Recently, I’ve been asked a few questions about my investing approach and a specific medtech situation. The questions are:
1. Is Medtronic a value play? Given my medical background, the reader was interested to know if I could bring any insights to the discussion over and above what a generalist would. Specifically, he asked if I could add anything via the scuttlebutt approach made famous through Philip Fisher, and popularised in his book - “Common Stocks and Uncommon Profits.”
2. How do I apply a margin of safety, and what do I think of using discounted cash-flow (DCF) in stock valuation, given none of us can predict the future?
3. As someone who works in the medtech industry, could I apply value investing principles in medtech and healthcare businesses?
I’ve decided to group these three questions into two posts. This post, “Part 1. The Numbers” will look at the numbers, valuation and margin of safety. “Part 2. Scuttlebutt“ will see if I can tease out any additional insights from working in the medtech industry and use of the scuttlebutt approach. By using Medtronic as a case example, this may help to illustrate some of my thinking.
First, then, to Medtronic – Is it a value play?
Medtronic (MDT) is a large cap medical devices business. The company manufactures and sells device-based medical therapies across 120 countries. According to the 10-K, the operating segments are:
- Cardiac and Vascular group - Cardiac Rhythm Disease Management (CRDM) and CardioVascular.
- Restorative Therapies Group – Spinal, Neuromodulation, Diabetes and Surgical Technologies
At the time of writing, revenues are $16.59 billion with net profit of $3.47 billion.
When looking at a potential investment situation, the first things I look at are a few ratios – valuation and economic moat. These are as follows:
- Valuation: P/E = 15.37x (trailing twelve months – ttm); forward P/E = 12.58x; EV/EBITDA = 9.18x
- Moat: ROE = 19.38%
I generally look for situations where the P/E (ttm) <15x, forward P/E <12x, low EV/EBITDA (<7x is good, accept up to 10x) and ROE >15%. These are part of a screening approach I use to narrow down potential opportunities for further research. On P/E measures, MDT is about OK but higher end of acceptable on EV/EBITDA. The ROE is attractive.
I look at the history of revenues, earnings and ratios over time (for 10 year historical values I use Gurufocus). I want to know two things – how do the ratios compare to historical values, and is there a history of growing revenues and earnings over a period of several years.
I’m not accomplished enough as an investor to be able to spot turnaround situations before the occur, or predict when a company with unpredictable earnings will be going into a positive cycle. I know some super-investors are able to do this but I prefer to stay with a system that helps me to follow Warren Buffett’s rule nos 1 and 2 of investing:
“1. Never lose money; 2. Never forget rule no.1″
On the revenues and earnings growth measures, MDT actually performs quite well. Revenues have grown year on year from $9.10 billion in 2004 to $16.59 billion in 2013, giving a CAGR of 5.9%. Net income has grown from $1.96 billion to $3.46 billion in 2013, with some small volatility but an upwards trend (CAGR = 5.8%). Not bad.
Valuation ratios have historically been higher – between 2004-2010, P/E ranged from a high of 35x to a low of 15.6x. However, the ratio has been below 15x in 2011, 2012, 2013, with a low of 11x in 2012.
The ROE over time has also been consistently strong – ranging from a low of 16.9% in 2009 to a high of 27.1% in 2006. There does appear to be an economic moat (more on this in the next post).
I also like companies with low or no debt. MDT’s Debt to Equity (ttm) = 0.57 and ranges from a low of 0.26 to 0.84 over the last 10 years. Interest cover is 28x (as measured by EBIT/ interest expense). Thus, it seems to be conservatively run in terms of debt.
As a sanity check, I also have a quick look at peer group valuation (though this is less important in my book – I’m more interested in absolute rather than relative valuations). This might highlight other opportunities in the sector, or give me a feel if a sector is generally over or undervalued.
How does MDT compare to its peers in terms of valuation. It’s quite low – St Jude Medical P/E(ttm) = 17.70x; Edwards Life Sciences P/E(ttm) = 21.23x; Johnson & Johnson P/E(ttm) = 23.0x, Stryker P/E(ttm) = 20.7x, Intuitive Surgical P/E(ttm) = 30.13x
So far then, we have a company that has attractive growth history, an apparent economic moat, and a fair valuation based on ratios. Its seems to be a company I might like to own, but what would be my buying price?
Margin of safety
How do I decide what is a fair valuation and hence derive a discounted value, giving me a margin of safety? This is where I may differ from conventional thinking by many value investors. I don’t usually go for a discount to book value (though, as a sanity check, I don’t like situations where P/B >3x; MDT P/B = 2.8x).
Basically, I set a fairly low hurdle – can this business earn me a compounded growth rate of 15% in the next few years (i.e. a return on my capital of 15%). I look at historical growth rate for net income, projected growth rate (with a heavier leaning towards the former) and project out the net income over up to 5 years. Then I use a moderate valuation multiple (usually around the 15x mark), and work out if it could double in price in 5 years.
Let’s look at this for MDT. Current price is $51.87. Historical growth rate for earnings is 5.8%, projected EPS growth rate is pretty close at 6.9% (from Yahoo! Finance), so I’ll use the lower figure. EPS (ttm) =3.37x. Growing this at 5.8% over 5 years, and applying a 15x PE multiple gives us a valuation of:
(1.058^5) * 3.37 * 15 = $67.
At a current value this would give 5 year growth of 29%, or CAGR of 5%. Hmmm. For MDT to be attractive for me to buy, it would have to have a share price < $33.5 (i.e. $67/2).
Addressing, one of the earlier questions, what about using a discounted cash flow (DCF) model? Well, this is somewhere I really struggle. I get the concept and given future cash flows, growth rates and discount rates, I can work out DCF valuations. But as Seth Klarman says, value investing is part art, part science – it doesn’t take much to learn the technicalities. I takes a lot to turn it into something effective and actionable in the real world.
But I don’t have confidence (or delusion) in my ability to project out a company’s earnings over 10 years. Management don’t even know what their companies will be doing 10 years out. Armies of analysts certainly don’t. I’m aware Buffett says if you don’t know what a company will look like in 10 years, you shouldn’t even consider owning it for 10 minutes, and he also refers to DCF as being the valuation yardstick. Yet, his long-time partner, Charlie Munger, says he’s never seen Buffett perform a DCF calculation. (Is Buffett so bright he does them in his head? Actually, by all accounts that actually may be possible).
Where I do find DCF is useful, is as another sanity check. If you use the historic earnings growth rate, compound for a few years (I use 5 years) then have a lower terminal valuation (say for 10-15 years), it can be a valuable comparator to my “can it double in 5 years?” measure.
If I do a crude DCF for MDT, using 6% CAGR for 5 years, 5% CAGR for 15 years (terminal value) and a 10-12% discount rate, the DCF valuation ranges from $38-$44 (variation due to different discount rate). If I want a margin of safety, buying at say 70% of fair value, this gives me a buy price ranging from $27-$31. Not too far from my buy price (I try to use fairly conservative assumptions in any DCFs).
Importantly, when performing any valuation exercise, the goal is not to be precise. The goal is: if you use conservative assumptions, and a situation still looks good value, then some good things may happen to you as an investor.
It’s a bit of “True wisdom is knowing what you don’t know,” in the words of Confucius. Or as John Maynard Keynes said – “It’s better to be roughly right than precisely wrong.”
As it happens, I did buy MDT, using the same broad rationale in August 2011 around $32. I subsequently sold around $42 in Oct 2012 as I felt it was getting closer to fair value. To be honest, I also got a little nervous and sold too early (it hadn’t even yet reached my estimate of fair value, but I had made some 30% gain over just a year, considerably more than my 15% target).
One big lesson I learned from this is that if an undervalued situation starts to pick up, there’s often more than just pure reversion to the mean at work. Often, the price changes due to positive perception changes (hopefully warranted), but as things undershoot below the mean, they overshoot on the upside too. So, in future, I’ll let such situations run, allowing a small amount of overvaluation (and maybe even a trailing stop-loss) to take advantage of the positive perception changes.
In summary, I like MDT – it has good consistent revenue and earnings growth, good ROE and low debt. But at the current price it feels like the high-end of fair value so I would not buy. Assuming the story stays the same, I would be interested again below $33.
“Part 2. Scuttlebutt“ will look at industry specific insights, scuttlebutt and assess whether these actually help me in my assessment of a situation. It will include some of the more qualitative aspects and suggest maybe why perception has started to change around MDT.
By Raman Minhas