Does depreciation matter?

[earthtodan]

I'm having trouble understanding why depreciation is counted against earnings. For example, let's say UPS buys a new truck. For the sake of simplicity let's say they pay for the whole thing up front. The cost of the truck gets subtracted from the company's earnings that quarter, because it was money out of pocket. The book value of the company does not change, except by the difference between the cash spent on the truck and its immediately depreciated value.

In subsequent quarters, the value of the truck declines, and the book value of the company along with it, but that doesn't cost UPS any money because it's already been paid for. All maintenance counts as OpEx. When the truck outlives its economic life and they sell it, they will exchange its depreciated value for cash, and the depreciation of the asset will have already been reflected in the book value of the company. In my mind, that should be the end of story. However in recording depreciation against EPS every quarter, it seems like companies are recording the cost of an asset twice.

The same should go for pipelines, drill rigs, railroad tracks, airplanes, office chairs, etc.

I was secure in this conclusion until I read the 2013 Berkshire Hathaway letter to shareholders. On page 14, Warren Buffet states:

Quote:Every dime of depreciation expense we report, however, is a real cost. And that’s true at almost all other companies as well. When Wall Streeters tout EBITDA as a valuation guide, button your wallet.

Warren Buffet isn't one to defend GAAP accounting in order to maintain the status quo; in fact, in the previous pages of this letter he makes clear his disdain for GAAP rules he does not agree with. Therefore it seems that he genuinely believes that depreciation costs a company money. So I ask the forum: How can that be?

Dan

[Dividend Watcher]

Part of the problem is the difference between the real world and the accounting world and it can get quite complicated. I'll try to make a brief stab at it.

First off, the goal of accounting is to accurately show the financial results of an entity. Further, it tries to match the inputs (revenues) and outputs (expenses) with the periods they actually affect the business. This way you can compare between other companies and standards and well as different periods of time for the same company's financial reports. The procedures for recording this has been accepted as the common method of reporting this stuff so everyone is talking the same language. Hence the Generally Accepted Accounting Principles (GAAP).

Hopefully I won't confuse this too much so soon but there are two basic equations that govern accounting:

• Assets = Liabilities + Equity (Balance Sheet)
  
• Profit = Income - Expenses (Income Statement or Profit & Loss Statement)
  

The first equation, affectionately called the "Accounting Equation", is a snapshot in time of what a company owns and what it owes. Equity is what's left over after all the liabilities are paid and commonly called the Book Value.

The second equation, the P&L for short, measures the profitability over a specified period of time; e.g., quarter, year, etc. and generally ignores what they own and owe. It only measures the money into and out of the business for the time period you're looking at.

You'll see these two reports for any company you're investigating. Both are important.

Profits can be used to pay off debt (liabilities) or put in the bank (assets), keeping the first equation in balance. That's why I like to use the concept of profit=equity to relate the two equations.

Now, back to your question.

(04-20-2014, 08:08 PM)earthtodan Wrote: For example, let's say UPS buys a new truck. For the sake of simplicity let's say they pay for the whole thing up front. The cost of the truck gets subtracted from the company's earnings that quarter, because it was money out of pocket. The book value of the company does not change, except by the difference between the cash spent on the truck and its immediately depreciated value.

In actuality, the truck is not deducted from earnings. All you are essentially doing is trading one asset, the money in the bank, for another asset, the truck (see equation 1). You haven't expensed anything.

That's great on day one. The truck is worth what you paid for it. Over time, however, the truck becomes less valuable because of wear & tear. It also is something that produces income for the company over a number of years. So, ten years from now, when they sell it, FedEx no longer can trade the asset for the original value in cash. You have to account for that decrease in value somehow. That is the depreciation expense which IS deducted from the revenues over time to reflect this decrease in value.

Remember I said you want to match the income and expenses when they actually occur? The depreciation time and values deducted are determined depending on the type of asset it is. Obviously, a truck will not last as long as a building so it will be depreciated over a shorter time frame.

(04-20-2014, 08:08 PM)earthtodan Wrote: In subsequent quarters, the value of the truck declines, and the book value of the company along with it, but that doesn't cost UPS any money because it's already been paid for. All maintenance counts as OpEx. When the truck outlives its economic life and they sell it, they will exchange its depreciated value for cash, and the depreciation of the asset will have already been reflected in the book value of the company. In my mind, that should be the end of story.

You're forgetting the truck is an income-producing asset. So, in reality, the decrease in the value of the truck (book value decrease) should decrease at a slower rate than the income it produces over time (book value increase). If not, why would the company even buy it? This profit (=equity or book value) should be greater than the decrease in value of the truck which should make the company more valuable.

(04-20-2014, 08:08 PM)earthtodan Wrote: However in recording depreciation against EPS every quarter, it seems like companies are recording the cost of an asset twice.

Nope. Never deducted from earnings in the first place. Depreciation records the decrease in value commensurate with the economic benefit derived from its use in the approprate accounting periods.

(04-20-2014, 08:08 PM)earthtodan Wrote: The same should go for pipelines, drill rigs, railroad tracks, airplanes, office chairs, etc.

It does and is handled the same way.

(04-20-2014, 08:08 PM)earthtodan Wrote: I was secure in this conclusion until I read the 2013 Berkshire Hathaway letter to shareholders. On page 14, Warren Buffet states:

Quote:Every dime of depreciation expense we report, however, is a real cost. And that’s true at almost all other companies as well. When Wall Streeters tout EBITDA as a valuation guide, button your wallet.

Warren Buffet isn't one to defend GAAP accounting in order to maintain the status quo; in fact, in the previous pages of this letter he makes clear his disdain for GAAP rules he does not agree with. Therefore it seems that he genuinely believes that depreciation costs a company money. So I ask the forum: How can that be?

Exactly as I described. It's the method of showing expenses for long-lived assets over a period of time which, theoretically, is the useful life of the asset.

Since it is a non-cash expense once it's been paid for, you can add the depreciation back to net earnings to get the cash flow and see what the company has to work with. What Warren is probably talking about is ignoring depreciation and taxes to see how well a company is doing. If FedEx needs 10 trucks to do all their deliveries but the slick truck salesman talks them into buying 100 trucks, the depreciation on those 90 idle trucks is being pissed away which will affect their ability to make a profit. If they only buy the 10 trucks, and they are doing the work of earning money for the company, then it is an expense but is being compensated for by the income. The same can be said for buying a business. If you overpay for it, those expenses have to be depreciated and can really hurt a company's performance.

Hope that provides some understanding. It's a complicated concept but essential when reading financial statements. When I taught small business bookkeeping for the Small Business Development Center, I had a 4 page primer that just covered capital transactions. You'd be surprised how many small business owners don't understand these concepts to their own detriment.

[Kerim]

Holy cow, DW. Great response! That is a better explanation than most I've read.

[Concasto]

Nice reply DW.

Depreciation may be a non-cash cost at the time earnings are reported, but if an entire fleet of trucks is wearing out, that non-cash cost turns into a real one(as they have to replace the trucks), so its good that the accounting takes this into consideration along the way.

I am not expert, but pipelines seems to be allowed excessive depreciation on their assets. DW, can you comment?

I am a fan cash flow from operations and free cash flow, which is the cash flow from operations minus the capital expenditures. Unfortunately all these metrics have hundreds of possible variables, so listening to conference calls and going over company presentations and of course watching dividend growth are all important.

[earthtodan]

Thanks for the response DW (weird, those are actually my initials). So to pull the main point that addresses my confusion, the purchase of a new fleet would not be counted against earnings for the quarter in which it was purchased? That's surprising because I keep hearing about EPS figures being thrown off by one-time capex. But it sounds like what you're saying is that only the balance sheet gets affected by a large expenditure, regardless of whether it's purchased with cash or debt.

I guess not all capex is spent on assets that depreciate. When SBUX had to pay a big legal settlement, or when CVA had to pay for unplanned repairs to a waste-to-energy facility, they didn't get anything for their money, and it couldn't be counted as depreciation in the future, so it had to be recognized immediately?

If say, BA hadn't signed a contract with the union in Washington and had built a new factory in Missouri instead, when would the cost of building the factory show up in earnings? Would it not show up in the earnings statements until it was finished, appraised, and started depreciating?

KMI declared a $0.42 dividend on $0.28 EPS, but from $0.55 of FCF available for distribution. If I look at FCF, it's well covered, but if I decide to ignore EBITDA and count depreciation as a real cost, then I should button my wallet, I'm told. However, then I feel like I'm counting the cost of pipelines twice, through depreciation (non-cash) and interest on debt (cash).

The question is relevant to offshore drillers because they have large fleets paid for with debt, and not all the same age. SDRL has levered up to buy a large fleet of $600m+ drillships, and therefore surely has a high non-cash depreciation expense. The depreciation expense would not be the same for all drillers. Diamond Offshore (DO) has the oldest fleet in the industry (and the lowest leverage), and therefore should have a lower depreciation expense. According to Finviz, SDRL is cheaper on a P/E basis than DO. However, I'm guessing that on an EBITDA basis SDRL might actually be more expensive than DO. A comparison of EV:EBITDA on Seeking Alpha strongly confirms this theory.

Again, thanks for the responses. I've never taken an economics or finance class, I'm just learning by investing.

[Dividend Watcher]

I'm glad some of you got something out of it. I have to add that I'm not a CPA nor a tax expert. All I know is self-taught with lots of reading and testing of the numbers. Actually, the more I re-read what I wrote, the more I could expound on what I had written and hoped it wasn't too simplified to explain what I was trying to say. That being said, let's see if I can explain your further questions:

(04-21-2014, 08:21 PM)Concasto Wrote: Depreciation may be a non-cash cost at the time earnings are reported, but if an entire fleet of trucks is wearing out, that non-cash cost turns into a real one(as they have to replace the trucks), so its good that the accounting takes this into consideration along the way.

First off, they (or any capital items) are usually placed in service and depreciated over a staggered time frame. Using the FedEx example, if they needed 1,000 trucks, they might purchase 100 every few months. This can spread out the cash outlay over several years and when the oldest are worn out, they start over again. This can produce a level and predictable use of capital that management can plan for. Smart management. Remember, these would be expensed using depreciation and, being productive assets, provide a steady stream of income greater than the depreciation expense covering those costs (cash or non-cash).

(04-21-2014, 08:21 PM)Concasto Wrote: I am not expert, but pipelines seems to be allowed excessive depreciation on their assets. DW, can you comment?

No, I can't explain your perception nor have I analyzed pipeline companies in particular nor the majors in such detail. Probably the pipeline depreciation is also including things like site prep, permitting & approval costs, site-specific insurance above general business insurance, etc. These are all things generally required to be amortized (depreciated) along with the actual costs of a physical asset.

Since I brought up a new term, I probably should explain it. Amortization is the proper term to use when depreciating non-tangible assets; patents, intellectualy property, digging the trenches for the pipe (afterall, once the pipe is buried, where is the hole?), etc. Depreciation is generally reserved for tangible items; trucks, factories, office furniture, etc. They are generally treated the same way as far as financial statements are concerned though each may have their own line item.

So, it may not be "excessive". The IRS and the SEC frown upon using depreciation/amortization that isn't justified. That's called cooking the books.

(04-21-2014, 08:21 PM)Concasto Wrote: I am a fan cash flow from operations and free cash flow, which is the cash flow from operations minus the capital expenditures. Unfortunately all these metrics have hundreds of possible variables, so listening to conference calls and going over company presentations and of course watching dividend growth are all important.

CFO and FCF are important. They can tell you whether there is enough cash to cover the dividend today. Keep in mind, though, that as Buffet said, those depreciation costs represent real money whether it's cash spent today or a couple years ago. They may have to spend it again.

(04-21-2014, 09:31 PM)earthtodan Wrote: So to pull the main point that addresses my confusion, the purchase of a new fleet would not be counted against earnings for the quarter in which it was purchased? That's surprising because I keep hearing about EPS figures being thrown off by one-time capex. But it sounds like what you're saying is that only the balance sheet gets affected by a large expenditure, regardless of whether it's purchased with cash or debt.

You are generally correct in your first sentence. I don't know where you're hearing all these reports of earnings off because of one-time capex, I hardly ever hear it. If it's 'talking heads', consider it bullsh*t or else management is trying to cover their a$$ for gross mismanagement. There are exceptions but it's few and far between in larger corporations.

There is a section of the tax law that allows you to deduct most or all of the allowable depreciation in the year it is placed in service. Section 179 of the tax code addresses this. I hardly ever hear large corporations taking advantage of it except for small capital purchases. The office furniture is an example. Large corporations have other ways to manage capex and depreciation that are much more flexible. Small businesses often use it extensively -- especially those that replace things like computers regularly -- or want to shield a particularly good income year to lower their tax bracket.

(04-21-2014, 09:31 PM)earthtodan Wrote: I guess not all capex is spent on assets that depreciate. When SBUX had to pay a big legal settlement, or when CVA had to pay for unplanned repairs to a waste-to-energy facility, they didn't get anything for their money, and it couldn't be counted as depreciation in the future, so it had to be recognized immediately?

For things like lawsuits, they are generally "reserved" over the period of time the lawsuit is going on (this can be years) and at the amount that managment and their lawyers estimate it's going to cost them. They add some every quarter to lessen the blow or all at once. This is how management can "manage the earnings" and also applies to depreciation. So, in essence it's expensed over time. Dig into BP's financials and you'll find this information.

Using my formulas above (since it's easiest to follow that way), liabilities are increased in a separate account. To keep the equation in balance, either assets have to go up or equity goes down. We know assets didn't go up (they're being sued) so equity had to go down. Using my relation of profits=equity to tie them together, then expenses had to go up for equity to go down. Using $1000 for the lawsuit as an example:

Assets (same) = Liabilities (up $1000) + Equity (down $1000)
Profits (down $1000) = Income (same) - Expenses (up $1000)

They may do this every quarter for several years

As for CVA, repairs are expensed when incurred. Upgrades are capitalized.

(04-21-2014, 09:31 PM)earthtodan Wrote: If say, BA hadn't signed a contract with the union in Washington and had built a new factory in Missouri instead, when would the cost of building the factory show up in earnings? Would it not show up in the earnings statements until it was finished, appraised, and started depreciating?

It wouldn't show up in earnings until the factory was placed in service. During the construction phase, costs actually paid may be placed in an asset reserve account. Once it is placed in service, all the associated costs are totaled up and placed in the asset list. There is no appraisal. You can only deduct actual costs which can include a lot beside building materials just as in the pipeline I mentioned above. Then it's depreciated over its useful life.

(04-21-2014, 09:31 PM)earthtodan Wrote: KMI declared a $0.42 dividend on $0.28 EPS, but from $0.55 of FCF available for distribution. If I look at FCF, it's well covered, but if I decide to ignore EBITDA and count depreciation as a real cost, then I should button my wallet, I'm told. However, then I feel like I'm counting the cost of pipelines twice, through depreciation (non-cash) and interest on debt (cash).

No, if you add back depreciation, you'll have a good idea what cash management has available to pay the dividend. EBITDA includes (correction: EXcludes - before) taxes which is a cash expense so I wouldn't ignore that. When you bring debt into the equation, you've now changed what we're talking about. Debt is a separate issue from the depreciation.

The principal part of the debt payments is a Balance Sheet transaction only. You're taking your cash (asset) to pay the bank or bondholder loan principal. This lowers your liability (loan balance) and increasing your equity (you now own a bigger portion of your asset). Interest on the debt is deductible when paid. Both P&I are cash transactions but have nothing to do with depreciation and the book value of your asset. You need to keep the two separate in your accounting mind.

Now, if you need that loan to purchase that asset, smart management will structure the debt to approximate the useful life of the asset. This way, when the asset is worn out and you want to get rid of it, the loan is paid off and you can sell it outright. If you still need that truck, wash, rinse and repeat the whole process. REITs are pretty transparent about trying to do this.

(04-21-2014, 09:31 PM)earthtodan Wrote: The question is relevant to offshore drillers because they have large fleets paid for with debt, and not all the same age. SDRL has levered up to buy a large fleet of $600m+ drillships, and therefore surely has a high non-cash depreciation expense. The depreciation expense would not be the same for all drillers. Diamond Offshore (DO) has the oldest fleet in the industry (and the lowest leverage), and therefore should have a lower depreciation expense. According to Finviz, SDRL is cheaper on a P/E basis than DO. However, I'm guessing that on an EBITDA basis SDRL might actually be more expensive than DO. A comparison of EV:EBITDA on Seeking Alpha strongly confirms this theory.

Don't base your investment decisions solely on EV:EBITDA. You're ignoring a big part of what management uses to engineer their financial position. It's a nice benchmark to add to the analysis but isn't a full picture of the business just as P/E, FCF and debt/equity don't tell the whole picture.

Am I forgetting anything? Yeah, lots. Hope this gives you a little more understanding.

[Dividend Watcher]

After re-reading this I think I now know why my classes required frequent breaks to keep them awake.

[earthtodan]

(04-24-2014, 09:26 AM)Dividend Watcher Wrote: After re-reading this I think I now know why my classes required frequent breaks to keep them awake.

I've had plenty of coffee today, so no problem there. I've been on the road for a couple of days so I haven't had time to get back to this discussion, but it has changed my understanding of depreciation. The P/E ratio isn't strictly the comparison of price to earnings, but an attempt to smooth out or amortize the affect of large expenditures over time.

I realized in my KMI example above that interest is a separate cost on top of depreciation, and EBITDA isn't that relevant to the discussion since that includes cash expenses, so I made it too complicated. However I wonder over how long the depreciation on a pipeline network is amortized. It's probably quite a while.

The explanation of lawsuit accounting makes sense. BAC has been setting aside a lot of money to cover possible legal expenses and fines, and it was deducted from their recent earnings statement even though none of it has been settled.

So, a company makes a big expenditure, the balance sheet stays the same (i.e., balanced), the debt goes up and the reported earnings are held down while the expense is recognized as depreciation over the expected life of the asset. I'll try to remember that and see how long it takes to become intuitive.