I’ve always limited mine to more like 10-15% but not based on much analysis.
@doschicos, you’re still at a better place on the risk-reward curve than I am :))
^Assuming the curve is meaningful. It looks legit with the graph and all except o come up with a graph, I’d think they had to assume a few things and the validity of the study would be highly dependent on the assumption. Is the way they assign a numerical value to risk valid? I get that the higher SD, the risk will be higher. But How do you extrapolate the SD to risk? Do you use linear scale? Would you have the same optimal point had you used a log scale? More fundamentally, is there any reason to believe it will follow a neat mathematical scale to our convenience? If not, have they studied why they used the scale they used in producing so-called sweet spot?
To complicate the matter, one also has to question the validity of using past data. I think it’s tricky for all stocks to infer from past data. When it comes to international stocks, it gets worse in that their economic growth also depends heavily on the internal politics as well as outside politics, is the US opening the market, do they give special tariff arrangement, etc.
There are as many graphs as opinions, and we know what they say about opinions :)). I don’t place 3 significant digits of validity on any of them, but the efficient frontier is worth considering.
Here’s another one that shows some different date ranges:
https://encrypted-tbn3.gstatic.com/images?q=tbn:ANd9GcSsGW_ApcWTPLpWS6uXvQ_UyOxWGHAUuqPaX_KAwaKLqZSM9Slm
@IxnayBob, very interesting. Apologies if this has been asked before, but where did you get your data? Not challenging it at all, I would just like to get more information.
That efficient frontier was what I was quoating about Bernstein’s 60/40 allocation a few posts back. I did pay attention but I also thought it could be a house built on cards. I think most of them are assuming some kind of linear scale namely risk goes up as SD goes up in the same way across all values oF SD. Couldn’t they be nonlinear and risk may follow SD in a different manner depending on the value of SD, risk may accerlerate or decelerate as SD gets too high? They have smaller number of data for extreme SD presenting a greater margin of error. Depending on the scale you use the curve can be bent differently producing different efficient frontiers. I think I can come up with a scale to produce any shape of the curve I desire putting efficient frontier any random place. They may have done the study for all possibilities and came up with what they present is valid. Or they may be putting up a fancy graph using a simple scale without backing it up with a thorough study. After all, financial world didn’t prove they are that trustworthy.
@Dave_n, I get my data from any number of places, usually second-hand at Bogleheads. I consider these graphs indicative rather than absolute truth.
@Iglooo, it would be interesting to see what happens under different scales, but as long as you use the same scale for international and domestic and there’s some rationale, I guess it’s fair. I would also drop countries (e.g., Venezuela) that are outliers.
When did CC start allowing embedded images? What tag did you use?
I tried “img” and it worked. I had previously used “spoiler” and saw that it worked also. I assume that, if it’s available, it’s okay. If not, better to ask for forgiveness than permission.
ETA: spoiler for an entirely different purpose 
I didn’t want to think about the data since I had a feeling there was a lot of handwaving. Now that I got into it, I can’t stop thinking. How do they define risk? Is it whether you maintain the value you put in or not? Like you put in $1,000 and you will have $1,000 or more? If that’s the case, wouldn’t it depend on when you check? Risk after 1 year would be different from 10 years later? If SD is annualized, does that mean risk a year later? Is that still useful if you are on a long term?
I think the bottom of that graph is volatility, which can be measured or computed.
Risk is a pretty nebulous term, but I think it is pretty well accepted that volatility is a proxy for risk. I think the idea is that, if I pulled out of the investment at any given time, what are the odds that I get less money than if I invested in something with the same returns but less volatility?
NRE, That would seem reasonable intuitively. The thing I don’t understand is that at any given time would make sense if stocks behave randomly. But do they? If you ignore day to day or month to month fluctuations, they go up in general, don’t they? That’s what historic data tell us. To me, it means you can’t really say what risks are without also specifying the time frame.
Interesting plan to save Social Security, by Sam Johnson (R-Texas), who is a member of the House Committee on Ways and Means where he serves as the Chairman of the Social Security Subcommittee.
http://samjohnson.house.gov/news/documentsingle.aspx?DocumentID=398516
Some analysis by Forbes:
http://www.forbes.com/sites/andrewbiggs/2016/12/08/at-last-a-new-social-security-reform-plan/
Basically, it
- raises full retirement age to 69 and maximum age to 72
- raises benefits for low and very low career earners
- cuts benefits for high earners
- limits the size of spousal benefits for high-income retirees
- tweaks taxability of benefits (although not for a while)
- eliminate COLAs for high-income retirees (makes Roths more important?)
- changes how the earnings are counted toward the calculating the benefit
plus some other stuff, but I think I got the highlights.
These changes make SS totally solvent for at least the next 75 years, without raising the payroll tax rate or income cap. There’s some docs on Rep. Johnson’s web site that show the calculations, but they are not easy to figure out.
The tl;dr version, as far as I can tell, is lower benefits by raising the retirement age, lower benefits even more for well-to-do retirees, and give more to poor retirees.
I feel like we are going to lose a lot more than the 20-25% haircut that happens when the “surplus” is gone, so I am not sure how I feel about this yet.
When the retirement age is raised, what happens to younger workers trying to get into the workforce? Any change to the SS system has to address both the fact that workers in industries that are hard on the body (as opposed to the indoor, chair at a desk position most policy makers enjoy) can’t realistically work at that age and that many older workers are pushed out of jobs and unable to find other employment. These proposals often make it sound like people have a choice.
Really, I think that’s the most dangerous element in discussions about SS, employment and workers benefits - the myth that most people have a choice.
@lefthandofdog , maybe I just woke up on the wrong side of the bed, but I think (meaningful) work will become a luxury over the next years.
I will be sorry to see COLAs go away. Maybe I’ll have to rethink TIPS.
I think the proposal keeps the early retirement age the same at 62, however the benefit amount at 62 will be smaller if full retirement age is increased.
Raising the retirement age, for all the talk about “people are living longer and working longer blah blah blah”, is really just an insidious way of cutting the benefits of those affected. Every year FRA goes up is something like an 8% reduction in lifetime benefits.
It looks surprisingly reasonable to me if they keep early retirement age 62. We do need to make SS solvent and the proposal puts pain mostly on high earners. Surprising move from a republican. What does full retirement age 72 mean?
Full retirement age would increase from 67 to 69, and the maximum age you can delay while increasing your benefit would increase from 70 to 72.
Changing the taxability of benefits is an interesting idea, but under this plan it wouldn’t start to phase in until 2043, and wouldn’t be fully implemented until 2053. Good for my kids I guess, won’t have that much affect on me most likely.
On this thread we need to keep communicating on what is currently happening! Trump is talking about privatizing Medicare? Wow, we all know what that would mean for us on premiums.
SS and Medicare are two areas that will affect all of us on our household income and household expense.
We will have to learn ins and outs of Medicare next year as wife becomes eligible in December. It looks like there already exists a choice between traditional Medicare and Medicare Advantage. Is Medicare Advantage a voucher-type program already? I believe any proposed changes to Medicare wouldn’t affect current or currently close to retirees, but would affect younger people in their future. Similarly, it looks like Johnson’s plan to save Social Security is “For retired worker and disabled worker beneficiaries becoming initially eligible in January 2023” so that would be 55 years old today. At least the numbers he crunched appear to include that assumption.