Morningstar plows ahead with ratings in $2.1-trillion model portfolio market --thanks, UBS and Merrill -- but not with its coveted star system

If Morningstar as a company thinks evaluating model portfolios has to be more qualitative than evaluating funds, then they're doing it wrong! First, you don't just judge what someone sends in as a history of model portfolio returns. You judge a history you retrieve over time...ie) model provider signs up on 1/1 and you get ongoing changes to the models as they occur; say 3/28, then 4/15. If the model submitters can't provide you with real time updates, they aren't included in the ranking. Period. Then you judge the modelers on a number of characteristics - and maybe even bucket them into style camps. I see no reason why one has to have the entire fund (or model portfolio) universe in order to evaluate either...so the statement that - We've not moved to do star ratings for models because of the challenges with data quality. With models, you need to ask more questions," Kephart explains. "Every US equity fund that falls in the large mutual fund category is reported. We have the entire universe. We're never going to have that with a voluntarily reported database," he adds. - is just wrong. You don't need to ask ANY questions of the modelers; just let the model portfolios themselves suggest the quality! And you don't need the whole universe against which to evaluate the quality of a model portfolio strategy. So this all begs the question - what exactly is the agenda of Morningstar when it comes to evaluating model portfolios? Are they charging a fee for those submitting models (like they do for mutual fund companies themselves?)