|
|
Post by silentbob on Apr 7, 2015 9:29:36 GMT -5
About a year ago, before approval, bobw made spreadsheet model to value Afrezza, discounting your own projections of future market share/income to today's dollars based on a discount rate you can choose yourself. I updated the spreadsheet with new diabetes numbers, a longer time-frame (the original only counted up to 2024 now 2040) and I put in my own numbers. The defaults seem attainable to me, but opinions on that may vary. The default numbers are only a starting point though. You should put in your own numbers and the spreadsheet will calculate what Afrezza is currently worth based on that. The calculation method has its limitations but I think it is a very good estimation tool. Note: - All green fields can be modified. Put your own numbers in there! For example: the initial value for T1 market share is 20%. If you think peak market share will be higher/lower then update this number. - Make sure to scroll to the right & bottom: the sheet is pretty big. - Important fields to look at are market share (for each type), revenue per patient, profit margin, discount rate, growth rate and future share count. - For the more adventurous you can look at ramp up percentages as well. By default market share peaks only by 2024 for most markets, which is fairly conservative. - Worldwide revenues are just calculated based on USA sales multiplied by a changeable factor (default 2.0). This is not obviously perfect since ROW will likely lag by ~2 years but 2.0 should be more then conservative enough to compensate for that. Remember that the original spreadsheet was created by bobw, so props to him! Please input your numbers and share your results!See attached : Attachment Deleted |
|
|
|
Post by silentbob on Apr 7, 2015 9:43:27 GMT -5
By the way, if you don't have excell, download openoffice, which is free and very good! www.openoffice.org/ |
|
|
|
Post by bobw on Apr 7, 2015 10:55:45 GMT -5
It funny that you posted this today. I just finished updating that spreadsheet to reflect the partnership and some updated assumptions. I will post it sometime in the next few days.
Bob
|
|
|
|
Post by gomnkd on Apr 7, 2015 20:32:32 GMT -5
As per 10-K by 2032, longest living patents expire. Many expire before that. If Afrezza is really successful, you may see some competitor's product approved by end of this decade.
I may have missed it, are you considering the fact that Sanofi owns 2/3rd economic ownership of Afrezza?
If I throw out income after 2030 and consider 1/3rd, I get today's value at $6/sh.
|
|
Deleted
Deleted Member
Posts: 0
|
Post by Deleted on Apr 7, 2015 21:04:03 GMT -5
As per 10-K by 2032, longest living patents expire. Many expire before that. If Afrezza is really successful, you may see some competitor's product approved by end of this decade. I may have missed it, are you considering the fact that Sanofi owns 2/3rd economic ownership of Afrezza? If I throw out income after 2030 and consider 1/3rd, I get today's value at $6/sh. So remind me why you're long again? Are you related to PA? |
|
|
|
Post by silentbob on Apr 8, 2015 11:05:51 GMT -5
As per 10-K by 2032, longest living patents expire. Many expire before that. If Afrezza is really successful, you may see some competitor's product approved by end of this decade. I may have missed it, are you considering the fact that Sanofi owns 2/3rd economic ownership of Afrezza? If I throw out income after 2030 and consider 1/3rd, I get today's value at $6/sh. Patents only last for 20 years, so it makes a lot of sense that the current patent moat doesn't extent beyond 2032. Mannkind is constantly adding patents, both on current methods and future ones. I'm sure that by 2030 Afrezza as it is now will not even be on the market, there may be multiple improved versions by then. I personally think it's more like a patent minefield. It will be very difficult for a competitor to come up with a biosimilar that can actually be mass produced without stepping on any of the powder/inhaler/production patent 'mines'. But even with a biosimilar on the market, margins and revenues will never be zero, as the investment required to apply and mass produce a biosimilar will be quite large in the case of Afrezza. That said, if you want to consider anything beyond 2030 as 'free upside' that is a valid preference. As for this : "I may have missed it, are you considering the fact that Sanofi owns 2/3rd economic ownership of Afrezza?" Are you joking? You assume I missed the most important marker in any valuation scenario? You must not think very highly of me. Maybe you missed the 'Profit Margin' field, a green space that you can adjust to your liking and is even referenced in the above post. By default it is at 18% which would mean the partnership entity sinks almost 50% of revenues into costs like COGS, sales, marketing, REMS, ... I think that's reasonable. Even if I set the profit margin at a very conservative 12% (which is only half of Matt's mid-20's royalty equivalent), and I assume zero income after 2030 but keep everything else the same then the model outputs Afrezza NPV of $12. Anyway at least you opened it and tried it, which is more then 99% of the board, so thanks... |
|
|
|
Post by gomnkd on Apr 8, 2015 12:46:49 GMT -5
silentbobThanks for the clarification. Fact is, I do think highly of you. If not, I wont be replying to your posts. I always love dialogs about valuations as it challenges my thinking and keeps me grounded. I got confused by column named "Potential MNKD revenue" , I think I should be reading it as partnership revenue. Since you used bobw's old xls, I assumed everything is for partnership.
|
|
|
|
Post by bobw on Apr 8, 2015 14:30:41 GMT -5
Dear Fellow Stockholders,
Here is my updated model that reflects the Mannkind profit sharing agreement with Sanofi. I also structured it so that the breakdown of patients, revenue and income is more transparent and straight forward. Note that all green cells can be modified.
In columns A through D, I entered what I thought were reasonable assumptions. It assumes that the diabetes growth rate is 5%, which is based on past data from the CDC that show a higher historical growth rate, see columns AA through AD. I assume a 60% profit margin for the partnership, which when multiplied by the 35% profit share gives a 21% revenue share.
I further assume that Mannkinds profit margin is 100% of the profit share. I justify this by not including the additional $700+ million in milestone payments that Mannkind should receive if successful. I am also not giving credit to the over $2 billion in net operating loss. Finally, no credit is given to the Technosphere platform. Since I am not valuing Technosphere, then I should not include the expenses of developing it. I do believe that this platform could potentially be very valuable.
I am using 480 million shares to account for warrants, company options, etc.
I am assuming an 18% discount rate to compensate for the risk of owning equity. 2015 has a 13% discount rate because over 3 months have passed.
Rather than try to estimate market share and adoption rate, in columns G through S, I put in assumptions of market share and the adoption rate that would make the model come close to the current share price. You can see from this simple model of Afrezza sales, that Afrezza needs to capture only 15% of the T1 patients, 15% of the T2 insulin using patients, and 10% of the T2 patients using oral medications only. Note that I have used a fairly slow ramp up assumption. I further reduced the income by assuming that starting in 2025, the ramp up declines and goes to 0 by 2028. The annual wholesale revenue per patient is assumed to be $2,000.
If you are comfortable with the low adoption rate assumptions that I have made and not including the value of the Technosphere platform, then Mankind stock is a buy at $5.47.
If you make less conservative ramp up, market share and Technosphere assumptions , then the present value of Mannkind is higher.
My feeling is that the fully ramped up market share will be north of 25% for the three subsets that I have modeled and that there is potential revenue from the pre-diabetic population as well. Setting the three subsets to a 25% market share, yields a risk adjusted share price of $11.48. Add to this the residual value of Mannkind Corp in 2028 and the value of the Technosphere platform.
Thanks in advance for any feedback, Bob
|
|
|
|
Post by otherottawaguy on Apr 8, 2015 16:10:25 GMT -5
Bob:
Took a quick spin through the sheet and just wanted to clarify that the 5.47 is the discounted revenue per share going out to 2028 and not share price? Would you want to venture as to what would be a fair P/E for a mature biotech?
Your sheet calc of the max market being 6.5M user at its height is also interesting as I have seen another number in the 6M user ballpark before.
Thank for the efforts,
OOG
|
|
|
|
Post by bobw on Apr 8, 2015 20:40:25 GMT -5
OOG,
Thanks for your reply.
The 5.47 is the discounted income per share. I assume that all the expenses are borne by the partnership and that the additional expenses that Mannkind incurs are offset by the remaining milestone payments and the benefit from utilizing the NOL. I think this a very conservative assumption. This assumption, which allows Mannkind to have a 100% margin, makes the 35% income share from the partnership = revenue to Mannkind = income to Mannkind. The $5.47 is the price that I would say is a buy recommendation, if you believe, that the ramp up and final market share assumptions in the spreadsheet are appropriate.
Remember, I entered the assumptions that would value the Mannkind income equal to the current share price. This was to see what it would take to justify the current price. If everything works out according to the assumptions in the spreadsheet, the share price in a few years will be much higher than it is now. If sales are higher then spreadsheet assumes, then the price will be higher still.
I believe that sales will exceed the assumptions in the spreadsheet and that there will be significant value from the Technosphere platform. If Afrezza is successful, the sales could be much higher than shown in the spreadsheet. My intent is not to try to predict the price, but rather to see if I am comfortable holding Mannkind as an investment.
I don't think it is appropriate, when valuing a company like Mannkind, to try to apply an earning multiple. The distribution of future earning projections is too wide to make a meaningful price estimate based on an earnings multiple.
I am not sure where you see the 6.5M patients.
Bob
|
|
|
|
Post by silentbob on Apr 9, 2015 4:53:21 GMT -5
silentbobThanks for the clarification. Fact is, I do think highly of you. If not, I wont be replying to your posts. I always love dialogs about valuations as it challenges my thinking and keeps me grounded. I got confused by column named "Potential MNKD revenue" , I think I should be reading it as partnership revenue. Since you used bobw's old xls, I assumed everything is for partnership. You have a point, I looked at the column headers and they are confusing. I should have updated them all. The last column is the one with the discounting and the profit margin built in, the one before that is the profit margin (by default 18% to MNKD) without discounting, and the one before that is revenues for the whole partnership (including SNY 65%) |
|
|
|
Post by silentbob on Apr 9, 2015 5:26:57 GMT -5
bobwCheck out the updated numbers on diabetes in the sheet I added in my original post. They are straight from ADA/CDC from 2014, I added the URL's. It looks like you're expecting generics starting from 2025, from the way you reduce market share at that point? Why do you feel this is the case? The original formulation/inhaler never got approved, so any generic would have to be proven biosimilar to the new formulation & dreamboat. Those patents extend a lot further. BTW your worldwide market cap formula doesn't account for your added ROW discount (=D5*$D$23 would be better)
|
|
|
|
Post by bobw on Apr 10, 2015 8:28:26 GMT -5
Thanks for your update silentbob. I forgot to update the market cap formula for the lag in the European approval; thanks for catching that. I will update with the latest CDC data on the weekend, when I have more time. The new numbers may not be that different, since I am growing the 2010 numbers at a 5% annual rate. I am reducing Afrezza's market share starting in 2025, just to be conservative. There could be generic competition or some totally new development in diabetes treatment. I agree that generics will probably take a little longer to impact market share and the same could be said for new treatments. One thing to consider, when discounting at an 18% rate, the income that is long dated becomes worth less and less pretty quickly.
I think the point that I am trying to make is that even with some conservative assumptions, Afrezza does not have to capture a vary large portion of the market to be successful.
|
|
|
|
Post by silentbob on Apr 10, 2015 9:47:37 GMT -5
bobw , "The new numbers may not be that different, since I am growing the 2010 numbers at a 5% annual rate." Oops - I just checked and I didn't remove this from the sheet I posted in the original post. This is a post that blows up valuation by 20% - embarrassing! "I am reducing Afrezza's market share starting in 2025, just to be conservative." Personally I feel the compounded discount rate is high enough by that point to include risk of generics and competitors, even more so when modeling smaller % of the market. "One thing to consider, when discounting at an 18% rate, the income that is long dated becomes worth less and less pretty quickly." Even at 90% discounted you will find later years still have a lot to contribute to NPV, assuming you don't 'ramp down' quickly like you do. I'm attaching the fixed sheet, where I also increased my discounting a little. Mods, and way I can update the attachment in the original post?
|
|
|
|
Post by liane on Apr 10, 2015 10:06:45 GMT -5
silentbob,
I swapped out the old attachment and inserted the new. (Hope I didn't screw it up).
|
|
