Multiples vs. Valuation, Growth vs. Returns, Cheap vs. Value, Drivers of Value Creation, Costco Case Study


Don’t Confuse Cheap With Value – Broyhill Asset Management, November 2016

  • Multiples are shortcuts, not valuation
  • Growth is not value without adequate returns
  • Moats matter, don’t confuse activity with accomplishment
  • Market’s P/E multiple has historically averaged around 16x normalized earnings
    • If you buy the market at an average price, you should expect long-term average returns and vice versa – this is common sense
  • Common yardsticks tell you little about valuation but they make great shortcuts for lazy investors and good material for talking heads – in order to understand what something is worth, need to understand two variables: cash in, cash out
  • Berkshire bought See’s Candy for $25MM in 1972; sales were $30MM; pre-tax earnings were less than $5MM; capital required to run the business was $8MM
    • From 1972 to 2007, sales grew to $383MM; pre-tax earnings grew to $82MM; capital required to run the business was an additional $32MM; additional capital generated $1.35B in cumulative pre-tax earnings
    • Sold 16 million pounds of candy in 1972 and 31 million pounds in 2007; that’s only 2% CAGR, yet the business created tremendous value (because it generated high returns on invested capital and required little incremental investment)
    • Over time, it’s hard for investors to earn returns much higher than the underlying business’ return on invested capital
  • Don’t start with growth, it is better to start with return on capital
  • Relationship between a firm’s growth rate, return on capital, and P/E multiple:
    • Company earning cost of capital will trade at “the commodity multiple” irrespective of growth (they can speed up or down but it makes no difference)
    • Growth is good for companies generating returns in excess of their cost of capital (in this case, faster growth translates to higher multiples)
    • Companies that earn below their cost of capital destroy shareholder value
  • Can break down the value of a firm into two components: steady state value & future value creation
    • Steady state value: this is when incremental investments earn the cost of capital; steady-state value can be converted to steady-state P/E multiple by inverting the cost of equity (8% cost of equity translates to P/E multiple of 12.5x); no guarantees but mean reversion comes close such that majority of businesses are average and average businesses ultimately see high returns revert to the mean (toward cost of capital)
      • Estimating steady value doesn’t even require forecasting but the problem is businesses are rarely stable (equilibrium itself has rarely been observed in real life)
    • Future value creation: for one reason or another, some companies can repel competition and maintain above average returns on capital for longer than average (moat); future value creation is a much bigger piece of firm value for these gems
  • There are three key drivers of value creation: spread between return on capital and cost of capital; magnitude of the investment; and how long a company can find investments at a positive spread
    • First two drivers dictate the rate of growth; the higher return a business earns on invested capital, the more cash it will produce and the more value it will create
    • Third requires more judgment; anticipated period of value creating investment opportunities is different for various industries
  • If you buy a bad business at a low enough price, an occasional positive surprise can give you the chance to sell at a decent profit, even if the long-term performance of the business is terrible
  • Time is the friend of the wonderful business and the enemy of the mediocre
  • Two ways to behave as an investor:
    • Buy something cheap, sell at a profit, and repeat (everybody does this to some extent); this requires making hundreds, maybe thousands of good decisions over the course of one’s career
    • Buy a great business – it appears far simpler, but is far more difficult to execute (almost no one does this despite the obvious advantage of lower Tx costs, taxes, fewer decisions, and higher quality of life); this requires the right investor base and requires a lot of patience
    • “All of humanity’s problems stem from man’s inability to sit quietly in a room alone” – Blaise Pascal
  • Costco (COST) Case Study:
    • Over the past two decades, COST generated a cumulative return of 1790% vs 337% in the S&P – probably safe to say that COST was not overvalued over this period
    • Costco membership costs about $45/yr and in exchange, people shop at Costco because goods are priced at a fixed 14% markup over cost
    • Costco’s operating costs are extremely low and this makes it difficult for even WalMart to compete as they can’t make money pricing goods as low as Costco
    • In order to make money selling at 14% above cost, revenues need to be very high
      • Competitive advantage is derived from what management does with this revenue advantage – passes efficiency gains back to consumer to drive more growth
      • Consumers benefit from firm’s expansion which drives supplier prices down
    • Over the past two decades, traded on average at 24x forward earnings estimates (stock looked always expensive)
    • We estimate steady state value for COST around $74/share which is nearly half of COST’s recent stock price – but this is conservative due to a number of factors temporarily depressing current profitability
    • COST stores are immature – newer stores only generating $80-100MM in sales vs $160MM on average and $180MM for stores open for 10+ years
      • Assuming new stores eventually ramp up to average, newer stores are under-earning by $8.7B
    • If you consider likelihood for a near-term hike in membership fees and maturation of existing store base, normalized steady state value increases to $80/share (60% of today’s value)
    • Costco is profitable enough to self-fund growth and has done so throughout history; so growth has been more measured in pace and more sustainable as COST is not dependent on capital markets
    • 5-year average ROIC stood around 13% vs. long-term average of 12%; despite the recent increase, believe normalized returns are greater than indicated by reported financials – assuming maturation of existing stores, estimated normalized ROIC of 16%
    • Assuming 16% returns, company’s earnings multiple should fall somewhere in the range of 15.7x to 25.5x depending on one’s expectation for earnings growth (4% to 10%)
    • Management appears comfortable with current pace of square footage growth – assuming 4-5% annual growth, equates to 1050 to 1150 stores in ten years
      • For perspective, HD and LOW have 4,000 combined stores in the US alone
    • In summary, future value for COST might be in the area of $40-80 per share – adding steady-state value of $80-90/share, puts fair value in the range of $120 to $170
      • Bottom line, would be happy to own COST at a 25% discount to our estimate of intrinsic value or roughly $100/share
      • At an entry point of $100/share, assuming mid-single digit comps and store growth, would expect to generate ~20% annual returns

Source: Broyhill Asset Management Research Presentation


Leave a Reply

Your email address will not be published.