anyone here doing valuations of privately-held/thinly traded/illiqiud assets for whatever purpose - financial reporting/tax compliance/marking to market/portfolio management - what methodologies do you use for quantifying these discounts?
I struggle with this, because it’s completely subjective. Depending on the company, I typically apply a 20-30% discount.
well i’ve seen such discounts ranging from 5% to 55%, with support from them ranging from “pull out of my a$s” to “applied an option-based model that gives exact answer”. somewhere in between is relying on emprical studies (restricted stock, pre-IPO) and picking something between 20-30% as you say. so some methods can appear to be very quantitative although arguably they are equally suspect. very diverse practices around, so that’s why i was polling opinions here
Traditionally, business valuation firms use data obtained from the public markets to quantify lack of marketability in a privately held entity. Each of these methods constitutes a proxy, rather than a direct comparison, for the lack of ready marketability associated with an interest in a closely held entity. As proxies, these methods inherently have both positive and negative elements associated with them. Some of these methods include, restricted stock studies, pre IPO studies, cost of flotation data, just to name a few. Our firm uses more of a comprehensive approach to quantify lack of marketability rather than just one or two methods which is typical of most valuation firms.
thommo77 Wrote: ------------------------------------------------------- > > Our firm uses more of a comprehensive approach to > quantify lack of marketability rather than just > one or two methods which is typical of most > valuation firms. elaborate
I cannot elaborate too much as our comprehensive LOM study is unique to our firm. All I will say is that most firms will use one or maybe two methods, lets say restricted stock studies and pre IPO data, to quantify lack of marketability. Our firm takes all the methodologies to quantify lack of marketability and comes up with weightings for each methodology.
ok, do you also use option-based methods or rely on empirical data only in your weightings?
We use both. The use of quantitative methods (option based models) as support for allowances for lack of marketability has been criticized for several reasons. Most importantly, quantitative methods have been challenged for their reliance on assumptions rather than hard capital market evidence.
thommo77 Wrote: ------------------------------------------------------- > We use both. The use of quantitative methods > (option based models) as support for allowances > for lack of marketability has been criticized for > several reasons. Most importantly, quantitative > methods have been challenged for their reliance on > assumptions rather than hard capital market > evidence. i am not saying they are perfect or the only solution, i was just wondering if you discard them completely or use them in addition to empirical studies. also i am not sure i agree with “reliance on assumptions rather than hard capital market evidence”. of course it relies on assumptions - it is a model. this looks like a general argument against any model to me. i can say the same about a discounted cash flow model over market approach/comparables.
We use them in addition to empiral studies. True, most criticism of DCFs is that they rely too much on assumptions. However, if the inputs can be measured with some element of certainty than an option based model can support an allowance for lack of marketability. We generally use an option based method to corroborate other methods. BTW, I am not criticising any method or any single solution, I am just stating what is out there in terms of opinions on these issues.
^empirical
ok last question - of the quant models, do you use protective put/longstaff/finnerty or all?
use the QMDM model developed by Chris Mercer. To be honest, we have not used this at our firm in a while so I am hazy on the details.
ah ok - so you use this DCF-based model, not option-based models
Yeah, I have not been exposed to the option based models as yet, but I know we have used them in the past.