Mannkind Valuation

[jpg]

May 29, 2014 21:12:23 GMT -5 bobw said:

jpg,

I also posted the same model on the other forum and have written a bit on the discount rate. Another way to think about the discount rate is to make the first year's discount rate 100% and the following 9 years make it 15%. If you do this the valuation is $15.97 instead of $18.48 per share which is not that different.

You can interpret the discounting this way: the 100% discount rate for the first year equates to a discount factor for one year of 0.5. So, an asset that costs $1 and will either pay $2 if successful or $0 if not with a 50% chance of being successful is fairly priced (except that you are not being paid for the taking the risk). If you discount back the $2 with the 100% discount rate, you will get the current valuation of $1. If you think that there is a 50% chance of success for Mannkind (FDA approval, partner, commercialization), then a 100% discount rate for the first year may be appropriate.

Since the market has returned 11% on average over the last century, maybe a 15% discount rate after the major risks are past is also reasonable.

If the valuation does not include all of Mannkind's assets, makes what you believe to be conservative assumptions and produces a valuation that is above the current price, then its price is cheap and you might want to buy. Leaving some of the difficult to value assets out of the model provides a margin of safety for the investment decision.

As for Deerfield lending at a lower rate, we must keep in mind that they also received the right to convert debt to equity at a favorable price which increases the effective rate that we paid for their financing. They also hold first lien senior secured debt, so they are paid first in a liquidation and are taking less risk than equity. I think if you took the time to value the embedded options that they received that the financing was not that cheap.

You are certainly correct that no model is precise, but it does give you a framework to access the risk/reward aspect under the conditions specified in the model. If you understand the strengths and weaknesses of the model you should gain some insight into the viability of the investment. I can't guarantee that my assumptions are reasonable and I don't expect the future to resemble what I have modeled, but it does allow me to make a more informed decision than I would without the model.

Thank you for the detailed explanation.

Changing discount rates as you do in different models (100 then 15 vs 25 then later 20) and ignoring value of other assets certainly is a valid way of coming up with a conservative value estimate and provides one big margin of safety! I did not understand that this was what you were trying to do. Both models you present are mathematically almost the same as obviously they come out to almost the same results... Obviously these types of mathematical models are statistically valid only when dealing with multiple similar events and not for 'one offs'. Kind of back to the 'spherical cow in a vacuum' problem.

Deerfield lends at a much much lower rate then you estimate risk for because, as you point out, they have an option to convert to equity. Keep in mind this equity conversion is worthless if things don't go well and the underlying assets don't have that much value if things go badly. It is mathematically risky to split this investment into 2 parts as if one was not related to the other. The risk and reward is global but depends on binary events that seem to carry much less risk then either of your models suggest. All this to say, I and many others including Deerfield I suspect, would not have invested if the probabilistic outcomes your model tries to mathematically demonstrate were representatif of the risks Mannkind faces going forward. Again if you label your analysis as being a extremely conservative value projection to guarantee a significant margin of safety then I understand.

Now that you have calculated your 'deep conservative margin of safety' analysis what would be your 'realistic optimistic' value analysis of Mannkind taking into consideration the value of all their assets?

JPG

[bobw]

Trend,

Here is a copy of a post that I made in 2012 on Agoracom. It is a little old, but it shows the share count up to that time. Since then there were the Deerfield conversions and possible future conversions, the ATM and possible future ATMs. I have not take the time to update the count since then, but I think we are getting close to the 500 million mark by now.


Here is my assessment of the expected number of shares based on what we already know and taking into account the entire capital structure of the company as of November 2012

We are currently up to 285,767,463 shares outstanding. 199,767,463 previously issued plus 86,000,000 issued on 10/23/2012. The issuance on 10/23/2012 was over subscribed at 46 million to the public, not 40 million.

I would also add the 3,002,778 shares issuable in connection with the 7/23/12 securities action settlement, but not the additional 1,000,000 shares if the share price is below $1 at settlement.

I think it would be safe to add the shares from the exercise of all of the warrants; an additional 86,062,500 shares. If the company is successful, all warrants will get exercised. If the share price is not high enough for the October 30, 2013 warrants to get exercised at $2.60, then the company will have to go to the capital markets for financing and the terms will surely be worse than the exercise would provide.

There were 3,782,122 unsettled restricted shares which will likely settle if the company is successful.

There are also 18,662,846 company options with an average strike of $3.47. Since I don't know the breakdown of all the strikes, it is hard to know how many will get exercised, but for safety, I would assume that they all get exercised

The 8/15/2015 convertible debt has a conversion price of $6.80, so if the company is successful, then by 2015 the share price should be above $6.8, so I assume conversion will happen.

I would not count the 12/15/2013 convertibles since the conversion price is so high, nor would I count the restricted shares that are available but not yet granted.

Summary of total number of shares in order of likely occurrence
From_____________________________Shares____Running Total
Outstanding___________________285,767,463____285,767,463
Other: from securities action___3,002,778____288,770,241
Warrants_______________________86,062,500____374,832,741
Restricted Stock________________3,782,122____378,614,863
Company Options________________18,662,846____397,277,709
8/15/2015 Convertible Debt_____14,708,590____411,986,299

So, by my count, we are up to about 412 million shares.


Here are the details: Unfortunately, I don't know how to make everything line up in columns.

Convertible Debt
Maturity Date Principal Coupon Conversion Price Total Shares if Converted
12/15/2013 115,000,000 3.75% 22.47 5,117,523 Up to 1,484,064 additional shares issuable make-whole premium
8/15/2015 100,000,000 5.75% 6.80 14,708,590 Up to 2,041820 additional shares issuable make-whole premium
Total 215,000,000 19,826,113

Mann Group LOC
Maturity Date Principal Owed Unused Principal Total Credit Line As of Date
7/1/2013 ?? 120,000,000 ?? 9/30/2012


Warrants
Expiration Date Quantity Strike Price Shares Per Warrant Total Shares if Exercised Total Proceeds if Exercised
2/8/2016 35,937,500 2.40 0.60 21,562,500 51,750,000
10/30/2013 46,000,000 2.60 0.75 34,500,000 89,700,000
10/30/2013 40,000,000 2.60 0.75 30,000,000 78,000,000
Total 121,937,500 86,062,500 219,450,000

Shares Outstanding
Date Shares Issue Price Proceeds
As of 9/30/2012 199,767,463
10/23/2012 46,000,000 2.00 92,000,000
10/23/2012 40,000,000 2.69 107,400,000
Total 285,767,463


Stock Options
Weighted Average Strike Price Total Shares if Exercised Total Proceeds if Exercised
As of Oct 2012 3.47 18,662,846 64,760,076


Restricted Stock Units
Total Shares upon settlement
As of Oct 2012 3,782,122


Other Stock
Date Shares
As of Oct 2012 9,338,040 shares available for future grant under 2004 equity incentive paln and option plan
As of Oct 2012 3,002,778 shares issuable in connection with 7/23/12 securities action settlement
As of Oct 2012 1,000,000 shares issuable in connection with 7/23/12 securities action settlement if share pr <1.00 at settlement
Total 13,340,818

[bobw]

To fix the formatting and make the columns line up properly, I am posting Word file that contains my last post.

[BD]

Well, because I apparently have too much time on my hands, I converted that Word file into a PDF for those who may not have Word (note that Word 2010 has a "Save as PDF" feature that I've found very handy!)

[coco]

Thanks for posting this . I could not view it before. Very interesting and insightful. You're awesome!
Coco.

[bobw]

silentbob,

I have been a little busy lately and wanted to consider your assumptions before responding. Thanks for taking the time to post them.

You used different revenue per patient assumptions than I did. I was curious how you came to these assumptions. My $2,000 annual assumption for all type 1 and 2 patients is based on what I thought the current price of injected RAA insulin is and that at most we will get a 10% premium over that. Why do you have a higher per patient revenue estimate for type 2 insulin dependent patients than type 1 patients? I assume that you have a lower estimate for insulin naive type 2 patients because they may not need supplemental insulin at every meal.



I noticed that you changed the number of pre diabetes patients from 80 million to 8 million. I got my estimate from the Centers for Disease Control website. They estimate 79 million in 2008, so it is probably greater than 80 million today; I am not sure by how much, so I just used 80 million.

"In 2005–2008, based on fasting glucose or A1c levels, 35% of U.S. adults aged 20 years or older had prediabetes (50% of those ••aged 65 years or older). Applying this percentage to the entire U.S. population in 2010 yields an estimated 79 million Americans aged 20 years or older with prediabetes."

Source: www.cdc.gov/diabetes/pubs/pdf/ndfs_2011.pdf


You have somewhat different ramp up speed projections from mine. My projections are a guess, so I have no comment on your modifications.

You changed the profit margin from 25% to 20%, which is 20% more conservative than my estimate. I won't argue with that assumption, but I am still hoping that it is closer to 25%.

Now for the discounting. I really feel that the discount rate cannot go below 10-15% ever, and I think it should be much closer to the high side of that range.

If you believe that the value of Mannkind will double after FDA approval and a partnership announcement, then you are sort of saying that you think the market is pricing in a 100% discount rate for the first year.

You can make a case that in low interest rate environment that the discount rate can be lower, but it gives me more comfort to use higher rates. I feel that my assumption that Mannkind can get 25% of the potential market is optimistic and want to offset this assumption with what I feel is a discount rate that accounts for the risk that this will not happen.

It is true that Mannkind has patents that go well past ten years, but there is uncertainty as to the value of those patents that far out. Other technologies could be developed in ten years that could displace Mannkind's patents.

I am not trying to estimate value, not price and I really do want to make a conservative valuation so that I have a margin of safety.



JPG,

Instead of trying to make an optimistic valuation as of today, you could project the value in the future by assuming that the risks that are modeled into the discount rates do not materialize. If you go back to using 100% for the first year and 15% for the final 9 years, you get a valuation as of today. If you set the first year's discount rate to something like 3% you are projecting the value in one year assuming FDA approval, partnership and the successful start to commercialization and sales. Setting the second year's discount to 3% will give you a valuation in two years, etc.

I really do not know how to value the Technosphere platform. It could end up being worth more than Afrezza if enough new drugs are developed using that technology, but there is more risk and less certain time frames when developing new drugs. Valuing new drugs with so many unknown risks requires discounting with high rates.

[babaoriley]

Relative to patent length, MNKD will not exactly be doing nothing from an R&D perspective, so they may well be able to get a new patent on a better product, making the version originally approved somewhat obsolete. Beyond that, though, if Afrezza works as well as the people who testified at the AdCom thinks it does, and if the FDA obliges us and let's us use it, and use it on more than people with six fingers and toes, that darn thing is going to sell like hotcakes. More than even pet rocks!

[biotec]

IDK about the more then the pet rock baba, I still have my pet rock.It sold a whole lotta rocks.Do you have a collection?

[silentbob]

Jun 1, 2014 14:29:39 GMT -5 bobw said:

You used different revenue per patient assumptions than I did. I was curious how you came to these assumptions. My $2,000 annual assumption for all type 1 and 2 patients is based on what I thought the current price of injected RAA insulin is and that at most we will get a 10% premium over that. Why do you have a higher per patient revenue estimate for type 2 insulin dependent patients than type 1 patients? I assume that you have a lower estimate for insulin naive type 2 patients because they may not need supplemental insulin at every meal.

I noticed that you changed the number of pre diabetes patients from 80 million to 8 million. I got my estimate from the Centers for Disease Control website. They estimate 79 million in 2008, so it is probably greater than 80 million today; I am not sure by how much, so I just used 80 million.

"In 2005–2008, based on fasting glucose or A1c levels, 35% of U.S. adults aged 20 years or older had prediabetes (50% of those ••aged 65 years or older). Applying this percentage to the entire U.S. population in 2010 yields an estimated 79 million Americans aged 20 years or older with prediabetes."

Source: www.cdc.gov/diabetes/pubs/pdf/ndfs_2011.pdf


You have somewhat different ramp up speed projections from mine. My projections are a guess, so I have no comment on your modifications.

You changed the profit margin from 25% to 20%, which is 20% more conservative than my estimate. I won't argue with that assumption, but I am still hoping that it is closer to 25%.

Now for the discounting. I really feel that the discount rate cannot go below 10-15% ever, and I think it should be much closer to the high side of that range.

If you believe that the value of Mannkind will double after FDA approval and a partnership announcement, then you are sort of saying that you think the market is pricing in a 100% discount rate for the first year.

You can make a case that in low interest rate environment that the discount rate can be lower, but it gives me more comfort to use higher rates. I feel that my assumption that Mannkind can get 25% of the potential market is optimistic and want to offset this assumption with what I feel is a discount rate that accounts for the risk that this will not happen.

It is true that Mannkind has patents that go well past ten years, but there is uncertainty as to the value of those patents that far out. Other technologies could be developed in ten years that could displace Mannkind's patents.

I am not trying to estimate value, not price and I really do want to make a conservative valuation so that I have a margin of safety.

Bobw,

<You used different revenue per patient assumptions than I did. I was curious how you came to these assumptions.>

I am really just guesstimating. As you probably know, Type1's are not insulin resistant so they probably need much smaller doses then insulin dependent Type2's, who very likely have advanced disease. All Type2's have insulin resistance, but in the early stages it is more mild and the pancreas does still put out significant amounts of insulin, so I modeled a lower dosage for early stage Type2's.
Of course I am also assuming higher dose cartridges will command a higher price then lower dose cartridges.

<I noticed that you changed the number of pre diabetes patients from 80 million to 8 million.>

I do not doubt the number of actual estimated prediabetics, however we are talking about diagnosed prediabetics only. Very few people that are prediabetic actually know it. Those that are not aware of their condition will not take medications for it.

< I really feel that the discount rate cannot go below 10-15% ever, and I think it should be much closer to the high side of that range.>

That's a matter of preference. In my own view the stock market is overvalued in general, and many other asset classes have much devaluation risk. Therefore considering risks are high across the board a lower discount rate seems acceptable to me, even though it is less conventional. I can see why you would want to model 10%, but 15-25% for later years is excessive in my view.

< Other technologies could be developed in ten years that could displace Mannkind's patents.>

While this is true, they would still need to go through the approval process.

It is certainly not wrong to estimate on the conservative side, especially in biotech.
But you should still be aware that you are compounding several conservative assumptions which affect the final valuation significantly...

[jpg]

But you should still be aware that you are compounding several conservative assumptions which affect the final valuation significantly...

[/quote]

I think this expresses my take on the above model well although the 'spherical cow in a vacuum' analogy is more colorful...


JPG

[papihoyos]

Great job bobw on your model. It was well thought out.

My question is regarding the assumptions, specifically the revenue per patient per year. Matt spoke today about the capacity of the plant. The current plant will generate 2 billion units per year or $1.5 Billion in revenue at full capacity (He actually said two lines will generate $375M, the two line only represent 25% capacity and the plants capacity is 2B units). A unit therefore is equivalent to $.75 per unit. Assuming 3 units per day, a patient will generate $821/year in revenue. Even if you double it, the amount would be $1,643 per patient per year. You use $2,000. Am I missing something?

They will need many more plant to meet the expected demand. Don’t we need to consider the costs of new plants in future cash flows?

In terms of valuation, is there a rule of thumb you can apply as a multiple to stabilize revenues to determine value? i.e. 7x, 10x, etc

Lastly, are you assuming in the 25% margin after the marketing partner’s share?

Thanks for sharing

[babaoriley]

Thanks for posting that, papihoyos, I see you started posting just in time to make this comment. I suggest you listen to Matt's talk, rather than rely on the transcript. Things get mixed up when transcribing sometimes. For .75, you might get a pack of gum, or half hour on a parking meter, but you won't get a dose of Afrezza, unless, perhaps I'm missing something. There is also another thread on this site that might help you out, where it's suggested that there was no "$" before the "375,000,000." Makes the whole thing a horse of a different color, sometimes it's so hard to tell exactly what color a horse is, don't you agree?

You're not in the medical field, are you? Just curious, cuz I'd like to cut my losses right now.

[liane]

Regardless of what was said (misspoke or poorly transcribed), my understanding has always been that full capacity is 12 lines ( 2B cartridges, 2M patients). I always thought they would launch with 3 lines (= 500M cartridges, 500,000 pts). The number they are now throwing out sounds like 2 lines (375M cartridges, 375,000 pts). The revenue has long been discussed as being ~ $2000 per pt.

[ashiwi]

The numbers I was using were:
$2/ cartridge. 3 cartridges/ day = $6/ day X 365 days = $2190 per year per patient.
375 million cartridges per year would bring in just under $800 million in revenue.
At full buildout 4 billion cartridges would generate $8 billion in revenue
Please correct me if I am wrong.

[brentie]

Ashiwi, I believe it's 2 billion cartridges at full buildout.

"So turning a little bit more towards commercialization. Obviously with the kind of market we are talking about, it's important to address where we are going to make this stuff and we happily have a very nice data facility located in Danbury, Connecticut that has been our commercial facility for some time.



We didn’t fully build it out because it's quite expensive to do so pre-approval, but the facility is scaled to do up to two billion cartridges per year fully built out and by that, we're really talking about filing the rooms with the equipment that they are sitting there waiting to have installed.



So the primary bottleneck facility has mostly the fill finish machines. You can probably imaging two billion cartridges is a lot of cartridges and you need a lot of equipment to fill those, almost no matter how fast you do it.



At launch, we expect to have three of those finish lines in place, which should give us a $375 million per year capacity, that's a quarter of the total capacity of the facility to some of the layer once it have a little higher than if you are doing the math and it doesn’t quite work.



You have to trust me on this, but it can go up to two billion cartridges a year at full capacity and we can add the capacity in a very nice relatively rapid modular way, which is one of the advantages of having built this very nice expensive facility with all the clean rooms and things in place, just waiting to drop the machines in.
"