Doesn't this just amortize out to be roughly the same amount of deduction over the long term?
All the big companies mentioned should be relatively unaffected over an N>5 year time period. Also this was something that's been in the works for years so their accountants should have been planning for it so it wasn't a financial shock (and company financials seem to indicate no such shock).
But more importantly, the article claims it was used as a tax shield to grow.
"Basically, as long as spending counted as R&D, companies could report losses to investors while owing almost nothing to the IRS."
"Once those same expenses had to be spread out, or amortized, over multiple years, the tax shield vanished. Companies that were still burning cash suddenly looked profitable on paper, triggering real tax bills on imaginary gains."
1: https://www.investopedia.com/terms/t/timevalueofmoney.asp
I get that this is bad for the VC monopoly bucks scene, but they were already down for the most part. If the changes are as the article alleges than all these big tech companies that are posting huge layoffs should mostly be fine because it's not a serious change from status quo for them.
This tax change just made it worse.
It hurt small businesses that were slightly profitable. No one else. VC shops aren’t profitable anyway, so no taxes to pay. Microsoft took a 4 or 5 billion dollar write off, but they can literally write a 5 billion dollar check.
The issue is that the IRS wants you to pay them today on profits and cash that literally don’t exist. You make $1M in revenue and pay 5 developers 200k/year? You have no money left at the end of the year, but you pay taxes as if you profited about 900k.
With steady enough employment numbers, sure. Google has a weird one-time cost where they get hit with extra taxes at 80%, 60%, 40% and 20% of their employee's salaries for five-years and then it's all balanced. You can turn the money Google needs to borrow (or not invest) at some interest rate into a known number.
Any startup that is cash poor and especially one that is growing struggles. In year 3 you get to write off 20% of year 1's salaries, 20% of year 2's salaries and 20% of year 3's salaries.
Most companies in question don't fit these criteria. They are either large public companies subject to the reactions of the market to quarterly earnings, or small private startups that have limited cash (a runway of far less than 5 years) and are facing a perfect storm of a historic rise in the cost of capital coinciding with this change.
In either case, their cost of labor just went up by a lot and will continue to cause layoffs, labor market shrinkage, and diminished ability to develop new products.
Yes, but if your business is not yet profitable, having to pay tax on money you don't actually have in the bank (because expenses exceeded revenue during the year) will cut into your runway, perhaps to the point that your company might not exist in five years... or even two or three.
It’s a pretty bad article general and to blame the law change on this is all kinds of disingenuous: “It’s no coincidence that Meta announced its ‘Year of Efficiency’ immediately after.”
None of this adds up. You're saying, the legislators were trying to cheat and because it's a "common tactic" that kind of cheating is somehow good, but it's bad when the cheating doesn't go through?
On the other hand, being a common tactic implies that the possibility of it remaining in the books was well understood, and the declared "expectations" carry zero weight as evidence, even less than zero when coming from politicians.
Legislation like that has far reaching consequences and pretend "surprise" just confirms the intent behind it. It's only prudent to assume that we have a common tactic case of throwing sheet at the wall to see for how long it'll stick. If there's no backlash the "tactic" will remain there forever.
As another example of the same common tactic, consider the fact that all popular browsers have been used as Trojan horses into the users' local networks for like forever. At some point back in 2015 somebody objected so the browser makers started talking about fixing the problem but then stopped talking without fixing it because public opinion moved on to other areas affected by abundant sticky materials... Thus, that particular sheet remained on the wall for another 10 years and counting, and the story may repeat itself again.
It's not cheating, it's playing by different rules to get most of what you want/need done and then sometimes those that played and gambled were intending to, or hoped to, make the changes later that require rules. Their hope is that 60+ Senators would be onboard for those changes because they (those that gambled and pushed the budget bill thru) managed to get what they wanted at the expense of #$%#ing something up that most others would then be willing to fix/address.
Changes to Section 174 happen rarely and are not a “common tactic.” Changes to tax policy in general are common, especially in the reconciliation process. They can have unforeseen side effects. As well as side effects that are foreseen but considered more acceptable than other side effects.
Not quite the sentiment you intended.
I don't think GP made any kind of value judgment either way; they were just stating how things seem to usually work.
i.e. some humans get the same tax treatment as humanoid robots, while LLMs ("AI") are always deductible as op-ex, regardless of function.
Draft 2025 spending bill in Congress would revert Section 174 changes for 2026-2029.
> And so, on schedule in 2022, the change to Section 174 went into effect. Companies filed their 2022 tax returns under the new rules in early 2023. And suddenly, R&D wasn’t a full, immediate write-off anymore. The tax benefits of salaries for engineers, product and project managers, data scientists, and even some user experience and marketing staff — all of which had previously reduced taxable income in year one — now had to be spread out over five- or 15-year periods.
Should they be able to expense all of those items that provide value for multiple years in a single year?
Does software development provide value exclusively in the year it's done? Or over multiple years?
The only possible justification for the Section 174 R&D changes is that employees working in R&D theoretically are producing something which does have a resale value, so there's a small tax dodge enabled by direct-expensing your R&D costs but then ending up with an infinitely-copyable asset that came out of it.
If that's what you're saying, then I'd reply to that argument by saying that paying humans to design new things has historically been a business strategy that the government has wanted to incentivize in a way that buying and holding physical assets has not been. I've seen no justification for the government deciding that from 2022 on we should actively discourage R&D, it just seems to be a mistake.
Removing a specific tax exemption to create a level playing field isn’t discouraging R&D.
That’s the thing, every year such exemptions exist the US taxpayers are handing out money. Just because we subsidize say EV’s or Corn doesn’t mean that’s the baseline forever more.
Restaurants weren't competing with R&D-heavy corporations in any way. R&D-heavy corporations competed with each other, on a level playing field where all of them can build new stuff without having to pay taxes on negative income in their early years.
The only change this has made is un-level the playing field in favor of old, established corporations that already have the revenue streams in place to fund their new R&D projects.
Taxpayers who end up with the bill and every company is competing for workers, office space, etc. Incentives across decades shift what people study, what business get created, etc. R&D sounds great abstractly, but it’s not some panacea where unlimited funding results in pure gains.
The economy is generally more efficient without central planning, and dumping money into anything that can be classified as R&D is simply inefficient.
My company is all-remote and none of us would work for a company that isn't doing R&D. Most of an entire profession now has to be amortized over 5 years.
> The economy is generally more efficient without central planning
The old tax code isn't "central planning", it just had the very reasonable property that the government wouldn't force you to pay taxes on a loss.
This scenario [0] is now possible. It wasn't before. That is a catastrophic level of stupidity, and you can't justify it with invisible-hand nonsense.
So you’d just be unemployed for the rest of your lives? That’s a possible edge case not worth adjusting the tax code for, but it seems unlikely.
> wouldn't force you to pay taxes on a loss.
R&D is an investment, you only pay taxes if the rest of the company is profitable.
If your company is spending 1M / year on R&D and not adding 800k in long term value then in theory you’d be correct. But at that point you either aren’t doing R&D, or are doing such a poor job of it that the government shouldn’t be encouraging that activity.
As a practical measure it’s really not. The transition is difficult for existing companies, but a future startup is going to be minimally impacted.
Year 0 you’re unlikely to have any profits, future years you have multiple years of R&D to offset with.
But let’s assume the worst case. Taxes are 21% of profits and at minimum deduction 20% of R&D so the theoretical maximum distribution is 0.8 * 0.21 = 16.8% increase in R&D expenses if profits = R&D year 0. But that maximum case is only year 0, you’d be able to fund R&D with those same profits and easily be profitable after that.
If profits where say 40% of R&D in year 0 you’d have to pay 16.8% of 40% so an increase is only 6.72% hardly likely to tank the business if it’s already generating that kind of income year 0, and again after that point you’ll deduct for multiple years.
More realistic numbers are going to be really low multiples here, more importantly they represent significant investments not operating expenses.
You're only unlikely to have no profits if you have no revenue. And you only get to break even 5 years in, which most startups will never reach.
In practice what is likely going to happen is that we'll see more and more startups deliberately avoid revenue in the early days. More and more free tiers followed by rug pulls when revenue actually becomes an asset rather than a liability.
There is no unplanned economy, only different outcomes from better or worse plans. And I'm having a hard time imagining a worse plan than one that intentionally disincentivizes businesses from adopting a sustainable business model early in their lifetime.
It’s much easier to have revenue than profits, set the price lower and suddenly zero profit. Some company avoiding profits because of the 21% tax on profit like that would be mathematically dumb.
> There is no unplanned economy, only different outcomes from better or worse plans. And I'm having a hard time imagining a worse plan than one that intentionally disincentivizes businesses from adopting a sustainable business model early in their lifetime.
There’s zero advantage to avoiding revenue or profit here. You’re tilting at windmills.
You simply need less investor money for R&D when other parts of the company are profitable. As to central panning, the mistake you just made is mitigated when many people are all independently making plans. Governments always need to get it right, the market is fine if some people get it right and therefore can reinvest in their success.
When the non R&D portion of the business is profitable they should start paying taxes. Assuming a company isn’t miss classifying operations as R&D it shouldn’t be a major issue.
This will of course discourage “riskier” startups and dampen innovation and give more power to profitable incumbents who will have less incentive to innovate. (Perhaps the result of this looks like Europe?)
You’re only paying taxes if the business is profitable ignoring investments like R&D spending.
Section 174 specifically made those R&D costs “ignorable” from a tax standpoint. When it ended R&D costs could no longer be used to offset income.
As to my other point, the highest risk category of startup has zero customers for years they also have zero revenue, zero profit, and zero taxes to pay here. On the 5th year they can deduct R&D from each of those years making the net effect on them minimal vs a startup with profits on year 0.
Big fat "citation needed" there. I know you chose the term "central planning" to try to invoke the communism boogeyman, but overall, free markets do not exist, and have never existed. Governments constantly use various levers (taxation being one of them) to encourage or discourage certain kinds of business activity. This is nothing new, and I find it laughable to suggest that this kind of thing should be done away with entirely.
Markets operate on revealed preferences, which is just a massive advantage in terms of giving people what they want. There’s definitely a role for governments in economies around information asymmetry, safety, etc, but allocation of resources specifically doesn’t work well.
If the end result of removing this exemption is that there is less R&D done in the US, then yes, empirically, removing the exemption discourages R&D. Assuming the mass layoffs were indeed fueled by the removal of this exemption (I don't know if the article is correct or not), then it is reasonable to assert that it is true that removing the exemption has reduced the amount of R&D done.
Or, you could also say that the "default state" is some low level of R&D, and the tax exemption encouraged and incentivized more of it.
Either way you slice it, though, the status quo prior to 2022 was some level of encouraged/incentivized R&D. That status quo changed to encourage/incentivize less R&D, and companies have followed these lack of incentives and have fired a lot of their R&D staff. Is that a good thing for the US? I can't see how it could be.
Not clearing a road means fewer people use it, but you not going out with a shovel to clear a public roads isn’t you discouraging their use nor is you canceling your plans to clear said roads.
Having zero subsidies is the default situation.
If Amazon delivered you a TV yesterday that doesn’t suddenly become the default where you can expect another one today and every day after that.
The US government does a new budget every year, making every year a new ballgame.
And my point about there being no natural state of subsidies is more important.
So if your argument is some subsidy will probably happen next year sure, but individual subsidies change over time. No specific subsidy is the default.
For any specific situation the default is no subsidy.
With millions of situations some of them are not going to be at the default.
In 500 years will some specific things be subsidized? Vs in 500 years will something be subsidized?
Under GAAP, construction labor is not immediately deductible as an expense in the year it is incurred if it relates to the construction of a long-term asset (like a building). Instead, it is capitalized as part of the asset's cost and then expensed over time through depreciation. Only labor costs not tied to asset creation (e.g., routine maintenance) are expensed as incurred.
Construction labor is generally not deductible as an expense in the year incurred if it is related to the construction or improvement of a capital asset (like a building). Instead, under the U.S. tax code (IRC §263A), these costs must usually be capitalized and recovered through depreciation over time. Exceptions may apply for certain small taxpayers or repairs.
§ 1.263A-1.a.3.A indicates that it's in scope: Real property and tangible personal property produced by the taxpayer
§ 1.263A-1.e.2 specifies that Direct Costs are subject to capitalization: Producers. Producers must capitalize direct material costs and direct labor costs.
(I'm just a taxpayer, not a tax lawyer or even an EA or CPA.)
What tax code references or treasury regulations did you find to support your belief that construction labor can be expensed in the year performed?
Same as if they sell the software, either as a copy or ownership.
But not being able to take salary as a business expense seems like as thing that would happen if software in and of itself has value, which is largely does not.
To me it seems like a thing that just wouldn't happen. Forget software.
Say you own a McDonald's, and as part of your operations you have some people on staff to take orders, prepare food, and clean the bathrooms. Why are their wages not a deductible business expense?
If the answer is "they are, don't be stupid", then... what exactly was the R&D tax break?
Imagine a restaurant spends money on employees to build 100 tables, 500 chairs, etc. Those tangible goods would be capital assets, so the labor costs of building them would also be capitalized.
This change to the tax code is just bringing the tax treatment of software development in line with how every other industry is treated. IOW, it was closing a loophole. A very valuable loophole, whose beneficiaries used it to get filthy rich, and bragged about how their industry was so much more valuable than everything else, even though a lot of that value was due to the exception software was getting in the tax code.
Notably, in the current version of the budget as of 6/6, the loophole is temporarily coming back, though given the Musk-Trump feud, it's very possible it will get pulled again to try to mollify the hardline deficit caucus.
Given the choice, Amazon would rather spend 100% of its profits on itself than allow any of its profits to be paid out in taxes. Section 174 was implemented without a minimum tax on corporate profits before voluntary deductions such as research. Therefore, it’s exploitable and all companies ought to hire and fire staff to ensure their profits show as 0%.
This tax code defect is now closed by accident, but could have been done much more intelligently than it was. Oh well.
(EDIT: My first sentence is potentially confusing when I reread it later. To restate: section 174 was defective as implemented due to the uncapped 100% deduction, but the concept of a significant research exemption is still excellent. Just need to close the effective 0% corporate tax rate loophole.)
What this change effectively did was make software developers significantly more expensive, without increasing the amount those developers get paid.
I think you're referring to Nordic countries which consistently rank as the happiest countries and also have relatively high tax rates (4 of 5 Nordic countries rank in the top 11 tax rates globally. Norway has oil.) The high taxes that "make everybody poorer" also fund extensive social services that contribute to happiness.
However, this conversation is about making (a class of) workers poorer by using tax policy that puts downward pressure on their salaries. Tax revenues will stay the same, so social services will not be increased. Economic inequality increases because the workers became poorer, the C-Suite and Board Members don't.
Yes it sucks for developers, but does it make any difference for any other employee? Why does Joe’s plumbing have to pay those taxes, but Jane’s AdTech company doesn’t?
Sure, there are benefits to investing in R&D in general, and tech has fueled a lot of growth, so incentivizing it has likely paid off for the whole economy. But will that forever be true? Maybe?
Why do I, the hardworking tax payer, have to subsidize Joe Plumber, who already has a big house with a pool?
But with the change, the cost of R&D employees is now only partially deductible (right now, you can eventually deduct the full amount over the course of several years), and software development has to be considered R&D.
And why is this bad, exactly? Money will be spent and will go back into the economy. Amazon will have to use the funds to build new offices, datacenters, do research, whatever.
And even if execs give themselves $10^11 USD in bonuses, they will be taxed as personal income, at even higher rates than corporate income.
I’m not sure what the answer is. The former is likely to drive some innovation, which I’m sure varies by company. Where the latter could also unlock innovation by giving the bottom-quartile of earners a chance to improve their situation.
Yes. Also, the salary will not go _only_ to highly-educated people. For example, if Amazon decides to build a new distribution center, it will employ blue-collar workers to build it, not software engineers.
> Or is it better for the money to go back into the economy through taxes, then disbursing the benefits to lower-income benefit programs?
No.
> I’m not sure what the answer is.
The answer is pretty clear: invest money into the private sector, rather than divert it into the Federal budget. Private actors are more efficient at allocating funds than the government.
I'm not against social spending, it's a necessary evil for any real state. Pure libertarianism leads to dystopian outcomes. But it should be understood that it's a very real artificial inefficiency that is imposed on the economy.
There are also situations where additional social spending is necessary, but they are VERY easy to detect: when your interest rate is near zero.
If you want a static economy that supports gradual decline (preferably with a mineral-based income stream), then a lot of state spending is fine.
State spending is not a panacea.
Growth has its own problems of course (I don't want to estimate the health impact of Coca Cola), but it's a prerequisite of a country not falling behind others.
“if we aren’t rich then no one else will be”
Imagine you are BigCarCo, you make cars. The salary for your factory workers that build cars to be sold is an expense, incurred in that year, to be matched against the revenues earned by selling those cars. But the cost to build the factory needs to be amortized over the lifetime of the factory - and that's true whether you buy a factory from BigFactoryCo or hire a bunch of people to build it.
Now, I'd argue that a) most software dev work is closer to the factory worker than the factory builder and b) the lifetime for most software is less than 5 years, but the idea that some cost of developing software should be amortizable is pretty reasonable.
Mostly developing software is about automating things that are expensive and slow to do manually. So, to stick with the factory analogy, it makes the factory a bit better and more efficient. If you stop doing that because it is too expensive, you fall behind with your factory.
Of course the whole issue in the US is that it outsourced much of what happens in factories to China and software has become one of the main things the country runs on.
Labor that operates the business day-to-day would be an expense, labor that creates a capital asset is more complicated.
I happen to think most employee time in software dev is more on the day-to-day operation side, and should be expensed, but I can see an argument that some should (or could) be amortized.
Big Beautiful Bill R&D Tax: Will tech go on a hiring spree again? - https://news.ycombinator.com/item?id=44028106 - May 2025 (19 comments)
The Consequences of Limiting the Tax Deductibility of R&D - https://news.ycombinator.com/item?id=43639202 - April 2025 (64 comments)
House restores immediate R&D deduction in new tax bill - https://news.ycombinator.com/item?id=39212650 - Feb 2024 (8 comments)
Ask HN: Best country to run a boostrapped startup from? (After Section 174) - https://news.ycombinator.com/item?id=39098371 - Jan 2024 (31 comments)
US tech innovation dreams soured by changed R&D tax laws - https://news.ycombinator.com/item?id=38988129 - Jan 2024 (3 comments)
Ask HN: IRS section 174 – cause of layoffs? - https://news.ycombinator.com/item?id=38957651 - Jan 2024 (21 comments)
Will US companies hire fewer engineers due to Section 174? - https://news.ycombinator.com/item?id=38931860 - Jan 2024 (37 comments)
Will US companies hire fewer engineers due to Section 174? - https://news.ycombinator.com/item?id=38870429 - Jan 2024 (20 comments)
IRS tax code change in Section 174: R&D is an expense - https://news.ycombinator.com/item?id=38642461 - Dec 2023 (23 comments)
New tax rules on R&D expenses may lead to layoffs for devs - https://news.ycombinator.com/item?id=38636866 - Dec 2023 (7 comments)
Tell HN: People laid off in my company due to IRS Section 174 changes - https://news.ycombinator.com/item?id=38633668 - Dec 2023 (6 comments)
Tell HN: Submit comments to IRS re tax treatment of software dev expenses - https://news.ycombinator.com/item?id=38120388 - Nov 2023 (225 comments)
Software firms across US facing tax bills that threaten survival - https://news.ycombinator.com/item?id=35614313 - April 2023 (981 comments)
Ask HN: How are you handling Section 174 changes for bootstrapped companies? - https://news.ycombinator.com/item?id=34627712 - Feb 2023 (187 comments)
https://hn.algolia.com/?dateRange=all&page=0&prefix=true&que...
What really changed things was the end of ZIRP [1] and even then it was opportunistic. Labor costs are a massive cost for tech companies. They have continually tried to suppress wages. In the 2000s, it was the anti-poaching agreement between Steve Jobs, Eric Schmidt and others. In the 2010s, high growth ahnd zero interest meant labor costs continued to balloon.
But then Covid came along and was a massive opportunity. A few companies may have needed to do layoffs but that created the opportunity for everyone else. Big Tech just went full Corporate America with a page straight out of Jack Welch: fire the bottom 5-10% every year. Call it "layoffs". It's a direct pay decrease for those who remain (who get assigned the work). Those are still there won't be asking for raises because they're now afraid of their jobs.
Very little of this was ever necessary. None of the big tech companies ever came close to making a loss. They've remaining insanely profitable, in total and on a per-worker basis. At different times Google's per-worker profit has approached or exceeded $1 million.
The other factor is these companies eventually reached their size limits where antitrust stopped them making any more significant acquisitions.
Consider the timing: this change came in 2017. Where were the mass layoffs in 2018? 2019?
Also, the 2017 tax cuts contained a massive tax holiday for the repatriation of foreign profits.
Mass layoffs are simply wage suppression. It's the end state for any company that can't keep growing the way the market demands: eventually it comes down to cutting costs to make those quarterly profit targets. And in that, they sow the seeds of their own demise.
[1]: https://en.wikipedia.org/wiki/Zero_interest-rate_policy
The bill passed in 2017, but the changes to R&D didn't kick in until 2022.
Big tech companies are both doing mass layoffs AND hiring. How does this fit the narrative that the tax change is at least in part responsible? The new hires still have the same deduction issue, right? So what impact does this really have?
Think of it this way: if this passes, will the layoffs end? Or reduce? Absolutely not. All this does is give line the pockets of shareholders. That's it.
I'm a big fan of tying certain benefits to NOT doing layoffs. This can include:
1. You get this deduction only if you've fired fewer than 1% of your workforce in the last calendar year;
2. You don't get to sponsor for an H1B if you've conducted ANY layoffs in the last calendar year; and
3. The tax deduction only applies to unionized workers.
And while we're at it, let's roll back this ridiculous tax structure where IP can be "sold" to a subsidiary in Ireland and then royalties paid.
Plenty of "big tech" already did it. Microsoft could not be more famous for stack ranking dating back to the 90s. Amazon have long had that kind of culture too.
Can someone explain this? What taxes do unprofitable US businesses owe that this would be deducted against?
Say you would have been exactly not-profitable ($0) if you could expense all of your R&D as in the old system, therefore avoiding tax. Now with the new rules you may be on-paper profitable because you can only deduct 20% of the R&D as an expense this year. The remaining 80% of that expense tips you over, becomes profit, and that's taxable.
In 2024, your business has $1m in revenue and has $2m in expenses. 100% of these expenses are R&D salaries (engineers you hire.)
Your company loses $1m/year. (You brought in $1m and spent $2m.)
Under the old rules, you'd owe no tax because you were unprofitable.
After Sec 174, what the IRS now says is:
You had revenues of $1m. But you only had $400k in expenses (because you now have to spread that $2m in R&D expense over 5 years).
So actually you had a profit of $600k! And you owe tax on that $600k profit (~$120k)
So you now have an additional $120k tax expense, making your business even more cash-flow negative.
.
Amusingly, if you're pre-revenue, none of this matters (you have no income at all, so it doesn't matter what your expenses are.) You get hardest hit by this change when you have some revenue and when you do a fair bit of R&D.
Otherwise I'm quite amazed that salaries can be carried forward as future expenses.
I know of at least two Western European countries where you don't have to do that. Don't worry, we pay enough taxes either way ;)
Sorry for the noise :(
The relevant paragraph from Section 174:
> (3) Software development
> For purposes of this section, any amount paid or incurred in connection with the development of any software shall be treated as a research or experimental expenditure.
how about "business process mechanization"?
Classifying them as non R&D doesn’t help saving taxes again.
That is surely wrong? Just because those salaries are for R&D?
I could understand if there was some additional tax break for R&D which was being removed. I can't see how basic operating costs cease to be expenses.
>That is surely wrong? Just because those salaries are for R&D?
The same would be true if you hired a bunch of scientists/engineers and got them to do R&D.
In the UK, business gets taxed on profit, which is what is left from revenue after subtracting costs.
If your employee expenses remained constant, then by year 5 you would be deducting $2m from your revenue since you'd be accumulating the deductions from the previous four years.
So in steady state it wouldn't necessarily be a big problem. But for a startup which is hiring many new employees and whose revenue is growing it's a huge problem.
The salaries are of course expenses, but they are exactly offset by the value of the IP created by the R&D activities.
It's a bit as if you spent money on buying some materials. As long as the material doesn't degrade, the cash is gone but the value is the same and therefore won't reduce your taxes.
If that IP is amortized over a single year, it does not contribute to taxation, but it does if it is amortized over a longer period.
Buying a truck is an expense, as is buying gas for the truck. But the former you have to amortize over x years, the latter you can expense immediately.
The law used to be "employee salaries for software are like buying gas" and now it's "employee salaries for software are like buying a truck".
The rationale behind amortization isn't exactly the idea that the asset can be sold, it's that the asset is producing revenue over multiple years. For software, the asset is the codebase.
Let's say you hire a single software dev, for one year, and they write Excel++, which you can sell for the next ten years. It would be entirely appropriate to amortize the cost of creating that software over those ten years, based on the matching principle (a fundamental idea of accounting, matching expenses with revenue).
The issue in the real world is that's not how the software industry actually works, 99% of the time.
You must have misphrased what you intended to say. If what you wrote was true, a software company's most valuable asset would be the specific programmers in its employ. If true, average tenure of a programmer would be way longer than 1.5->2 years as companies worked really, really hard to keep their most valuable assets from walking out the door into the doors of another company just to get reasonable pay increases.
Perhaps your opinion is influenced by doing post-collapse-sales of a whole bunch of software houses that built just plain bad software? I can't see why else you'd be selling "software IP" independently of the rest of the business.
Anyway. Given that information, how should you have phrased what you wanted to say?
What would be a more appropriate model from accounting perspective?
An unprofitable business doesn't pay income taxes. Businesses are taxed on their net income (i.e., profit).
People are railing against this as the cause of tech's recent underperformance, but it was a non-factor for the vast majority of tech companies, because most tech companies aren't profitable and wouldn't have paid taxes anyway.
The employer makes less profit due to salaries, but they won't "lose less" or make more money due to salaries.
Under that argument, the government would have a direct incentive to dictate how businesses do business to maximize taxable revenue.
After that, we can nitpick: should the development costs of new software be encouraged the same as maintenance costs of existing software. If you want to encourage startups, then yes they should. If you want to discourage startups or very temporarily increase tax collection, then no.
lol
Discouraging starting new businesses would be unconstitutional. All freedom in the US is derived from being able to participate in controlling capital.
1) Accounting rules are to match revenue with the expenses responsible for them, which I think is a good principle. If your workers make something now that provides revenue for 5 years, it makes sense to spread that expense over 5 years too. In many cases, you would want to do that as a business, makes it more clear how your business is profitable vs not.
2) Decisions whether to "build vs buy" a capital asset should not have massive tax implications. If I buy CoolSoftwareProduct from someone and resell it for the next 5 years, I'd have to amortize that. Should be similar if I hire a coder to write CoolSoftwareProduct instead.
(This doesn't mean that "salaries should always be amortized" is the right answer, of course, I think it's a very silly law)
I do not like many things in BBB, but I am glad to know there is at least something in there that I can be glad for.
The better question is why the tech industry seemed to forget that the first Trump administration was terrible...
Temporarily, for 5 years.
Now Trump second round fixes it, but expires in next (presumably) Democrat administration.
> tl;dr on Section 174, Research & Experimentation costs went from being fully deductible in the year incurred to being deductible over a 5 year period.
Larger tax bills and a tightening on what roles/activities are deductible as R&E are likely what OP is pointing at with his comment.
To the best of my non-inside baseball research, Section 174 changes were simply one part of a package of revenue generating measures to offset the large tax cuts from the broader tax act they were a part of.
The changes came from The Tax Cuts & Jobs Act of 2017 that was introduced to the House of Representatives by Congressman Kevin Brady (R) Texas. The bill passed both houses of Congress along party lines. Then President Trump signed the bill into law. Section 174 changes did not take effect until 2021.
https://news.ycombinator.com/threads?id=heymijo&next=4332098...
1. I start “Facebook for dogs” It’s gonna be massive. For the first year me and five guys code away in the garage and I use my savings / credit card / family trust fund to pay them 100k each. Expenses are 500k, revenue is, amazingly, 1.5M and taxes owed is 500k.
At this point turning round and saying the development was R&D, and claiming 500k of tax breaks is just (to me) ripping off the American Taxpayer.
And I’m not even an American Taxpayer.
If the revenue was zero would anyone suggest that the taxpayer give me 500k to help ? (Ok I would because I like free money but most people won’t)
Or am I missing something?
You can only expense $100K of the salary costs this year, so even though you're break-even, you pay taxes on $400K in income.
Or, even worse, imagine revenue is $250K, and you spent $500K on salaries for the team.
You can only expense $100K of the salary costs this year, you're already -$250K on the year, and now you're paying taxes on $400K in income. You're destroyed.
VC-backed startups aren't designed to get profitable quickly, and I don't see that as a problem for the American taxpayer, and nobody is saying the taxpayer is giving money or helping. A business losing money should not have to pay taxes on income, as if it's not losing money.
in this case, your taxable income is $1.0M, and cash flow is $500k ($1.5M - $500k salaries - $500k taxes).
now you have to amoritize it over 5 years. so your taxable income is $1.4M ($1.5-500k*20%), taxes are 700k, and cash flow is $300k.
Uncle Sam just reduced your cash flows by 40% by a simple tax change. You eventually make up the difference, but for fast growing tech companies, that's a large shift in current flows and significantly changes their investment strategy.
The most important bits:
* Subsection (a) requires amortizing "Specified research or experimental expenditures" over 5 years (paragraph (2)) instead of deducting them (paragraph (1))
* Paragraph (c)(3) is a Special Rule that requires that all software development expenses be counted as a "research or experimental expenditure".
That's it. All software expenses must be treated as research and experimental expenses, and no research and experimental expense can be deducted instead of amortized. Ergo, all software expenses must be amortized over 5 years.
I strongly recommend reading the section before forming an opinion. It really is quite unambiguous and is unambiguously bad for anyone who builds software and especially for companies that aren't yet thoroughly established in their space (i.e. startups).
Also note that this makes Software a special case of R&D. It's the only form of R&D that Section 174 requires you to categorize as such and therefore amortize.
Everyone assumed it was a traditional accounting hack. But given the timing and the reinitialization, it's clearly political, not economic.
The code is a strategic time-bomb designed to cause a high-profile economic downturn during a presidential election cycle, specifically when the following president is a Democrat and Republicans have a house majority.
It was used to harm Biden's economy, and it will happen again in 2030 if the next president is a Democrat. While deferred, it will be spun as a major Trump "economic achievement" for the midterms, because companies will be able to afford to hire again.
The tech industry is merely high-profile fodder for extreme politics. It really is that petty.
https://en.wikipedia.org/wiki/Filibuster_in_the_United_State...
With Republicans usually being dominant in a number of states, if Democrats have a Senate majority, it is usually both narrow and dependent on a very small number of Democratic and/or Dem-leading moderate independent Senators from Republican-majority states who vote with the party on leadership, but are soft (or firmly opposed to the progressive preference) on a number of issues important to progressives.
If the US were approximately an equal democracy, this might be less of an issue.
Equal to what?
Trump himself admitted it's better for Republicans when fewer people vote.
This thread is talking about the Senate. The senate isn't gerrymandered. Both senators are state-wide races.
If you want to view it that way, you can view the senate as "pre-gerrymandered". But the last time that was an option was in 1959, and both of those are just "the entire area the US owned, but wasn't a state yet. To get senate gerrymandering, you have to go back to 1912 and the admission of New Mexico/Arizona.
And thinking about it more, though I haven't seen if there are studies on it: there are probably manpower/fundraising effects from gerrymandering.
If you're able to protect your political power in one area that probably better enables you to amass resources to use in the area you can't gerrymander.
But all that said, both parties practice gerrymandering and I don't think there's strong evidence of a significant advantage over a major party from current gerrymandering at the national level.
[1] https://da.lib.kobe-u.ac.jp/da/kernel/90008864/90008864.pdf
[2] https://electionlab.mit.edu/articles/gerrymandering-turnout-...
[3] https://stateline.org/2022/05/20/check-your-polling-place-re...
That is quite explicitly the history of the US Senate (and House), FWIW.
The Connecticut Compromise was reached to give low-populations states outsized legislative power in the senate. This is the main reason the senate exists.
Building on that, the 3/5th compromise was reached as part of this to give slave states outsized legislative power in the house.
The state of Maine used to be part of Massachusetts, but it was later set up as an independent state in order to increase the number of anti-slavery states in the senate (the Missouri compromise).
How? Evenly divided voters and representatives are the issue. Each side can barely afford to lose 10% or so during votes
The issue being discussed in the Senate is not a symmetric issue resulting from near balance in support between the parties.
If you ask me "should corporations pay more taxes?" I will say, yes. Famously so does Warren Buffet, is he also a progressive?
If you ask me, "hey should we gut tax incentives for R&D spending in the USA?" I will say, uhhh no? probably a bad choice?
It's a 10% tax cut for big corporate America, with some economic poison for blue states in the future.
Generally, in tax bills they try to keep them "neutral" where any tax cuts or tax breaks are coupled with tax increases elsewhere BUT they tend to report the 10-year affect for whatever reason. This bill provided a ~30% cut in corporate tax on profits, with a delayed increase in tax cost on Software R&D pushed to the next term.
If the next party wants to reverse it, they'd have to find the money with an increase in tax - directly undoing it would be a ~50% increase in corporate tax rate, which (I guess?) would be a tough sell politically. Meanwhile, the tax code on software engineering sounds too niche to expend political capital on.
Either way, its another example of how corporate America is trading long-term growth (R&D, product development) for short term gain (lower taxes today).
Given the history of prior presidents winning 2 consecutive terms, it seems like Trump could have reasonably expected a 2022/2023 tax change to be his problem.
Public opinion can change daily, and external events can appear with no warning. These things can make a prior path of action vanish, or even make it madness to pursue.
If you try to plan everything long term, I bet you hit a lot of disappoint as a politician. If you only see today, then you're not fighting for things that are now not possible.
I imagine one would be far less stressed as a result. And maybe more popular than otherwise.
Why?
Prices haven't gone down at all nor will bringing manufacturing to the US do this (likely causing them to go up) but his approval rating is 50%
Trump is blaming Biden for the obvious outcome of Trump's tarrif nonsense. What do you think Trump would have done?
Are there any parties running in a track record of functional government?
What about Detroit?
And Detroit... well, I guess now that they've bulldozed all the abandoned buildings it looks less like a post apocalyptic hellscape and more just abandoned. An improvement I suppose.
I'm not saying it's the actual story, but the timeline does track.
[0] Page 60, Sec 1306(e) sets the date: https://www.congress.gov/115/bills/hr1/BILLS-115hr1enr.pdf
My argument is simple: Occam’s Razor
The Republicans in congress put the provision in solely as a gimmick to get past the CBO.
Frankly I don’t think legislators in either party are competent enough to have foreseen the consequences and even if they had been they wouldn’t have put a bomb like this in that would be more likely than not to backfire and affect them.
I just think that too often people interpret incompetence as malice, especially nowadays when things are so polarized that it’s fashionable to hate people who differ with one’s political opinions.
For those around when this went into effect many business owners were surprised. Our accountants told us they seriously thought congress would fix this before it went into effect.
Get a better accountant.
It's completely unimportant. Nobody is getting fooled "on paper" by amortized salaries.
I’ve seen it used in UK listed companies to massage the profit numbers and make divisions of the company seem more profitable than they are
In the US, it remains the case that programmers salaries must be treated as an expense (i.e., cannot be amortized) when calculating the company's income statement, balance sheet, etc. Not following that rule will get the accounting firm signing off on those financial reports in trouble (with the SEC, the Public Company Accounting Oversight Board, and maybe even the Justice Department if the purpose of the violation was to defraud investors).
Here's some food for thought:
* Global financial crises: Banks were paying (bribing) ratings agencies to rate junk bonds AAA.
* Savings & loan crisis: widespread fraud & insider abuse.
* Bernie Madoff: Ran the largest Ponzi scheme ever, with an estimated fraud total of $65B raking in $17.5B in invested cash.
* Enron: straight up accounting fraud sprinkled with intentionally causing brownouts in California to pad their pockets with a side bonus of making Gov Davis unpopular & get him recalled (Enron was closely aligned with the Bush administration).
* Nixon straight up using psy-ops against Democrats & finally trying to burgal the DNC offices.
In terms of stats, the FBI does a few hundred bribery and corruption cases annually. Are they good at catching white collar crime? Well such crimes regularly take more than 5 years to investigate.
And hell, some things that are basically lying and cheating are straight up legal. Usury is legal with minimal to no regulation of payday loans. Pyramid schemes are legal as long as you call it multi-level marketing.
The list goes on and on.
I thought you were being sarcastic here at first because, good lord, there is plenty of corruption here in the US (though those doing it used to care more about hiding it). The US, especially in its current state, is certainly not a place I'd describe with "almost no lying or cheating". I do understand that Russia is on another level, though, given the open assassinations and doing things like what was done to Navalny.
You've never been in Russia. There is no clear law abiding business there. That opens a lot of opportunities for those with some power. Corruption is one of them, selective punishment is another. I'm sure in most 3d world situation is not better, but they at least don't have laws to cheat and bribing isn't a crime.
The funny thing, is that people not from America say that there IS corruption, but at least it happens in the open. I think OP is saying the same.
How is that corruption?
A good definition from AI "Is when government officials misuse their power for personal gain or to benefit their friends or associates"
Are you living in an alternate world?
I'd say we're slightly behind western Europe as far as rule of law goes, not really sure about the advanced east (Japan, Korea), and miles ahead of just about everywhere else (eastern Europe, Russia, Africa, China, etc). Yes, even with Trump in office, though he really makes me worry.
This is not slightly behind Western Europe. This is miles behind any developed country. China may be corrupt, but Xi Jinping hasn’t yet sold beans or cars via press conference.
It's not perfect, but you could do so, so much worse.
Being better than others really isn't the only thing that matters.
The amount of daily activities in the US that just work 99.999999% of the time that would have a corruption aspect in some other countries is mind boggling.
The closest analogy I can come up with is imagine if every money transaction involved cash tipping the parties involved. And that’s just the beginning.
This hurts small companies (like mine) that were priced out of the US developer market.
I'm curious if contract work is really exempt, would look like a major loophole to me.
I can’t believe this still exists, and no one has changed it. We truly are governed by morons
this was done to fuel their tax cuts to a small group of a certain people.
you can see all of the sponsors here: https://www.congress.gov/bill/115th-congress/house-bill/1/co...
This is actually a really common intention in laws like this. Get the tax cuts during your term, and then kick the can down the road so your successor's term is marred by the bad law. If your successor wants to fix it, they need to pass a different tax to recoup the costs, and incur the publicity of "raising taxes".
https://www.congress.gov/bill/115th-congress/house-bill/1/co...
I think the purpose of the change was to "increase revenue":
> Requiring that certain research or experimental expenditures be amortized over a five-year period or longer, starting in 2023, would increase revenues by $109 billion over the period from 2023 to 2027.
https://www.congress.gov/congressional-report/115th-congress...
It's a specific tax, on a particular class of better educated workers in specific jobs.
Yes, but in a specific way: they were trying to offset the tax cuts they wanted so they could pass it via the reconciliation process and avoid the Senate filibuster. They didn't actually care about this revenue and the assumption from most people was that the specific carve-out would disappear in some future bill.
Each and every startup will have a year 0 where they're spending more than they earn, and under the new Section 174 they will only get to deduct 10% of their employee's salaries that year. In year two they get to deduct 20% of year 1's salaries and 10% of year 2's salaries, which is still 30% of what the established players will be able to deduct. By year 4, if they make it that far (which most startups don't) they'll finally be at 90% of a full deduction.
Add to that the fact that startups also by definition have a much higher rate of growth than established companies and you'll find that a startup almost definitionally will be paying substantially more in taxes as long as it remains a startup, because they only get to deduct an average of the last 5 years of expenses from this year's revenue in order to calculate this year's profit. That's fine when your last five years are more or less similar to this one, but it's terrible when you've been growing.
The net effect of this change can only be to disincentive startups and cement big, slow established players.
Yikes. Does that apply to outsourcing?
From what I've read, not for software fixes to ongoing products, but for new products and I can't remember for new feature work. Also if you contract for someone else I heard you can still write off expenses without amortization.
It had a huge impact on my personally, I'm a small R&D shop and basically I have had to end all risky long-term research projects.
In addition to the research costs, I'd also have to pay taxes on the research costs mostly up-front. Significantly, if the project doesn't work out, I'm still out of pocket for the tax money. It's a penalty for taking a risk, and it kneecaps American innovators in a globally competitive technology race.
The rules are even worse than the article notes because it double-dings open source developers. See Section 6.4 of https://www.irs.gov/pub/irs-drop/n-23-63.pdf. The relevant bit is here:
> "However, even if the research provider does not bear financial risk under the terms of the contract with the research recipient, if the research provider has a right to use any resulting SRE product ... costs paid or incurred by the research provider that are incident to the SRE activities performed by the research provider under the contract are SRE expenditures of the research provider for which no deduction is allowed ..."
The rule as written means contractors who write Windows drivers could deduct their expenses (as they would have no residual rights to a closed-source work product), but contractors who write Linux drivers may not (as they would have some rights to open-source Linux drivers).
The problem is that the license assigned says that anyone is free to use the code. Anyone is a set of people that includes the contributor, which then triggers the interpretation that the research is incrementally in the contributor's benefit and thus disqualified from preferential tax treatment.
You'd need a custom license where everyone in the world could use the results except for the contributor, and then like, a source control system that hides the source files from the contributor's view of the repository.
As I understand accounting, this means that reported profits would be higher, and therefore incur more corporate income tax liability. Cash flow isn't effected besides tax.
A startup isn't likely to be making a profit yet, under either accounting rule. Is there a benefit to reporting a larger loss?
My first thought is that this effects Google and suchlike, not startups. But... assuming steady state "r&d" expenditure... it's not that much. Everything gets deducted within 5 years anyway.
So... maybe this hinders more modestly profitable, and fast growing companies most. Those that can't afford to carry 5 years worth of paper profits as easily.
Otoh... I am curious about how the difference between r&d expenses and operational ones are determined irl.
This should be quantifiable. How much extra assets are software companies actually booking?
It seems questionable that this "silent killer" had actually affected employment so much.
So previously, some 20% of all revenue would be owned as corporate income tax, and startups would deduct it all as they're spending much more on R&D than they owe in corporate income tax. But with this tax change, the deduction would be much lower (80% lower IIUC).
But just as an accounting note: R&D expense has nothing to do with the company having revenues for an existing product, which already is allowed to deduct cost of goods sold, selling and admin expense. It is a cost related to future business and in that regard, it is not crazy to say it should be amortized. That in the past this did not happen, or that accelerated depreciation for other assets is in the IRS code is a function of the government wanting to effectively subsidize business investment.
https://pro.bloombergtax.com/insights/federal-tax/rd-tax-cre...
https://tax.thomsonreuters.com/news/the-future-of-rd-expensi...
Doesn't that make software engineers one of the few employees with much worse tax treatment?
That's fundamentally different from regular software development outside of agencies where there is no direct relationship. Software development is closer to an investment than an expense.
Amortization sucks in general, yes, because the money is gone and it doesn't affect your taxes to the same amount, but that's not different for any company doing manufacturing or anyone needing specialized tools or vehicles that cost significant amounts.
What if the chef invents a new signature dish that makes your restaurant famous for the next 10 years?
For chains like McDonalds where they actually research and develop ways to make pink slime look like a burger, maybe? But do you call them chefs?
EU provides a large pool of experienced developers seeking new opportunities on salaries well below SV. Why pay 500K for a burnt out "rockstar" who spends more time on twitter than doing actual work when you can hire highly skilled people in Eastern-EU (or even in Berlin).
Section 174 seems unlikely to progress unless attached to broader legislation.
> "More promising is the Tax Relief for American Families and Workers Act of 2024 (H.R. 7024), which proposes restoring immediate expensing for U.S.-based R&D investments through the end of 2025. " -- https://www.pwc.com/us/en/services/tax/library/tax-committee...
If I could start anywhere in the World, Switzerland would be above all the war-torn and crime-ridden places, but business-wise it's no good for a tech startup.
I'm wondering, if such a movement doesn't doesn't exist already, do I need to start it myself?
- Bribe the right people
I hate to provide such a cynical and lazy response but we've got until midterms (maybe) before you really have a shot at 'democratically' influencing the system. For the time being you'll have to work with the mafia that's currently running things and outbid whoever wanted this to happen in the first place.
Didn't AAPL, GOOG and FB all create products _before_ they had any taxable income? Would this change have had any actual impact on their foundings?
I don't have an answer for you. But I support your intrigue.
DARPA was working on Project LifeLog starting in 2003, was to be "an ontology-based (sub)system that captures, stores, and makes accessible the flow of one person's experience in and interactions with the world in order to support a broad spectrum of associates/assistants and other system capabilities". The objective of the LifeLog concept was "to be able to trace the 'threads' of an individual's life in terms of events, states, and relationships", and it has the ability to "take in all of a subject's experience, from phone numbers dialed and e-mail messages viewed to every breath taken, step made and place gone".
The program, at least officially and publicly, was cancelled on February 4th, 2004, the exact same day that Facebook was founded.
https://en.m.wikipedia.org/wiki/DARPA_LifeLog
https://en.m.wikipedia.org/wiki/Facebook
You can call it a coincidence if you want, I just tend to be very skeptical of "coincidences" where massive, powerful, unaccountable, immoral, unethical institutions like the US intelligence community get exactly what they want at the expense of our civil liberties.
[0]: https://medium.com/insurge-intelligence/how-the-cia-made-goo...
It applies to things like configuring your internal tools too. Good luck at audit time.
Most likely neither: It is its massive trade deficit, the one it strangely wants to get rid of now, that has allowed US consumers to consume more than they produce (i.e. you can take something with no real expectation of having to give anything back in return). Which, as it relates to tech, has enabled offering services for what is effectively free to dominate the market. Nobody else in the world can compete with that.
> Didn't AAPL, GOOG and FB all create products _before_ they had any taxable income?
Wouldn't you say they had no taxable income because of it? If Facebook brought in $100,000, and paid $100,000 to developers, then there would be no taxable income under normal regimes. But if the developers were not tax deductible, then that $100,000 in revenue would be taxable, even though the bank account is empty. This isn't nearly so simple, but it has changed the calculus in a similar way. The business models of old no longer work because of it.
A bankrupt company can still sell their computers. Selling you code, lol -- code is more of a liability really :)
It's important to consider that lawmakers (who are not well informed or downright stupid) might think code has intrinsic value because of media married with a lack of real-world experience.
This was a Blue Ribbon School 1992-1993 yup. https://www.ed.gov/sites/ed/files/programs/nclbbrs/list-2003...
I am nitpicking but since a microscope or a computer is a tangible asset, the correct term is depreciation. Amortization applies to intangible assets.
I have a few friends who specialize in it with 2 ongoing contracts for splitting off pieces of software.
What is that? Software sold by companies that have HQ in the US? Or software created by someone in the US? Because if it is only the first, good riddance.
Over long time scales (and big company revenue streams), this is sort of a wash. I think this hurts startups a bit more due to the long timescales involved which eats up much needed cash in the short term.
Since this is done on annual buckets it's very common to try to move items in both columns between years to minimize tax.
It is sort of between the two in my view and is highly dependant on what the software engineer does each day.
Are they fixing a bug, helping a customer, refactoring? I think that is operational.
Are they building out a new feature? That is capital. But it is not quite like buying equipment because it adds no value to the books. So depreciation seems off.
But the same issue applies to other roles. Is a sales persons day trying to land a sale, or trying to develop the business.
It all comes down to "intangible assets" and whether you are making them.
I think it is easier to just say if you are paying someone to work then you can deduct. There must be better ways to claw it back.
The whole reason for most business to exist is to use operations (operational costs) as a lever to increase the growth and intangible value of the business.
Consider a contractor in a software maintainer role; accounting for this as capex makes zero sense.
Small business owners are very impacted by the R&D schedule.
It's clearly not enough to cover all of the expenses that are required to generate your "revenue", but it's a gesture in that direction.
It’s a fudge to make projections look better to allow congress to pass a budget neutral reconciliation bill with the intent that congress would remove the fudge before the consequences triggered.
Governments in general are pushing for capital gains tax normalization where instead of requiring a taxation event the capital gains tax would be levied yearly. In such a scenario the only difference remaining would stem from the difference taxation rates.
You’re alluding to wealth taxes, right?
Because taxing unrealised gains are wealth taxes.
Or maybe I’ve misunderstood?
No, wealth taxes are a tax on retained wealth (a stock). Taxing unrealized gains is a tax on income (a flow), it just changes the point at which taxation attaches from a realization event to the actual gain.
The unrealized values are a fiction. There is significant value in treating values as unknowable when they are, in fact, unknowable. Forcing people to make up a fake valuation creates a lot of adverse incentives.
Then instead of taxing the gains, you'd accept the government nationalizing the assets by eminent domain and paying fair compensation that was significantly less than the "fictional" unrealized value?
Or if someone unlawfully deprived you of the asset, you'd accept as restitution or seek as civil damages for the loss something significantly less than the "fictional" value?
Or, when it was no longer an excuse to avoid fair taxation, would that "fiction" suddenly be a lot more real to you?
It’s much easier to do because there is no disputing the assessment since the person implicitly agrees to the valuation. And it allows people to forgo realizing any benefit from the unrealized value at all to avoid taxation.
Say take x% of the top of the money lent to someone who uses their unrealized gain to secure a loan. Make the money paid count against any tax they owe if they sell the asset later.
Take the value of the asset assessed by the bank and the price paid for the asset to find the total value that is unrealized gain.
Divide that by the total value to get percent of collateral that is unrealized gain. Multiply that by the loan value. Then multiply that by the tax percentage.
All you need is for banks to report secured loans to the IRS and it’s easy.
But if those skyrocket in price from tax, they'll be more subtle about convincing banks they're good for the money and pay a slightly higher rate for unsecured loans.
Or maybe they'll just treat the asset securing the loan as having the pre-gains price. Get the bank to agree it's worth at least what you paid, with no further analysis.
If you try to plug those loopholes you lose the "much easier to do because there is no disputing the assessment since the person implicitly agrees to the valuation" factor.
Are you oblivious to the extensive litigation that occurs in cases like eminent domain because there are substantial differences of opinion on even the notional value, never mind the realizable value?
Notional valuations are fiction, everywhere and at all times. Treating them as some kind of objective reality is just enabling a lot of abuse and motivated reasoning.
Yes, you have. You have an asset of greater value which you can leverage in a number of ways without liquidating it and "realizing" the gains. That's a real gain, with real value.
> you could be taxed over and over again
Only if you make new unrealized gains.
> and if the stick drops or hits zero then what?
Then you have a negative unrealized gain, or, equivalently, an unrealized loss. If you are taxing unrealized gains instead of taxing gains when realized, then the natural assumption would be, just as is done with taxing gains at realization, that negative unrealized gains are either offset against current income or against future unrealized gains, and so effectively create (considered on their own) negative (current or future) taxes. The simplest form of this is to offset only against future gains, by the simple mechanism that when gains are recognized for tax purposes, they adjust the basis value of the asset, and when unrealized losses occur, they don't effect the basis value at all, so you don't have a taxable unrealized gain again until the market value exceeds the basis value established at the prior peak.
More complex versions would allow you to offset some or all of the unrealized loss from the prior basis value against current income of other forms, but the amount of that offset would reduce the basis value of the asset.
Most laypeople grossly conflate notional and real value. Taxing notional value massively inflates the adverse impact of tax incidence on expected returns relative to people’s casual intuition based on the relative tax rates for realized and unrealized gains.
A tax on unrealized gains is in effect a way of laundering a steep tax rate so that it looks “small” and therefore reasonable to the unsophisticated.
Being difficult to assess value is a problem they’ll make you pay an accountant for and punish you if you get it wrong, it’s not going to stop them.
Even in the case of real estate, a large amount of value is locked up in extremely non-liquid markets. You might get a vaguely representative market-clearing transaction once per decade, with high price volatility that makes it nearly impossible to predict what the next market clearing transaction will look like. I’ve owned assets in these types of non-liquid markets; differences in subjective valuations can vary by an order of magnitude and there is no evidence from the market to support any of those values.
If you only include extremely liquid markets for tax purposes in order to make valuations vaguely plausible, assets will be made non-liquid such that they are excluded from consideration. Ultimately this is why taxes on unrealized gains have been a challenging proposition in practice. We have no way to accurately model realizable value for the majority of assets and current simple approaches produce extremely wrong estimates a substantial percentage of the time.
Of course this is a prediction of something that hasn’t happened before but looking at the chess prices move this does appear to be an intended destination.
tl;dr: Many assets have no meaningfully assessable fair market value. These are investments with extremely long and indefinite time horizons before the asset value can be assessed in a reasonable way. You can look at it as a peculiar type of risk capital portfolio with an extraordinarily long time horizon.
No, its an actual thing, measurable by some mechanism. Otherwise, this would be a non-discussion, as taxing it would be impossible, not a possible thing that we can argue about the merits of.
> The notional value is often not remotely realizable.
Whether it is or is not immediately realizable is immaterial to the desirability of taxing it; it may be material to designing the forms of taxation that should be acceptable. E.g., if the difficulty of realizing the value is, across the tax base, likely to making collecting the tax in cash or equivalents difficult, it would argue for permitting a fallback option for the tax to be collected in-kind, e.g., by the taxing jurisdiction acquiring a proportional interest in the asset equal to the share of the value of the asset represented by the taxes not paid by other means.
> A tax on unrealized gains is in effect a way of laundering a steep tax rate so that it looks “small” and therefore reasonable to the unsophisticated.
If you allow carry forwarded losses, even just by the simple method of adjusting basis values, and include taxes on realized gains (and carry forward, offsetting against current income with perhaps a negative net, etc., for realized losses), then taxing unrealized gains is identical to taxing realized gains if the gains are eventually realized, but simply avoids the ability to find maneuvers to benefit from leveraging the value of the asset without paying taxes by avoiding realization. It doesn't make a "steep" tax rate look small, it makes the tax rate look like exactly what it actually is, unlike taxing only realized gains, which makes an effectively non-existent tax on capital gains look like something more, when people can benefit from assets without realizing the gains.
Before this change, tax for software development was calculated against:
* Profit = Revenue - Expenses
And software developer salaries fell neatly into Expenses unless you were looking for an R&D tax credit.
After this change, tax for software development is calculated against this new equation:
* Profit = Revenue - (1/5 * YearlyExpenses[-1]) - (1/5 * YearlyExpenses[-2]) - (1/5 * YearlyExpenses[-3]) - (1/5 * YearlyExpenses[-4]) - (1/5 * YearlyExpenses[-5])
Which means that if you are in Year 1 of operation, your values for YearlyExpenses[-2:-5] are all 0 and you only get to deduct 1/5 of your actual operating costs for the year from your "profit". So you can be in the hole but still owe taxes on your "profit" for the year because what you spent money on was classified as R&D.
Why should money spent on software _development_ not have to be deprecated over time like other money spent on _development_?
I get that it sucks from a cash flow standpoint but the same is going to be true of other R&D expenses. It's just that we're more exposed to this specific R&D expenditure and not others.
And, as a sibling points out (and as I pointed out in a comment at the top level), software is in this regime singled out from all other possible R&D expenses, making it particularly vulnerable. A skilled accountant/lawyer can probably turn big chunks of other R&D expenses into something that doesn't fall under 174. No amount of skill can do that for software, because we're singled out.
It is if the only thing your company does is create software. No operations, no sales, no physical assets to purchase sell or manage.
The new rules would apply from 2025 to Dec 31, 2029:
https://www.crowell.com/en/insights/client-alerts/house-comm...
Just a reminder that Congress, even now, can rapidly act on a laser focus when it is sufficiently motivated.
It was higher profile because Congress decided it should be higher profile.
Thinking about it, the US government is going exactly the same way of the Roman republic.
We're gonna conquer and annex Egypt? What an awesome time to be alive!
I saw let Trump’s ugly bill die and then a small fix up to the tax code could be this. Should be able to pass.
And I’m tired of pretending like we aren’t going to be beneficiaries
Every Congress increases the debt, we can acknowledge that the cuts they picked are going to wreck the lower class especially with the medicaid, we can acknowledge that it won’t meet its goals of cuts
but are you guys just scared to acknowledge its going to super charge things that you are a beneficiary of too? so busy saying it just benefits billionaires as if we’re trying to avoid guillotines. not gonna happen and many people here are going to try to take advantage of new programs
No, it's not, because giving its authors any power or wins or ability to execute on their agenda is disastrous for nearly everyone, including tech and the middle class. The only people that its illegal acts are good for is a tiny minority of crooks, fascists, and oligarchs.
Given the firehose of illegal stuff they are doing that is impossible to push back on, it is utterly imperative to push back on every little thing that is possible to push back on, and to hold consent hostage to an end to the former.
You're letting them burn your house down because they promised you a bottle of whiskey.
Regardless of whether it benefits our industry or socioeconomic status, it'd be incredibly shortsighted to just do all of that at the expense of the lower classes.
Congress can always pass anything else at any speed. This slow motion filibuster thing is a choice, and the powerlessness of doing anything about that choice just means everyone else should have a single they care about too to correct the laws and riders that shouldn’t have passed.
Really? You can’t think of anything you wouldn’t support in order to get this thing that benefits you?
I'm fine admitting that I would benefit greatly from this bill. I also hope to heaven it doesn't pass because an additional trillion dollars to suit me sounds asinine. I don't need help.
And yet only one party campaigns on fiscal responsibility, debt concerns, and reduced spending
174 is so small it can't go through both chambers on its own so it needs to get attached a larger bill like OBBA.
It's unfortunate because it appears both sides want this repealed to allow immediate amortization of domestic R&D expenses.
Well, yes, but then everyone doesn't really want it, do they? Someone wants something else, and wants that something else enough that it is worth jeopardizing the supposedly universal goal for it.
If you've ever negotiated, I bet you've done the same thing of jeopardizing something you want in order to get something else you want. If you never do that, you'll make a lot of deals where you're riding the edge of just barely acceptable and the other person is taking advantage of you. But in this case, with a standalone law, doing it gets pretty rude and we'd be better off if nobody did it.
There's a minimum size for laws?
Arguably, that's the whole of politics: why should I give you something if you don't give me something?
The people involved are, generally, not deep thinkers, aren't aren't thinking much beyond their direct short-term advantage. The system selects against that.
Other attempts that come to mind: 1. Tax Relief for American Families and Workers Act of 2024 (H.R. 7024) 2. American Innovation and R&D Competitiveness Act of 2025 (H.R. 1990)
This article is informative: https://www.cebn.org/media_resources/section-174-sign-on-let...
So now it seems its like a pseudo tariff against any other freelancers and producers for software outside of US.
Deciding whether labor is a capital or operating expense and deciding how to depreciate it if capital is also not a subsidy.
Dealing with Section 174 amortization in those first one to three years is a real headache (and your tax bill ends up higher than if it didn’t apply). Once your startup survives that the first few years of doing Section 174, things do get easier... but, sadly, most don't make it that far.
Also, the 10+ years before the layoffs started tech companies were on a hiring binge. Much of big tech was hiring to keep people off the market and off their competitors payrolls (this is from friends of friends in FANG HR departments). These were high paying jobs, too.
All of which make hiring engineers unattractive
It's effectively 6 years too. You only get to depreciate 10% in 1st year. This might have killed my company if it was around during first years.
See my comments on the previous discussion (Nov 2023) here: https://news.ycombinator.com/item?id=38145630
This would be a no go for startups though.
Query any search engine for "are US income taxes direct or indirect taxes"
Every one will tell you that they are direct taxes. This is false. The supreme court has exclusively held that income taxes have always been indirect taxes (excises specifically, read about what an excise is in any authoritative source on tax law) in a constitutional sense. (See Brushaber v Union Pacific RR Co. 1916, Moore v U.S. 2024)
The sixteenth amendment did not give congress the power to directly tax citizens (or domestic corporations) and the complexity of the tax code is an attempt to obfuscate this fact, but the code is not inscrutable, it has rules.
Unsure of why this matters? Look up the difference between direct and indirect taxes in US law. None of these deductions matter unless you are a foreign corporation. I have tried commenting about this in other income tax related threads (this is my alt account), but people here don't like the idea that there is government propaganda in the US, or that most people are wrong and blindly accept the socialization about taxation without verifying what the law says.
I realize this is a disturbing truth to accept, not least because it involves accepting that most people who have been prosecuted for income tax crimes are only guilty of ignorance of the true legal purpose of the forms they signed. You can easily verify that most accountants and tax attorneys do not know what they are talking about by asking them this simple question about direct vs indirect taxation.
This is not legal advice, it is a wakeup call.
The masses are in a persistent state of slumber, so they will never wake up. Depressing but true.
Some people will point out that AI will fix this, no it won't:
1) The real cost is higher than anything you'd pay for a person an there is not likely any real change there.
2) AI will be lies like Actual Indians that won't scale
3) Here's the kicker: If AI does succeed, now these multi-billion dollar firms will have to compete with multi-billion dollar single person businesses, that eat their lunch
Its a race to the bottom right? That means you need to invest in the business and all these layoffs are exactly not that, and will leave companies unprepared for the next 10 years.
Remind me in 2035.
This is an artificial subsidy. That’s not how the tax code treats other types of investments that generate recurring income.
Also, finally programmers with the right to live and work in the US catch a break: salaries for US-based programmers can be amortized over only 5 years as opposed to the 15 years of non-US programmers.
It has effectively become a lot more expensive and difficult to employ a programmer. Once this change went into effect we started to see hundreds of thousands of layoffs.
Then tech executives started aggressively talking up how you could use AI to write code instead of having humans write it.
Now of course reducing headcount and the associated expenses and replacing them with a bot sounds tempting to executives no matter what. But it sounds REALLY tempting when you've been on a hiring freeze since 2022 due to the fact that you can no longer deduct employee salaries in the year you pay them out.
Bear in mind that both Republicans and Democrats say they want to fix this and haven't done so due simply to gridlock and government incompetence.
I think most software businesses are taking a wait and see approach. Don't hire until this thing gets fixed. In the meantime, double down as hard as you can on automating those programmer jobs out of existence, in case the law never gets fixed.
Section 174 is the root cause.
I used to work at a small startup, and most of our spending went toward engineers’ salaries. If we had to amortize that over several years back then, I don’t think we would’ve made it.
It’s surprising how a single line in the tax code can quietly make it harder for small teams to hire. Makes me wonder how many other policies are silently shaping things behind the scenes.
This has been a slow moving disaster for years now and people have repeatedly tried to raise the alarm.
Just crickets and layoffs.
This provision can and does lead companies to owe significantly more in taxes than they make.
The only reason it hasn't been bigger news, is because most companies are pretending it doesn't exist and just sweeping it under the rug, hoping it will get fixed before enforcement gets serious.
Why pretend that it doesn’t exist? Why not vocally lobby for a change in the tax code?
There is very little pressure on elected officials because big cos can afford it and it bankrupts their tiny future disruptors.
Why would you let it be fixed?
Hasn't Ben Thompson of Stratechery spoken about this a number of times? I'm aware of this 'feature' and I'm not even in the USA, let alone a COO at a private-equity-backed yada yada.
Trump is just getting started. By the time he is finished, your economy will be shot to pieces. The US dollar will no longer be the reserve currency for global trade.
AFAIK it was also affecting more freelancers outside of US since amortisation is 15 years. For EU citizen IMHO this is equivalent of US putting tariffs on outside world. I wish EU at least try to fight back and revenge on US Tech by increasing taxes or also making all US tech bought by EU companies to be 15 years amortised so they have taste of their medicine.
dtagames•2d ago
The games industry, while hugely profitable and bigger than TV, movies, and music combined, laid off tens of thousands of people. It's unmitigated greed is all it is.
tjchear•2d ago
According to the article, as long as the tech workers contribute to improving or creating a product (be it games or apps), they count as R&D cost.
dtagames•2d ago
gregw2•2d ago
This sort of thing appears to be self-reported; I don't know if it ever gets audited. I don't know if big tech lies or creatively interprets what counts and that has contributed to the issue. But this article sort of over-represents what qualifies as R&D for US tax purposes.
ghc•2d ago
https://larsco.com/blog/section-174-updates-navigating-the-i...
ndriscoll•2d ago
robocat•2d ago
The rule says if you pay someone $200k to develop software: then you now have a $200k asset that then devalues to value of $0 over 5 years (starting midyear). That's just plain weird.
For our example a depreciation table might look like:
The final effect of the 174 rule change is that you still finally end up with a software asset worth $0. However you now have taxable income of $200k in year one and expenses equalling $200k spread over 5 years. The taxes paid could be a lot: although the taxation money is really just being lent to the government for a few years at 0%. The actual financial costs are fucking complicated.Understanding accounting and taxes are two absolutely essential skills if you ever wish to be a founder (and useful anyways).
Finding a solution to dealing with the valuation of assets is difficult. The historical solution of depreciation is broken for software, intellectual property and goodwill. In theory, taxes on dividends and capital gains taxation already deal with the issue (company taxation at x% kinda ends up at $0 because the shareholder pays y% and claims back the x% through imputation).
And remember that salaries are properly taxed.
ndriscoll•2d ago
If it turns out it's not useful, we could then allow companies publish the source and release it into the public domain to immediately "destroy" the asset (the copyright) and claim their deduction. So failed r&d projects would be deductible right away as long as the public gets them, and ones that result in a useful asset get depreciated based on how long they actually last, which is currently potentially multiple lifetimes.
jameshart•16h ago
Software built by a business is a trade secret independent of its copyrightability. Even after the expiry of copyright a business can continue to exploit it as a proprietary asset.
kelnos•15h ago
Amortizing development cost over the useful life of the software is maybe a reasonable thing to do (I don't think it is, but let's for a minute say I agree), but determining "useful life" is not simple.
nitwit005•14h ago
Which, I think is an overlooked part of this. They must constantly have gotten feedback that people were lying to them.
jewelry•2d ago
dtagames•2d ago
There is no justification for "cost cutting" when it hurts the larger economy. If the company were losing money, that would be different, but these mass layoffs are all from firms that make obscene, enviable levels of profit. It's greed.
RobGR•2d ago
But they were making high profits for decades, and being greedy for decades. Then there were a lot of layoffs. What changed ?
orangecat•16h ago
There is no justification for "cost cutting" when it hurts the larger economy.
It is not good for the economy to have people doing work that doesn't produce value.
jokethrowaway•2d ago
But it's not "greed": it's the end of zero interest rate policies.
https://newsletter.pragmaticengineer.com/p/zirp
mywittyname•16h ago
I bet they were classified as R&D for accounting purposes. Product development largely falls into R&D - it doesn't matter what the product being developed is for.
Every job I had at a megacorp was classified as R&D, and I know because I had to track hours against such.
demosthanos•16h ago
It's not just that. Section 174 now explicitly calls out Software as always being an R&D expense:
> (3) Software development
> For purposes of this section, any amount paid or incurred in connection with the development of any software shall be treated as a research or experimental expenditure.
https://www.law.cornell.edu/uscode/text/26/174
kelnos•15h ago
Even though it sounds unintuitive, that activity is considered R&D for tax purposes.
jen20•15h ago
Wrote software, like, you know, "developed" it?