HN user
Solve what's meaningful for you today.
If something more meaningful arises, you can switch course in the future.
State upfront that you will not hold regrets, for to regret a decision implies that you had knowledge of how it would turn out when you originally decided. You do not have any such idea - so there is no room for regrets.
Great job. Looking forward to checking it out!
Cool idea that I'd like to see gain product/market fit in some way - how do you plan to monetize this?
that another cool thing about the Kivalia framework. We have time-stapped market views, so we can confidently go back and say "Here's what we would have recommended for your plan 9 months ago".
I think you're obligated to do something like this if you're in any way providing active advice.
I do like the idea of strategically selling volatility in a portfolio. I've been toying with the idea of adding some of these orthogonal (or nearly so) parameters in the Kivalia framework...as it would be cool to see how a portfolio is positioned with regards to things like vol, momentum, etc.
I'm sure we'll come back to this, but we'll want some stable ETFs to use as factors. Russell had a slew of interesting ETFs but closed them down due to lack of interest.
Interesting catch, Paul. If you look at the plan itself (E*Trade Commission-Free ETFs) there aren't a lot of fixed income options available in the list and all of them seem to be international. So, the first point here is that it's tough to get a truly low-risk, low correlation instrument within this list.
The chinese yuan is an edge case (ie. we don't classify it well because it's currently outside the granularity of our framework - that will change), but if you think about it, probably not a bad choice. The yuan is pegged to the U.S. dollar and has a return (unlike US money market).
So simply, yuan is being used as a U.S. fixed income proxy in this case. The benchmark for the conservative models is about 40% bonds, 60% stocks.
Tell you what you can do - subscribe to the model and add a U.S. treasury ETF (eg. TLT) to the list...then re-run the models. My hunch is you'll find it heavily weighted in the models.
Currently fund ratings are determined as risk adjusted excess return versus the fund's best fit index. A "C" is average (no curve here!), so you'll find most index funds will be rated C. Funds that have returned more than 3% more on an RAR basis over the past year are deemed A.
We'll want to revisit ratings as we build out - to look at the graphic representation of excess returns in a universe and then customize ratings around reasonable percentiles.
On the +/++ and -/-- portions of the ranking: these are simply momentum scores for the sector in which the fund participates. So, an A++ fund is a manager who is doing well on a risk adjusted basis, in a sector that has strong current momentum.
We'll need to be more explicit on the biases. You're right they don't tell me much. The biases are clearly based on the forecast 10 year returns. At present we are negatively disposed to international generally. You can actually read through our views on the blog: http://www.kivalia.com/blog/post/2013/07/11/Investment-Outlo...
See if that helps.
Will take usability thoughts into account as well. Thanks!
A couple thoughts -
re: managed funds or what you're getting in a particular fund. Our style maps for each fund (shown in the bottom right of any fund detail screen) break down how a fund is behaving relative to the overall market, showing % weights in 30 different sectors. Sectors at present are only as granular as small cap growth, emerging markets, international value, etc. but it's a start. We'll make it more granular over time.
Re: Allocations The models will not adjust (ie. decrease beta) like target date funds over time. The various models are meant for personal risk tolerance more than time to retirement.
Low fees are clearly better than higher fees, but you probably won't be able to build a very diversified portfolio based on selecting investments based solely on fees. For example,international investments will always be more expensive than domestic, small caps will be pricier than large, etc.
If you're selecting lowest fee fund within specific classes, that will probably serve you relatively well...of course then you need to weight the fund classes. How do you handle that decision?
We're not shooting for bulletproof security, at this time, as there is no personal info here. If we were, we'd probably be hiding the model portfolios behind the firewall as well.
That said, we would like those who find the information of value to sign up so we can update them as the models change.
From my knowledge Brightscope rates the plans themselves based on expenses and other criteria.
Kivalia provides plan participants ongoing, actionable advice on how to invest the funds held in their individual accounts.
thanks for pointing that out, elindell...I've noticed that every once in awhile as well and we'll stamp that out.
Well, that's exactly what Kivalia does; helps you decide which funds to own at any point in time - and how to combine them. Of course you have to create the list at Kivalia, but then you're good to go - immediately and quarter after quarter.
That seems like a complete conflict of interest. Do you mean that they require you to have your account managed, for a fee, in order to get matching funds?
Thanks, Bradley. Tutorial explanation comment is noted - we have some ideas on that front.
There are a few themes that we're playing on with regards to the advice, and we'll try to express them more clearly on the site as we move forward. These are:
1) Most of the investment options available to participants are really a mix of a number of market sectors...for example, Vanguard Wellington (http://www.kivalia.com/assets/vwelx/vanguard-wellington-inv). The name tells me nothing, but if one evaluates it, it looks like the mix of assets shown at the bottom right of the page referenced above...a mix of stocks and bonds. So we do a good job of mapping what various investments look like.
2) Getting the overall allocation right is the most important aspect of investing...and explains maybe 80% or more of one's overall returns. How much do I want in large cap growth, small cap value, international, etc.? That's incredibly hard for an individual to do when you have a bunch of vaguely named funds. Of course it's much easier to do when one has a list of sector-based index funds to work with.
3) Philosophically we're indifferent between active and passive funds...they're each just combinations of asset classes with a fee tacked on. What's most important to us is the exposure to various sectors a particular fund provides you, after fees.
4) Yes, our algorithms take into account each fund's respective management fees; so all else being equal we'd be biased towards the lower cost (better performing) alternative.
5) Past performance is definitively NOT a predictor of future performance. We simply show results for transparency sake in an effort to allow users to see whether or not we're adding value to the process.
6) We do build in "tilts" to our advice based on expected returns over a substantially long time horizon. These tilts are intended to bias portfolios towards sectors that look more attractive at any point in time, and away from sectors that are less attractive in our view. That said, it is a simple process to tighten down our algorithms to force the recommendations to a more benchmark-y look and performance.
anandabits - that's the point of our service! It is specific advice customizable to your specific 401(k) plan. Check out the site to see what I'm talking about.
I've got to agree w/ your last point - it is typically best to move money into an IRA. There are cases however where a larger employer may be able to get lower cost funds than would be available to you in a stand-alone IRA.
What is most important is putting away enough money to maximize any company match you might get.
But you're right, the limited options can be problematic; and that's why we've created Kivalia.
Then you might benefit from the advice we provide that takes advantage of the "no commission" ETF lists at the leading brokers (Fidelity, Schwab, E*Trade, TD Ameritrade) to start.
Our advice is applicable in a number of ways.
Agree completely wj! Feel free to chat w/me if and when you'd like.
Brightscope rates plans based on cost and other attributes, while we provide actionable advice for participants in any plan. The actionable advice is in the form of model portfolios which are updated each quarter (with notifications sent to all subscribers to a particular plan).
Hi all,
"Edit: I'm the co-founder of Kivalia." We've spent about 1 1/2 years developing the above-linked community-based 401(k) advisory service offering, which is now live. We'd welcome feedback from the community to help us refine the design/flow/usability of the service offering.
The specific problem we're solving is "How should I invest my 401(k) money, given the market environment and the options available to me specifically?"
My partner and I have deep domain expertise (quant finance & investment advisory experience - 25 years or so between us) and as such, run the risk of creating something intuitive to us, but too geeky for mass consumption.
I’ll be active in the comments section
Thanks! B