Interesting links this week

Sanjay Bakshi - Bayesian thinking piece - Must Read

The most recent Howard Marks memo

Recent podcast of Julia Gatlef on Phil Tetlock's Superforecasting book - great site and Julia has some very good videos on youtube for understanding cognitive functions and scientific stuff.

After Amazon's recent earnings result I found re-reading this great post on their business model

And of course, Andrew Left of Citron Research on Valeant - this is a good introduction for anyone who, like me, has zero knowledge of the biotech world

Pebbles of perception: Book review

Pebbles of Perception by Laurence Endersen is a little big of wisdom and, not surprisingly, stemmed from the author's reading of Poor Charlie's Almanack and his resonation with Mungers thinking. It is a very easy read and gets you to stop and think about all aspects of choice within life and steps to take to ensure you do the best to make the best choice.

Life is about decisions and this book gets you to think more about perception, incentives and emotion in order to go through life making good decisions. I have highlighted a few main points from the book that really made me think:

1. Incentives - 'the rabbit runs faster than the fox, because the rabbit is running for his life while the fox is only running for his dinner'. The chapter on incentives is great and immediately made me think of a bit of advice I got from a former colleague that 'every decision and problem in the real world stems down to incentives and expectations of the parties involved'. The author links incentives to the 'eat what you kill' culture in finance and how there is nothing more dangerous tan underestimating incentives.
2. Perspective - ' too far east is west and too far west is east'. Understanding perspective and contextual differences in decisions will help understand outcomes.
3. Fear and uncertainty - The future is uncertain. The author uses a great analogy of fear of the future in life. 'We are the captain of our own ship. Fear of the future is the anchor that holds us in the harbour. Fear-ruled ships stay in safe harbours, but what use is a ship that won't set sail?' Uncertainty is forever present in all decisions. But without risk there is no return. Understanding the forever presence of uncertainty can not only help us calculate or imagine probable scenarios but also help us strive to eliminate as much uncertainty as possible. This point also resonates with Seth Klarman's comments on uncertainty which can be found in the link in my previous blog post here. 

Munger, MITMCo and Klarman: Interesting links from this week

First link is from Munger on Coca Cola and one of the best examples of 'inverted thinking'. No explicit mentions of basic valuation metrics we would see in most analysts' work and is a must read for anyone attempting to think differently:

Charlie Munger: Coca Cola Analysis

The following link is an interview by Manuel of ideas with MIT's Investment Company MITMCo. Great insight into what they look for in managers and how they analyse the thinking process and characteristics of managers. Great read for money managers and aspiring investors.

MITMCo: Perspectives of Aspiring Superinvestors

The third link is an article by Seth Klarman on uncertainty written in Feb 2009, the midst of the crisis. I think this piece resonates well with recent spouts of high volatility and uncertainty in markets today. Understanding uncertainty and the limits to your analysis helps understand the risk of the investment and also spurs you on to eliminate as much uncertainty as possible.

Seth Klarman: The value of not being sure

Lectra - earnings power of a quality business

Lectra – Free growth in a quality business.  

Lectra is the world leader in integrated technology solutions dedicated to industries using soft materials. The firm designs, produces and markets full-line technological solutions – compromising software, CAD/CAM (computer aided design and manufacturing) equipment and software and associated services such as PLM. This mainly covers automakers, fashion and apparel and furniture amongst other industries. Two founding brothers have a combined 30% ownership of the business own the company.

They have been heavily spending on R&D recently, planning to benefit from the longer term trend of brands modernising their manufacturing processes, using seamless data and technological solutions to improve cost efficiency and product quality. Lectra estimate they have around 25-30% market share in CAD/CAM (computer aided design and manufacturing) software and equipment business. This company is not statistically cheap on basic valuation metrics, although the quality and earnings power of the company provide value.

Business

In 2011 they decided to invest for the long term and transform the company. They have committed €50m to spend on innovation, research and marketing in the years 2012-15. This is a company that is investing for the long-term future of the business and is noticing the rapidly changing requirements of the sector. They have actually spent around €20m per year since 2012 on R&D, although French taxation and research credits mean the net spend is actually around half this value. The company sells machinery with software, which is then contracted annually, providing Lectra with recurring revenues. Revenues from deploying new systems for new clients is the growth revenue of the business.

How much of a margin of safety does the recurring revenue give and what are reasonable growth rates of the growth revenue?

As always, lets think safety first. The recurring revenues are from contracts of existing systems, spare parts and consumables. In 2014, Lectra earned €122m in recurring revenues, which just about covers the fixed overhead costs of running the business entirely. This fixed cost, which is mainly salaries for the engineers and brains behind the solutions, amounts to around 75% of gross profit for the last 10 years. This recurring revenue has also been around 55-60% of total revenues consistently and provides some safety, although obviously subject to some risk.

How reliable are these revenues? How replaceable are they for consumers, i.e. what is the replacement cost of the service and product Lectra provides?

Lectra has around 23,000 customers from all industries, and although they do not equally contribute to revenue, it would take many small or a few large customers to literally shut down operations for Lectra to lose a huge chunk of revenue. FY2014 – no single customer represented more than 7% of revenue, the 10 largest customers account for less than 20% and the 20 largest less than 25%. The sales are fairly evenly spread geographically with 46%, 24% and 23% coming from Europe, Americas and Asia respectively and are evenly spread between both automotive and apparel sectors, with a small portion in aeronautical and furniture. This provides some sustainability in revenues and predictability in the underlying business.

I believe the replacement cost of the machinery Lectra provides is fairly high. I am assuming here that once Lectra do employ a system (machinery and software) for a client, like Oracle and other back office operating systems, it takes huge cost and efficiency incentives for the client to change. A brand is not going to replace all their systems, plus educate and teach their employees to use a new system, with a new competitor once Lectra have installed their software into the brand’s processes. Therefore, Lectra benefit from great first mover advantages and the innovation and R&D provides high barriers to entry. The main issue is recruiting and educating the clients on the solutions they provide.  

Growth

Selling new systems, the machinery and the software with it, is the growth driver in this business. They have COGS made up of purchase and freight costs which amount to around 25% of revenue giving a 5-year average gross margin of 75%. Operating costs are c55-60%, which involve fixed overheads and fixed and variable wages. This gives operational leverage to the business and if they begin to sell many new systems, a large amount of marginal revenue flows to bottom line. FY2014 the firm had operational leverage of 2.27 (fixed costs / variables) and therefore for every 1% change in revenue, operating income will increase 2.2x. These fixed costs have been around 55% of revenue for the last 10 years consistently. The company has negative working capital requirements as they receive payments for their software and equipment and then shift the goods/service. As long as gross profit from recurring contracts cover personnel costs, a security ratio management refer to, the risk of this investment is reduced. Lectra benefit from 30% R&D tax credit and employment tax credits from French government enabling them to literally spend all cash generated on growth.

Instead of asking the obvious questions around growth; why/how will they sell these new systems and what kind of growth rates can we expect, I am going to invert the question and say how is it possible they will not grow and in what scenario would this happen?

Lectra estimate they have around 25-30% market share in their core business of CAD/CAM equipment and services. I have mentioned before about the replacement cost of their services and that it would need a seriously highly efficient, low cost service provider to take Lectra’s clientele. Another way to not grow is to implode and this would cause a global crisis to halt all manufacturing in the fashion and apparel and automotive industries. 

The company’s cost base is largely in euros in France, which has given Lectra an edge over its US competitor for the last year. Adjusting for Lectras intangible strength, mainly customer relationships and brainpower in this niche field, and the remaining assets shows the reproduction asset cost of business is around 1.5x book value. The value of the assets actually needed to start a competitor in this niche field is around €160m. The earnings power and reproduction cost show the company have a competitive advantage. The majority of the assets are intangibles and therefore the main risk is a large competitor entering the market with greater innovative solutions. This is a risk, although the niche nature of the business makes it highly inefficient for a large player to spend as much time as Lectra to enter a small market. 

I feel there is definitely spare capacity in the demand on the micro level for Lectra’s services as global brands move towards using technology and data in all processes. I am confident in the long-term growth in new systems as it is clear more and more brands will need to adopt systems that enable the full use of seamless data and technology to maintain their brands cost efficiency and quality. The business is sustainable on recurring revenues and operational leverage is going to accelerate earnings once new systems are employed. Also, Lectra have publicised a 10-year plan to speed up China’s product development and add value to their design and manufacturing activities. Their strategy to targeting growth seems to be adequate although valuing this growth is difficult. To attempt an estimate of the value of growth I use the basic Greenwald method:

Lectra have spent c€20m a year on innovation, engineers, marketing and revamping the business. This R&D is fully expensed each year and actually only amounts to 70% of the total value due to tax credits.

Assuming 5% growth rate and 10% WACC:

Future cash flow = (ROC – g) * Capital invested

                                          = (.35 - .05) * 55m = 16.5m

Value of a growing firm = 16.5m * 1 / ( .1 - .05) = 330m or this can be proven by:

(ROC – g ) * Capital / (r –g).

This value of future growth gives 25% upside to the current value of the firm.

Valuation

Greenwald states that ‘when earnings power is above market value, growth comes for free’, and Lectra seems to provide this. The firm has great potential to evolve into a high growth stock. Average operating margins for the last 5 years have been around 11-12% and the operational leverage in the business provides space for expansion as they grow.

Management claim to have around 25-30% market share in the CAD/CAM software and equipment sector, and they have patents running until 2022 that protect this share and therefore margins. The power of these patents were recently proven when Lectra won a battle with a German manufacturer claiming they had infringed on Lectra’s airbag cutting system. I feel there is a high chance Lectra will remain global leader in the CAD/CAM sector and I do not consider any growth in the PLM market as this is a very competitive service. 

The current price of Lectra is not cheap by any means. Trading at around 9.8x EBIT and 16.8x FCF with a PE of 16.8 with all metrics higher than their 5 year averages. However, it’s clear you are paying for quality here. Insider ownership, 35% 5-year average ROIC and consistent cash flow proves the quality of this company. Lectra's main competitor is a private company called Gerber. Vector Capital bought Gerber Scientific, in 2011, for $282m. This price was at around P/S 0.5 of the last 4 years total revenue, and a huge PE of over 40 but no recent data is available on Gerber.

I have calculated the earnings power value of Lectra to be around €12.09, giving an 11.5% margin of safety to the market price. This is using a 10% cost of capital and being conservative with adding back only 10% of SG&A and R&D. This current earning power eases some of the uncertainty against future growth in new systems. I am happy to pay for quality here. Also, looking plainly at the facts of the firm investing over €60m in the last 3 years, with an average ROIC of around 35%, this has a high probability of returning at least €21m in cash in the coming years.

A valuation narrative

" What sets pricing? Supply and demand. Mood and momentum. What sets value? Cash flows, growth and risk. Can the two give you different results? Absolutely. "

The quote above come from the Google Talks by Damodaran, the King of Valuation.

Valuation is subjective. I find DCF's and any other technical measures bring out the worst heuristics and biases that investors face. Searching for a WACC or cost of equity is easy to use a round number such as 10%, and assumptions for growth can be backed out of your perceived (the price you aim the valuation exercise to give you) target price.

I've made my valuation process this 'simple': concentrate on finding a narrative for growth, risk and cash flows. The story should give logical reasoning for all assumptions, for example how much does a company need to reinvest to earn these rates of growth and cash flows? I am now in the process of making a checklist that will help me build this narrative.

DCF comes in two parts: the cash flow, and the discount rate. Risk, growth and cash flow. Tell a narrative to get your assumptions for these three factors that drive the valuation. Find ranges of assumptions based on bounded rationality. Use 'laws of the universe' when building a narrative, i.e. an operational margin of 50% is not sustainable in the long term and doesn't come for free.

Building a narrative reminded me of Munger's quote:

'People calculate too much and think too little'

Berkeley Group Holdings

Berekely Group Holdings (BERK)

Px - £35

Target Price -15x to 19x 2015 NOPAT.

Fundamental drivers of the business

Housebuilding is a capital-intensive business. The more capital you invest, the more profit you can make just as in any commoditized business. The challenges in the business come from the competition you have purchasing the sought after land, and the margins you earn are constrained by the demand and availability of the houses nearby.

It seems logical to value homebuilders on the capital they have on their balance sheet and the ability of this capital to generate cash. Tangible book value is the assets that generate earnings for homebuilders, not intangibles so much as it would for a company such as Microsoft. It seems as if the market understands the cyclical nature of house builders and therefore never gets too excited and bids them to high PE multiples. Forward PE is relatively low and P/TBV is at a significant premium, which actually indicates that the market questions the sustainability of recent earnings.

Company	ROE	Price to Tangible Book	Current ROIC	EV/NOPAT
Barratt Developments	11.45	2.45	20.59	10
Bellway PLC	17.42	2.11	22.03	15.15
Berkeley Group Holdings	27.33	2.87	39.89	9.6
Bovis Homes Group	12.6	1.63	15.34	11.5
Crest Nicholson Holdings PLC	21.78	2.6	25.41	14.4
Persimmon PLC	21.03	3.28	33.54	15.56
Taylor Wimpey	16.81	2.69	20.09	16.45

Therefore, buying homebuilders today expecting a re-rating in market multiples is nonsensical as investors know the cyclicality of the business and therefore even in the boom years the PE ratio is never hugely inflated. Berkeley has a 5-year average PE of 12.8 and PESN of 13.8 BERK current PE is at 12 and therefore I do not expect a large expansion in multiple to driver return.

Sustainable earnings growth

If we cannot expect re-rating in terms of multiple expansion, we need favourable fundamentals to drive earnings. These are mostly driven by mortgage availability, government policy, FDI and overall supply and demand of housing and land. I feel that we have been in a pretty ideal environment for the past couple of years, are we too late to join the party?

One fact remains, there is a shortage of housing in the UK, especially London and South East, and the government are publicly supporting housing formations. The future path of rising rates will probably net off, I hope, increases in real wages and therefore an investment thesis here is based on a continuation of the fundamental factors that drive the business.

Paying over book value for a house builder such as BERK does not worry me as much, due to the geography and type of housing they build. Firstly, London and the South East is an area that is subject to persistent demand due to urbanization and foreign direct investment. The UK government and it’s Help to Buy scheme has also pledged to complete £12bn of guarantees over the life of the scheme and as of June 2015 they have only completed £1bn. Spare capacity for mortgage growth to continue its trend and I am somewhat confident the BoE rate rise will not affect this too much.

The Office for National Statistics report that in London there is an estimated 2800 hectares of land available for housing in London that is not in use or available for redevelopment, and over 6000 hectares in the South East which can provide around 330,000 and 210,000 houses respectively. Seasonally adjusted housing starts are still 32% below the March quarter 2007 peak with completions 26% below the peak, highlighting some spare capacity within the industry. So the supply is limited, although available, albeit at a certain price, and therefore the issue comes with not finding the supply to build, but finding it at favourable prices. This is where the business model of BERK will prosper. By taking on complex plots of brownfield land, they have full support of the government who introduced policies in 2010 and extended in 2014 to target brownfield sites for housing and build around 200,000 houses on brownfield sites by 2020.

Business Strategy

With regards to land cost, BERK need to ensure the spread between housing sales and land purchase is adequate, and their timing of purchases throughout the cycle is not prone to basic behavioral biases, i.e. overconfidence buying at peak of market. This emphasizes the importance of ROIC in this business and the expertise of management in capital allocation being crucial to the success of the firm. BERK emphasise their long-term outlook on their business and claim to understand the cyclical nature of the market, which determines their buying habits.

BERK has consistently increased capital invested and therefore inventories too, as these are the assets that generates return in the business. Capital has increased from around £1bn in 2006 to over £3.5bn today and ROIC has followed from 10% to over 35% in the same period. Having a consistently increasing asset base de risks the business from any huge fluctuations in underlying housing prices. On this capital, the high and increasing levels of ROIC highlight the profitability of the business and the great capital allocation of management.

Understanding the markets in which we operate is central to Berkeley’s strategy and gives us the confidence to buy land without an implementable planning consent where we understand what local stakeholders want. (BERK 2015 10k)

BERK is in the business of placemaking, not just housebuilding. They claim to create value by identifying and purchasing land and to then build and sell. They take on complex sites and ‘brownfield’ land (areas that previously had commercial or industrial use) and then seek to create developments. This is a very different business model than its competitors who seem to buy ‘consented land’ or housing sites.

The firm takes the risk of buying land that has not yet been consented to developments. This is a huge risk, which can weigh on profits and margins if the acquired land has planning permission rejected. BERK is delivering 10% of all new homes in London and has unconditional contracts for sales that deliver £3bn in cash over the next 3 years. The firm currently has 37,473 plots of land plus strategic options on 5,000 plots. At an average selling price of £456,000 this gives gross margin of £5.2bn (c30%) on the balance sheet. Gross margin is driven by house prices and land purchasing strategy. The cheaper you buy the assets, the higher the margin when developed and monetized. Land supply is constrained and in demand, therefore margins will definitely be squeezed going forward due to new entrants and scale. Land plots and inventories have a CAGR of 5% and 13% respectively for the last 6 years and capital employed has increased 2.2x over the 6-year period. Capital is crucial to buy the land that eventually generates the gross margin and flows to earnings.

Valuation

EV/NOPAT – concentrating on the ROIC of Berkeley to value the homebuilder provides some metric of valuation to target. Berkeley has been able to grow rapidly post-crisis due to their successful and shrewd business of buying property at depressed prices during the housing crisis and therefore have been able to greatly expand margins in recent years as they have sold these properties. In the long run these high returns on capital of over 30% are not sustainable and therefore to find a fair value we use the 10-year average ROIC of 19%.

Using a WACC of 10% and long-term growth rate of 3% (inflation + greater asset price growth in London and South East). Using 10-year average range of ROIC 15-19% this gives a fair value range of 11.4-12x 2020E NOPAT.

Current valuation is 9.7x Apr 2015 diluted NOPAT. Now how much do we expect Berkeley to grow? Revenue and EPS has a CAGR of 33% and 44% respectively over the last 5 years. Assuming BERK grow at 10-15% per year for the next 5 years, this implies that BERK’s fair value is 1.6x-2x 2015 NOPAT. Hence the fair value and price target of BERK for the next five years is 15.5x to 19x 2015 NOPAT which gives a price target range of £52-57.5 providing a 46% upside potential over the next 5 years.

BERK also pledged in 2012 to return £13 per share to shareholders by 2021. They will have already committed £4.34 by Sept 2015; therefore £8.33 in dividends from 2016-21 is on average £1.73 a year which yields 5% at current prices and when discounted at 10% this gives a PV of £6.57. This provides 18% of return on top of any potential price appreciation driven by earnings growth.

Sanjay Bakshi Podcast

One of my favourite blogs to read on a Sunday morning is Shane Parrish's Farnam Street a blog covering all disciplines, with plenty of Mungerisms, great book recommendations and learning material. This week covered a Podcast with Sanjay Bakshi, an Indian Finance professor and successful value investor, founder of Value Quest Capital LLP (audited 24% gross return from 2002-12). Found a post on his career and his journey to become a value investor here. Fascinating.

10 points from the talk:

1. He reads on a Kindle, which when he makes annotations, sends the notes to the cloud which he can then access when the book is finished. This seems to be much easier and more efficient than underlying text in the book which can be a pain to find at a later date. 

2. Charlie Munger's multi disciplinary approach is essential to becoming a complete investor, i.e you cannot understand economics without understanding psychology and incentives. 

3. Financial independence is key to being able to think truly long-term like Berkshire. 

4. Talks about a 'physical space', somewhere with no distractions that enables one to make rational decisions. He refers to Guy Spier's room with no electronic devices and Buffett's move to Omaha from New York. 

5. He looks to minimise noise. He doesn't have a Bloomberg terminal or watch CNBC. Short term results, quarterly reports and next quarters EPS estimates are nothing but noise. He takes See's Candies as an example. A wonderful business that loses money 8 months a year but is hugely profitable for the remaining 4. Looking at short term info is noise. 

6. Low cost provider is the most sustainable moat. Speaks about Costco's business model. Sale of low cost brands, highly scalable, customers provide the business a float, pays customers better than competitors. Will look to do a post on Costco soon. 

7. Slow changes go unnoticed. Boiling frog syndrome - myth that placing a frog in boiling water it will jump out, but place it in luke warm water and then slowly boil it will not perceive death and die.

8. Give descriptive names to mental models to help retrieve them from your memory. 

9. Looking for a business/entrepreneur that is risk averse but not loss averse. Finding managers that are making asymmetric bets, not scared to take the right risk. This led me to think of a point made by Mohnish Pabrai about Bill Gates and Microsoft. Pabrai argues that Microsoft was not risky in the early stages. The actual amount of capital injected into Microsoft was very small but Gates dropped out of Harvard and concentrated solely on the project. He argues where the true risk of this project lies? If it succeeds it can change the world, literally, and if it fails, well Gates can go back to Harvard, complete his degree then find a job. Low downside risk, huge uncertainty. Good combination of a business or venture. 

10. Poor Charlie's Almanack is the book that has influenced him most.