Hacker News Reader

Not counting depreciation as part of your loss in a business is about like all of the Uber drivers who don’t take into account the wear and tear on their cars when calculating how much money they are making. Depreciation is a real expense. If you depreciate an asset down from $1000 to $0 in 3 years, you’re accounting for the fact that in three years you are going to have to replace it.

There is a reason we have GAAP, to keep crap like WeWork’s “Community Adjusted Ebitda” and Uber from saying that their financials really aren’t as bad as they look if you ignore a dozen expenses.

> If you depreciate an asset down from $1000 to $0 in 3 years, you’re accounting for the fact that in three years you are going to have to replace it.

but does a company have to be "honest" about such depreciations? What if the asset isn't actually losing value at the stated depreciation rate? Then at the end of the depreciation period, the company may still extract the residual value by either selling or continue using the asset. Does that somehow then count as revenue? Or did they just magically managed to tax dodge using a fake depreciation schedule?

The asset will eventually have to be replaced. In the long term, it doesn’t make a difference - ignoring the time value of money.

Theoretically, the only reason we have depreciation expenses at all instead of just expensing the entire cost at once (cash accounting) is that we decided that accrual accounting was more representative than cash based accounting.

I am not an accountant. I’m an MBA dropout after almost finishing.

Also, taxes. If we were able to fully expense capital plants and equipment, there wouldn't a single company paying taxes, they would continually purchase more to offset profit, eliminate taxation, and accelerate growth.

Is that a bad thing?

As far as governments are concerned, yes, as they will never get tax revenues from companies if they are allowed to fully expense large purchases.

Accounting was a decade ago for me, but as I recall if you sell an item you've depreciated to $0, you will incur a "Gain on Item Sold" which is taxable.

This might be different in various countries, but in general companies cannot play games with depreciation. The financial audit is done according to certain rules and common practices. In reality some of these investments might be hard to properly calculate and audit, but simply reporting them as zero is malicious. EBITDA numbers should provide the insight into profitability without hiding anything.

It's generally fine to mark down your assets faster than they actually become obsolete, so you can take all the depreciation in year 0 if you want. Marking them down slower and overstating profitability is not so cool.

There are rules in place which assets depreciate at which rate. Otherwise companies could just cheat their way through accounting.

On a cashflow basis, you incur 100% of the depreciation at the moment you pay for it.

But wouldn’t your profit be awesome the years you didn’t have to claim depreciation?

Hence you’ve invented EBITDA and EBITDA less capex :)

Damnit that was a major self own I just did wasn’t it?

It just means you’ve internalized the concepts!

Imagine you're a growing startup, and you have a yearly recurring investment (you're growing, after all!) of $1000 that's linearly depreciated over 5 years.

Let's say your income is 1100$ each year.

Your profit, according to accounting, would be 900$ for the first year, counting only $200 of the investment, then for the following years you'll see a profit of $700, $500, $300, and $100, as the investments accumulate. Oh no! A downward trend!

The cash flow, however, will simply show $100 profit each year.

Which one is more representative of the growing business with recurring investments?

That’s exactly the story that YC darling DropBox has been telling over the last decade. They still aren’t profitable and according to their own disclosures, have no idea when they will become profitable.

Just as as microcosm, how many YC backed companies have reached profitability? Only two have gone public - DropBox and PagerDuty.

I'm not sure that I understand your scenario. If you actually need to replace the equipment after 5 years, the accounting approach seems like the most valid approach.

For my understanding, do you mean that cash flow would show $100 profit each year on the 5th year, whereas in the first year it would show $900 profit? Thanks.

In year one, you’re generating $1100 of income with $1000 of capital invested (that will last 5 years).

By year five, you’re generating $1100 of income with $5000 of capital invested (that will require $1000 each year to keep up).

You’re _way_ better off in year one here, so it seems the GAAP approach is actually showing the decline accurately. This isn’t a growing startup, it’s a startup needing more equipment to make the same money each year.

You're right, growing capital but not growing income... I made a poor example.

(Imagining a bottom pricing scenario, while keeping a positive cashflow.)

Good point. At year 6 this logic breaks down. (Then again, no longer a 'startup' at that point.) Better hope the replacement equipment is double worth it's money. :)

Should make some spreadsheets with more scenarios.

Amusingly, assuming the example company's pricing remained the same throughout rather than cutting prices each year, the client base ends up fixed, the clients get 5x better service by the 5th year, while accounting shows a decline in performance.

External perception of business health (5x better service!) unintuitively is masking the the accounting reality here (worsening returns).

(Now look at all these cloud services giving better services and pricing every year, seemingly very healthy...) :)

The thing is that depreciation is not a cash expense, companies will push the depreciation to the highest reasonable amount to reduce the amount of tax on their profits.

A commonly used valuation technique (discounted cash flow or DCF) relies on projected free cash flows and adjusting for expected CAPEX.

Also there’s GAAP depreciation and tax depreciation, and firms will try and use different depreciation schedules between the two in order to create deferred tax liabilities which boost cash flow early on.

I know about WeWorks deceptive accounting/reporting practices, but not sure what similar thing Uber has done when it comes to accounting?

https://investor.uber.com/news-events/news/press-release-det...

We define Adjusted EBITDA as net income (loss), excluding (i) income (loss) from discontinued operations, net of income taxes, (ii) net income (loss) attributable to non-controlling interests, net of tax (iii) provision for (benefit from) income taxes, (iv) income (loss) from equity method investment, net of tax, (v) interest expense, (vi) other income (expense), net, (vii) depreciation and amortization, (viii) stock-based compensation expense, (ix) certain legal, tax, and regulatory reserve changes and settlements, (x) asset impairment/loss on sale of assets, (xi) acquisition and financing related expenses, (xii) restructuring charges and (xiii) other items not indicative of our ongoing operating performance.

Thanks. I am not a financial professional but do run a medium sized business and have a lot of respect for being cash flow positive.

I agree with depreciation and other non cash expenses obfuscating P&Ls, but can we agree that in an article focused on securing debt financing, which will incur real interest expense, the term “free cash flow profitability”, “loosely” meaning “profitability” is a little misleading?

> I agree with depreciation and other non cash expenses obfuscating P&Ls, but can we agree that in an article focused on securing debt financing, which will incur real interest expense, the term “free cash flow profitability”, “loosely” meaning “profitability” is a little misleading

It depends on the industry. DO is in the capex heavy industry so depreciation is not a funky accounting cost, it is actually something that's applicable to the vast majority of their assets.

Yeah, this was my first thought. Technically if you go back in time and model past Cash Flow Statements, you could layer on what you think the real go forward infrastructure cost is to their CFS.

If they are consistently laying out capex, it actually could be very reasonable to just use CFS and not opex.

Software is not typically depreciated though. DO is writing code now that will last for decades but they have to take the accounting hit for it in one year.

Code rarely last decades without maintenance and complete rewrites over time - unless you’re the government.

I maintain some LOB PHP5 apps, that started as PHP3 - these internal systems are over 10yr old. Just chugging along, producing value for the business. But yeah, pretty rare stuff, and everything I see is small/internal.

Plenty of enterprises are running decades old code that powers their core business.

At least in my small European country, I believe you can use a deprecation schedule for software as long you buy it externally, instead of creating it in-house.

The same is true for other creative work/IP. The reasoning being, I believe, that it is hard to appreciate (as in: set a monetary value) for such work. If you buy, you’ve given the open market a chance to work its magic and the price is believed to be closer to something like truth.

This might go a long way towards explaining why company management in Europe often have some kind of mysterious love for vendor supplied software and doesn't see any in-house development as a competitive advantage or worth spending on maintaining.

Agreed that free cash flow can be misleading some cases, and digging a bit deeper helps us understand what those cases are.

Free cash flow and net profit should equal each other if summed up over the entire lifetime of a company.

The trickiness of course comes in the timing of when revenue and expenses are recognized. Whereas a large capital investment hits free cash flow the same year, it only impacts net profit over the course of its depreciation.

Free cash flow can be misleading for companies that require a disproportionately large capital investment upfront but negligible expenses thereafter. This can be illustrated by taking an edge case to the extreme: Let’s say a company required a $1,000,000 capital investment in Year 1, $0 in subsequent expenses, and brought in a piddling $1 in annual revenue until it shut down in Year 10. The company would have been free cash flow positive in Year 2 through Year 10, which would be misleading because in fact the company was never profitable and also had negative ($999,999) free cash flow in Year 1.

> but can we agree that in an article focused on securing debt financing

We could also agree that its a tech-blog article, which means 99% of the people reading it simply do not care about financials, other then "huge raise! wow, revenue! great profitability potential".

That’s what I was taught when I took accounting. Every measure has drawbacks, so you can’t rely on just one.

Income statement, balance sheet and statement of cash flows together give you a pretty complete picture.

Almost all companies re-invest at the rate of depreciation, which is why depreciation rates are a good approximation of cash you are going to have to spend. Move on down to "investing activities" on your cash flow to see all that "non-cash" depreciation bleeding out. Better yet, check out a balance sheet. Net income = change in balance sheet YoY.

Which makes perfectly sense. A company can be cash positive and profitable, cash negative and profitable (worked at a place like that, didn't end well), cash positive and not profitable (at least that company would be by default alive, and if revenue is growing quite healthy IMHO) or cash negative and not profitable (in that case death could happen any moment).

I understood the OP that mixing cash and profitability in one term is wrong. So yeah, I agree that the CEO told BS.

The rules around deprecation and so on are at least intended to give a “better” picture about the health of some business, so it’s somewhat strange when we start going back the “lesser” data of cash flow.

As just one example, if you pay for supplies in one accounting period, but only sell (and get paid) in the next, cash flow will fluctuate widely, even if your business is entirely stable.

"consider cash flow positive a better representation of actual profitability than the actual net profit/loss reported on the P&L"

Wow. What a sad indictment of modern accounting practices that its accepted practice to create 'noise' to prop up a P&L (or avoid paying tax).

I suppose you could do this with household expenses, but if you don't look at P&L in a household you would be ignoring things like credit card debt or car loans.

This kind of debt in either a household or a company represents risk and shouldn't be overlooked, IMHO.

> What a sad indictment of modern accounting practices that its accepted practice to create 'noise' to prop up a P&L

Income statements include things like depreciation. That's real, in a sense. It's future CAPEX. But for the present health of a business, particularly a levered one, cash is king.

One way to think of it is in time frames. Cash flows are immediately relevant. If they're out of whack, it's an urgent problem. (They're also the most difficult to mess with.) Income statements are longer term. An unprofitable, cash-generating business may have structural issues with its PP&E, or it may be overleverd, or a bunch of options might have been issued or exercised. Balance sheets are the longest-term statement. (They're also the easiest to mess with.)

Keep in mind, too, that income statements for investors are different from those prepared for the IRS.

"Income statements include things like depreciation. That's real, in a sense. It's future CAPEX."

Depreciation is past capex, not future capex. You could argue that the two are equivalent, because the assets being depreciated will need replacement in the future. But:

1. The assets' useful life may be longer (or shorter) than the depreciation schedule.

2. The future replacements may cost less (or more) or just not be needed at all.

> Depreciation is past capex, not future capex

Technically (from an accounting perspective), yes. Practically (from a financial/economic perspective), not quite.

Past capex is a sunk cost. Investors care about future cash flows. If you strip out depreciation, you’ll be surprised by the bill when your machine breaks down. (You may still be surprised. Life is unpredictable and depreciation schedules are an estimate. But that’s one of the problems the concept tries to solve.)

It's not really a "sad indictment", there's nothing shady about re-investing your money into the business. It just so happens that if you are a fast growth company doing that, it is no longer beneficial to use the same tools to compare to traditional large established slow growth companies. Amazon fairly and legally and morally (mostly) re-invested their positive cash flow back into their business, but looking at their P&L you can't tell if they are a year 0 startup or an amazing giant company - the cashflow, however, reveals a different story.

It's perhaps not nefarious. A lot of GAAP and the standards came about when manufacturing was a lot more prevalent. Concepts like depreciation make a lot more sense when you mentally place yourself in the 1920s.

Depreciation still makes plenty of sense today. A company with very little capital equipment will simply not have a lot of depreciation. Digital ocean as a cloud hoster has a lot of capital equipment (servers etc) to worry about as well.

Do they run their own servers? I thought they were wrapping t other big cloud providers.

DigitalOcean owns their own servers and networking infrastructure inside existing datacenters.

https://www.quora.com/Does-DigitalOcean-have-its-own-datacen...

Many analysts will capitalize R&D spend.

Which makes a lot of sense, especially when it comes to compensation where not capitalizing end would lead to incentives to not invest in it even if it is profitable. It doesn't make sense to expense it right away when it's something that, like capex, is meant to essentially purchase assets that can pay off later. If those revenues fail to materialize, then the bottom line takes a hit then, just as a company would take a hit if expenditure on a factory proves to be worthless. The goal of this sort of accounting is to reflect economics better, we have statements of cash flows to keep a track of cash and you're more than free to use that in valuing a company (and many analysts do ignore everything but cash)