Here’s to our wives and girlfriends… may they never meet! –Groucho Marx
MCX and Iridium Case Study
Our first discussion of Maintenance Capital Expenditures (“MCX”) occurred here: http://wp.me/p1PgpH-6t
One method of learning is to EXHAUSTIVELY analyze and read about a subject so we can master the topic and understand the principles and subtleties in applying those principles.
We are focused on Return on Invested Capital which has been defined one way as Operating Earnings (Earnings before Interest Expense and Taxes, EBIT) or better yet, (Earnings before Interest Expense, Taxes and Depreciation & Amortization, “EBITDA” – MCX) divided by tangible capital or (Net Working Capital + Net Property, Plant and Equipment). We have covered EBITDA thoroughly in a 36 page discussion here: http://www.scribd.com/doc/66843869/Placing-EBITDA-Into-Perspective.
Now we review MCX as part of the (EBITDA – MCX) calculation.
The link below has a PDF that further analyzes how to calculate one aspect of Return on Invested Capital–(EBITDA – MCX) divided by Tangible Capital.
Also, this case study will be placed in the VALUE VAULT
The Bridge of Death
Please refer to the prior post. A reader generously sent me this analysis of capex from Sanjay Bakshi (Fundoo Professor). Go to his blog (recommended) http://fundooprofessor.wordpress.com/
. The more contributions from readers, the more we will all learn about specific aspects of investing. Thanks!
Capex is given in cash flow statement in the investing section. All capex which is not growth capex must necessarily be maintenance capex. So if you can estimate growth capex, since you know total capex, you can estimate maintenance capex.
Estimating growth capex is sometimes easy because one knows that the company is expanding capacity from x units to y units and its capex plans are known. Even when we don’t know exact capex plans, sometimes to we know how much money is needed in an industry to expand capacity from x to y.
Sometimes we know that almost all the capex is maintenance capex because of competitive pressure. For example what happened in the “Shutdown of textile industry” example, or in the petrol pump example i gave in class. The functional equivalent of these examples are very very common in the business world so one must not take the accountants’ definition of capex as sacrosanct. Accountants don’t like to make estimates. They would rather have something precise even if its wrong. You don’t have to be like them.
When you look at capex numbers you must ask the following questions:
1. Is this expenditure likely to result in a sustainable rise in economic earnings in the future? This is not the same as asking if margins will improve or not – they may improve temporarily but the nature of the business may be such that such gains would be temporary and the cost savings will flow through to the customers instead of the owners. So what may look nice on DCF analysis in an excel model, will not translate into sustainable economic earnings jump. You really need to do this “second step analysis” by asking how much of the benefits of the capex will go to the owners and how much to customers. If the benefits will go to customers, then its NOT Capex for US. For US its and EXPENSE. something we reduce from operating cash flow to arrive at owner earnings.
2. How competitive is this industry? If its extremely competitive or has a lot of foolish competition, then its very likely that capex won’t result in improvement in long-term sustainable earning power. In some businesses, you just can’t be a lot smarter than your dumbest competition. And there is plenty of DUMB competition in some industry. Witness for example, what’s happening in the airline industry in India right now…
3. Is there a lot of inflation? Historical cost accounting and inflation result in under provision of depreciation in accounting books.
4. Are there substantial productivity improvements? Inflationary effects are sometimes offset by productivity improvements. This was discussed in class in detail.
5. Is the plant really old and dilapidated and probably needs replacement? Sometimes there are very old plants chugging along for a long time but ultimately they have to be replaced. The plant may have almost zero value in books because they have already been written off. But they need to be replaced and replacement cost could be a bomb. When you spent that money, should you record it as Capex? Well the accountants will ALWAYS record it as capex. But they key question you have to ask is this: will this money spent improve long term earning power, or only help the company maintain current earning power. If the consequence of the money spent is going to be mere maintenance of earning power, then its NOT CAPEX FOR US. Rather its an EXPENSE.
6. Whats the rate of Obsolescence in this industry? Industries with very high obsolescence rates require frequent capex just to keep up with the competition which keeps on inventing new applications of technology making old technology obsolete. We talked about what happened to Moser Baer and Samtel Color. Well the unfortunate people who invested in those companies learnt the hard way that essentially these companies made NO REAL PROFITS because when you CORRECTLY TREAT the money spent on frequent capex programs of such companies not as CAPEX but and MAINTENANCE CAPEX, you would have figured out that essentially there were NO OWNER EARNINGS. And when there are no owner earnings there is NO VALUE. This is true even though interim value in the market may end up being BILLIONS OF DOLLARS!
A reader asks about calculating capital expenditures and Buffett’s owner’s earnings. I believe only maintenance capex is deducted in determining owner’s earnings not growth capex because maintenance is mandatory while growth capex is discretionary.
This document is 11 pages and it includes other links.
This post responds to a reader’s question from Lecture #1:
Lumilog | October 21, 2011 at 8:46 am |
I’ve been wondering more about the estimate of Maintenance Capex Greenwald gives on slides 13-14 and its reliance on revenue. A company could theoretically grow revenue, but due to higher costs, end up with about the same EBITDA. In that case, I’d want to assume that all was Maintenance Capex despite the revenue growth. So I’m wondering if it might be better to estimate Growth Capex from the delta in EBITDA that CapEx produces, not the delta in revenue.
From page 96, fn. 1: Value Investing From Graham to Buffett and Beyond by Bruce C. Greenwald
Companies generally report capex in their statement of cash flows. We assume that each year, a part of this outlay supports the business at its sales level for the prior year, and part is needed for whatever increase e in sales it has achieved. Companies generally have a stable relationship between the level of sales and the amount of plant, property, and equipment (PPE) it takes to support each dollar of sales. We then multiply this ratio by the growth (or decrease) in sales dollars the company has achieved in the current year. The result of that calculation is capex. We then subtract it from total capex to arrive at maintenance capex.
Also, you can simply ask the company’s CFO what amount of total capex goes to maintenance capital expenditures. Certain companies like Iron Mountain “IRM”(Document Storage-physical and digital) will specifically break out the two type of capital expenditures. In IRM’s case, its maintenance capital expenditures are low compared to revenues or any other accounting metrics because their storage facilities have long lives with minimal upkeep.
To: Lumilog, you do not want to assume in your example (A company could theoretically grow revenue, but due to higher costs, end up with about the same EBITDA. In that case, I’d want to assume that all CapEx was Maintenance Capex despite the revenue growth) that all capex is maintenance capital expenditures (“MCX”) because you want to segment the company’s capital spending. You could have non-productive growth capex with stable MCX. For example, what does it cost to maintain old stores at their current sales level vs. the cost to build and develop new stores?
Another point, you have to understand the industry and competitive forces to calculate/estimate true MCX. Take, Iridium, the satellite company that is launching new satellites into orbit, for example. That company may need to replace their new satellites sooner than expected due to new technology in competing telecom industries. Investors who own Iridium could be vastly underestimating MCX and therefore, overestimating normalized earnings. Earning power value and thus asset value may be lower than what current investors estimate.
Hope that helps……….