Why is D&A usually taken as a proxy for Maintenance CAPEX?

[manuelbean]

Hi guys,

 

Why exactly is it assumed that the Maintenance CAPEX will roughly equal the D&A expense?

 

*This question is linked to another one I've posted about ROIC, but given that there might be other investors with the same doubt, I thought that creating a new thread would be a good idea.

 

Let's picture a company with 1 asset. A house that was bought by $1M. Let's say that the company depreciates the house in 25 years to zero and it won't replace it after that period. If the maintenance CAPEX equals the depreciation charge, it means that after 25 years the company has spent $2M on that house, right?

 

Why is this so? Why should it match?

 

I know that finding the Maintenance CAPEX is a long time quest for several investors and I don't want to overcomplicate. I just want to understand the reason people use the D&A as a proxy.

 

Thank you

[Munger_Disciple]

Why exactly is it assumed that the Maintenance CAPEX will roughly equal the D&A expense?

 

There is no such relationship and it should not be assumed. You REALLY need to understand the business behind the numbers. Maintenance CAPEX and its relationship to D&A depends on the business; to me that is obvious. For railroads, maintenance CAPEX routinely exceeds D&A. For other businesses like real estate the opposite might be true. For some others the two might be roughly equal.

[Guest cherzeca]

Why exactly is it assumed that the Maintenance CAPEX will roughly equal the D&A expense?

 

There is no such relationship and it should not be assumed. You REALLY need to understand the business behind the numbers. Maintenance CAPEX and its relationship to D&A depends on the business; to me that is obvious. For railroads, maintenance CAPEX routinely exceeds D&A. For other businesses like real estate the opposite might be true. For some others the two might be roughly equal.

 

well put, and this goes double for amortization of intangibles

 

edit:  and goes triple for the non amortization of goodwill

[gjangal]

In my view, it’s a just a proxy for maintenance capex. This can be used to figure out incremental capital invested in the business and what kind of return are they getting for this incremental invested capital. It’s not very useful nowadays if a lot of the investments are funneled through the income statement rather than capital expenditure

[SHDL]

I agree with what the others said.

 

As to why this can be a decent proxy however, imagine you had a small rental car company that owns a fleet of 1000 cars. Suppose each car lasts for exactly ten years and costs $10,000 to replace. Assume no growth and no car price inflation. Then the book value of the fleet is $10m and annual depreciation is $1m. And your maintenance capex is also (approximately) $1m because each year you need to buy about 100 cars to replace the ones that went out of service.

 

[SharperDingaan]

The expectation is that over the long term, the business will continue in its current form. For most businesses, this is just not a practical expectation, as businesses evolve over time (different product lines, market shares, etc.).

 

Its more relevant in the extractive industries (mining, o/g, etc), as there is a minimum level of annual exploration/development/drilling that must be done to maintain production at current levels. Complicated by rapid obsolescence/revaluation as the MV of the resource moves up/down.

 

SD

 

[D33pV4lue]

I agree with what everyone above said. It really boils down to each specific company. The one thing I would add is you have to know the company extremely well. If you know Company X is building a new factory that will add capacity (growth) I would not penalize the company for making the investment as it will drive higher revenue in future periods. Whereas if a movie theater company that has 500 locations is renovating seating once you get to the 500th location you will have to go back to 1. In this scenario you are generating higher revenue through upgrades but the costs are really maintenance not growth. In certain cases, you can estimate maintenance capex from the companies PPE schedule.

[scorpioncapital]

I think it's environment dependent. When I look at companies today i see capex less than d&a. Buffett wrote an interesting article about Inflation swindling the equity Investor by causing businesses to require ever increasing amount of capex to simply maintain their earning power (ie just maintain even 0 percent real growth). So if the environment will change , you may sniff it out in a change in the capex to d&a ratio.

[mjs111]

If you look at a theoretical environment with no inflation, where 100% of capex and acquisitions can be depreciated and/or amortized, then once a company hits terminal growth annual capex and d&a charges should level out.  In reality, I've found this rarely to be the case, even for 100+ year old firms like Coke and Johnson and Johnson.  All it takes is a little inflation or the occasional acquisition with a little bit of unamortizable goodwill or land value to boost capex and acquisitions higher than d&a indefinitely.

 

In my valuations I'll typically have capex plus acquisitions get closer in line with d&a long term (this should happen as growth comes down), but I can't recall ever bringing capex down to completely match d&a.  There's too much friction from inflation and unamortizable stuff for that to typically happen. 

 

What helps bring this into focus is to look at 3, 5, and 10 year averages of d&a vs. capex plus acquisitions.  Year to year capex is too lumpy but when you look at a string of years, what I've generally found is that capex always averages a bit higher than d&a.

 

 

Mike