s getting more stressful and unpleasant than in the past.

Is there an ETF that follows these strategies? I know that Mebane has GTAA and SYLD ETFs. Supposedly the expenses are high but it may be worth it for the simplicity and peace of mind. It sounds like you are basically following the strategy of the GTAA etf but doing the trading yourself, is that correct?

Is there an article in your blog that is most appropriate for someone who has yet to get started with IVY?

Thanks

Dave

_____

• libertatemamo says:

Dave,

I understand where you’re coming from. And I do think an IVY portfolio, whether it be the buy and hold version or the timing version, can bring you some piece of mind.

You can see the intro post I did on this topic a while back here. You can see all my posts on the IVY model by searching my blog for “IVY” (here are the search results). I also post an update on the basic portfolio once a month, at the end of each month.

The GTAA ETF follows a more complex version of the IVY portfolios discussed here or even in Mebane’s book. The jury is still out on the ETF due to its short existence. I would not recommend it until it has a few more years under its belt.

Paul

4. Tom W says:

Paul,

I’ve been an admirer of your work on this blog for almost a year now, and it’s about time that I commented and thanked you for all of your efforts. You have provided a ton of food-for-thought, and at a particularly critical point in my own life (career transition, uncertain future employment, possible “early retirement”, etc.).

THANK YOU very much for this post, and all of your previous posts, that have contributed to the overall knowledge base and financial thought of so many of us. Thank you!

Tom

• libertatemamo says:

No problem Tom. Happy I could help.

Paul

5. Vall&Mo says:

Paul,

First of all, thanks for the outstanding work! I’ve been reading your posts for more than 2 years and learning a lot from you.

Second, do the above numbers include costs (commissions, fees, etc)? what about taxes? I’ve been talking with a friend who’s partial to buy-and-hold and he’s skeptical of momentum strategies because they could generate much more transactions which would end up costing a lot more due to costs and taxes and eating so much into income as to make it worse than buy-and-hold.

Apparently this is confirmed by the performance of the GTAA ETF, which I understand implements the IVY5 momentum model: it’s listed on MorningStar[1] as having 1.41% a.a. costs (more than twice the 0.69% category average), and coincidentally or not, since its inception in 2010/10 it has had a net loss of 3.1% (NAV varied from \$25 to \$24.23 yesterday, again according to MorningStar[2]).

Third, when you say GTAA are you talking specifically about the Cambria Global Tactical ETF I referred to above, or are you using the acronym to just refer to general, investor-managed IVY5 momentum strategy?

Thanks,

Vall.

[1]http://financials.morningstar.com/etfund/operations.html?t=GTAA
[2]http://etfs.morningstar.com/quote?t=GTAA

• libertatemamo says:

Val,

Great questions. Thanks for asking them. I’m sure others have them as well.

Fees, commissions, and taxes are huge questions to consider in any portfolio. None of these results even for the basic S&P500 or 60/40 portfolio include fees, commissions, or taxes.

Lets take fees first. In terms of fees (annual operating expenses for the ETF chosen) it is not any more expensive to invest in say the IVY 5 buy and hold vs the IVY 5 timing model. They are the same ETFs with the same expenses.

As for commissions, most if not all the ETFs proposed for the IVY models are commission free at most discount brokers.

The big consideration is taxes. Taxes can make huge differences. Of course, if one implements these models in IRAs or 401Ks the issue of taxes goes away completely. For investing in taxable accounts the more active timing or momentum models taxes will take a bigger bite out of returns vs the buy and hold models.
Mebane has a good discussion of this in the paper on page 36. Here is the key paragraph:

“Gannon and Blum (2006) presented after-tax returns for individuals invested in the S&P 500 since 1961 in the highest tax bracket. After-tax returns to investors with 20% turnover would have fallen to 6.72% from a pre-tax return of 10.62%. They estimate that an increase in turnover from 20%-70% would have resulted in an additional haircut of less than 50 basis points to 6.27%.
There is some good news for those who have to trade this model in a taxable account. The system results in a high number of short-term capital losses, and a large percentage of long-term capital gains. Figure 17 depicts the distribution for all the trades for the five asset classes since 1973. This should help reduce an investor’s tax burden.”

Buy and hold portfolios generate taxes as well; from dividends, annual re-balancing actions, and capital gains distributions. The question becomes how much more incremental taxes do the timing and momentum models generate. According to the study referenced by Mebane its ‘only’ an extra 0.5% per year. So, if a timing or momentum model generate more than an extra 0.5% per year in returns it looks like its worth it.

That’s the theory anyway. And the theory seems to support timing and momentum even after taxes, fees, and commissions. Now there is reality. Reality and the facts say that no matter what the theory says the majority of investors cannot stick to buy and hold no matter how hard they try. In many ways buy and hold is a paper tiger. Most people can’t do it. The timing models should be looked at as primarily risk reduction, sleep at night, strategies.

And then when you start talking about investing in retirement the discussion changes yet again. The differences in SWRs for the timing and momentum portfolios way overcome the differences in fees, taxes, and commissions. In retirement its mainly about reducing or eliminating negative returns which buy and hold portfolios do not do at all.

My 2 cents. Maybe I’ll make this another post.

Paul

• libertatemamo says:

When I mention GTAA in my post I’m referring to the terminology used in the research paper by Mebane and NOT to the ETF, GTAA, which he launched a bit ago.
The jury is still out on the GTAA ETF. He implements a more complicated model than GTAA 6 which I discussed. And so far the performance after fees does not match what you can do on your own implementing these models. The ETF needs a minimum of 3-5 years before passing judgement on it.

Paul

6. Jeff M says:

Paul,

Your website has changed my life. I can smell the fresh air and freedom down the road! I am implementing the IVY GTAA AGG 3 or 6 in my Vanguard IRA and Trending Value in an OptionsHouse taxable account to access in my 30’s. Upon retiring, I will switch Trending Value over to Enhanced Dividend Yield or Consumer Staples/Utilities. Although maybe I should do IVY in taxable and quant in IRA for tax efficieny since IVY seems like it could grow fast enough. I recently discovered your blog from Jacob Fisker’s Early Retirement Extreme and I am happy to hear the good people before me convinced you to keep blogging!

A sincere thank you,
Jeff

• libertatemamo says:

Thanks Jeff. I like your strategy.

Paul

7. Bryce says:

Paul,
I think the returns for the Aggressive portfolio’s are nothing short of remarkable (they almost seem too good to be true). The thing that stands out the most to me (other than the stellar returns), is the fact that since 1973 the AGG 6 only had 4 down years (-5.76% being the worst) and the AGG 3 only had 3 down years (-4.17% being the worst). That’s crazy! I don’t think that the MaxDD being twice what it is for the Conservative and Moderate portfolio’s should be a consideration. When you look at the calendar year returns you can see that any monthly draw down is truly short lived.

Bryce

• libertatemamo says:

Hey Bryce, totally agree. Its very unfortunate that most investors can’t even keep a time frame of one year. Most investors do react on a monthly and even shorter basis which puts them at a great disadvantage. I’m constantly amazed at the changes in fund flows based on the most minute news.

Paul

8. Ellis says:

Hi Paul,

I have two problems

1. MMTM do not have enough history for the momentum calculation, do you have any suggestion for other ETF?
2. any suggestion for the secondary data source for calculation for the signal? I am using yahoo data.

Cheers,

Ellis

• libertatemamo says:

Hey Ellis,

MMTM is the best choice for that asset class. You can just use the returns that are there so far to calculate the averages. I think this is a better option than substituting an inferior ETF.

I just use Yahoo data myself.

Paul

9. Scott Wharton says:

Hey Paul, with the July numbers in, I looked at the GTAA 13 (i.e. 12) portfolio and got 6 funds to invest using the 200 SMA. Those same 6 were momentum ranked and by coincidence were 1 thru 6 (momentum ranking 1st then compare price to SMA for the agressive models). These 6 are the aggressive 6 and the top three would be the aggressive 3.

So here is the rub, MMTM liquidity, MMTM would be in both aggressive models as “Invested”. Take now (about noon CST on 8-1) the B/A is 1X24 with a 0.10 spread and no trades yet today. That may work with a 6% of portfolio investment under the GTAA 13 but with one of the aggressive models, well, don’t know you could get in/out when you wanted.

I guess there are lots of options, all of which move a bit further away from the original portfolio structure. Any thoughts?

Scott

ps – impressed with your excel skills!

• libertatemamo says:

As reader Bryce found, MTUM is a better option for the GTAA 13 model. It is also young with not a lot of history. That’s the ETF I would use for now.

As for excel skills, I stand on the shoulders of others. Mainly copy/paste from reader Bryce’s spreadsheet with some edits.

Paul

10. Mike says:

Thanks for the work you are doing on this site. It’s incredible! My question relates to the weighting on the GTAA AGG 6 and GTAA AGG 3 models. Do you weight each of the top 3 or 6 ETF’s equally?

• libertatemamo says:

Thanks Mike. Yes, the AGG3 and AGG6 equally weight the ETFs.

Paul

11. John says:

What is the risk of ruin cutoff that is used when calculating the SWR?

• paul.novell@gmail.com says:

John, the SWRs I calculate are 100% success rate SWRs – the max SWR for a 30 year retirement period that led to no failures, i.e. no portfolio going below zero dollars.

Paul