# Get Active in Our Amazing Community of Over 450,000 Peers!

Schedule mock interviews on the Meeting Board, join the latest community discussions in our Consulting Q&A and find like-minded Case Partners to connect and practice with!

# Mckinsey real case

MBB McKinsey
New answer on Sep 07, 2019
3.2 k Views

Hi, Can anyone explain the solution of global pharm case they have in their website?

https://www.mckinsey.com/careers/interviewing/globapharm

I don't get how they came up with 40% increase. . .

Expected Probability of Success

Phase1: 70% > Phase2:40%>Phase3:50%>Phase4:90%

Global pharm will improve the product by investing 150M USD to Phase2. How much should the success rate have to be increased for investment to pay of. (Assume that if the drug is successful and sold its worth:1.2B USD)

Thanks

## Overview of answers

• Date ascending
• Date descending

Hi Anonymous,

Here's a slightly different approach to the calculation that might make more sense:

Currently, the probability of success for a candidate drug before Phase I trial is 70% * 40% * 50% * 90% = 12.6%

If a successful drug is worth \$1.2B, then the expected value of a candidate drug before Phase I trial is \$1.2B * 12.6% = \$151.2M. I would likely round this to \$150M in an interview.

In order for the \$150M additional investment to break even, we need the expected value to increase by the same amount. So the new expected value needs to be \$150M + \$150M = \$300M.

What increase would we need to see in the Phase II success rate in order to double the expected value from \$150M to \$300M? You could set this up as an equation, but it's pretty apparent just looking at it: it needs to double from 40% to 80%, an increase of 40 percentage points.

Once you have the answer, it's then important to comment on the implications: if we need the Phase II success rate to reach 80% just to break even, this is probably not a good investment.

-Matt

Seems to be a hard one! Now, I almost understood the logic vs the explanation on the site. Almost, except for "In order for the \$150M additional investment to break even, we need the expected value to increase by the same amount. So the new expected value needs to be \$150M + \$150M = \$300M." Can you please explain how do we know that?

Think of yourself as the person in charge of this decision: you can invest an additional \$150M to improve the success rate. By how much does the expected value need to increase for you to want to make that investment? Assuming you're risk-neutral, it should increase by at least \$150M also. For example, you wouldn't invest an additional \$150M if the expected value grew by only \$100M. Similarly, you would want to make the \$150M investment if the expected value grew by \$200M. Because we're trying to calculate the tipping point between those decisions (the "break even"), we calculate the success rate for which increase in investment = increase in expected value.

The solution presented on their website is not very intuitive for some people. Here is a different way of understanding the math.

We know that once a drug reaches market, it's worth \$1.2 billion. Using the success rates of the 4 different stages, we can calculate that the expected value of a candidate drug (before it passes any trial) equals to the product of \$1.2 billion and all those success rates, which equals to roughly \$150 million.

We also know that the investment required is \$150 million. Therefore, to justify this investment, the expected value of a candidate drug has to increase to \$300 million, essentially doubling the original value. This in turn means that the overall success rate has to double. Because the investment only affects Phase II trial, the success rate of Phase II trial has to double. Hence the 40% increase.

Hope this helps.

Seems to be a hard one! Now, I almost understood the logic vs the explanation on the site. Almost, except for "We also know that the investment required is \$150 million. Therefore, to justify this investment, the expected value of a candidate drug has to increase to \$300 million, essentially doubling the original value". Can you please expalin how we know this?

How likely are you to recommend us to a friend or fellow student?
0
1
2
3
4
5
6
7
8
9
10
0 = Not likely
10 = Very likely