Post Category → Formation

When LLCs Make Sense for Startups

TL;DR Nutshell: In the vast majority of instances, tech startups are best served by starting out as Corporations (C or S-corps, but usually C-Corps) on Day 1, and lawyers suggesting otherwise are usually generalists who lack tech/vc-specific domain expertise to understand why. However, there is a narrow set of circumstances in which LLCs make sense for a startup.

Background Reading:

This post is about the “almost” part of that tweet. But to get there, it’s important to address the “simpler” and “tax efficient” aspects, because those are the two core reasons that I often hear pushed onto founders for why they should be LLCs.

LLCs may be simpler generally, but Tech Startup LLCs with investor capital and equity compensation never are.

Here’s a hypothetical: Imagine you’re an athlete who’s signed up for a football camp held in Boston in the middle of February. Your general knowledge of Boston weather tells you that it is going to be a** cold. You ask a few other people with knowledge of Boston, including me (I went to law school there), and receive confirmation that Boston is a** cold in February. So you show up to the camp with only your winter gear… but it turns out the camp is entirely indoors in a heated facility. Whoops. Should’ve asked someone with true domain-specific knowledge of that camp, not just people with general knowledge. 

That, in a nutshell, is what happens when lawyers and other business people tell tech founders to use LLCs. LLCs are extremely common in the general legal world. For simple operations with one or a small number of owners, they are by far the dominant legal structure, because they usually are simpler. However, for tech startups, who very often (i) use equity as a significant part of their compensation for employees/service providers, and (ii) often raise capital with multiple equity classes, complex preferences/rights, etc., things get extremely complex under an LLC structure, much more so than with corporations. The amount of tax and legal analysis that has to be done to issue equity compensation and/or raise capital in an LLC is (without exaggerating) 10x that of a corporation.

So, if your plan is to raise capital and use equity as a form of compensation for employees and contractors (which is usually a hallmark of a tech startup), do not delude yourself for a second that an LLC will be simpler than a corporation. 

The “Double Tax” issue usually only matters if your startup is a “cash cow.”

Yes, in a general sense LLCs have one layer of tax and C-corps have two. That is another reason why (as stated above), LLCs have become a very dominant legal structure, not just for simple companies but also for many large businesses as well. Again, though, context is key. The “additional layer” of tax that corporations face is on net profits; after accounting for expenses, including salaries. No net profits, no corporate tax. So if a startup is going to be generating substantial profits (taxed once) with the end-goal of distributing those profits to shareholders (taxed again at individual level) as a dividend, the two layers are a problem.

But how many high-growth tech startups do you know that, instead of reinvesting profits for growth, pay profits out as dividends? Not many; certainly not in the first 5-10 years of the company’s life. Most high-growth tech startups deliberately operate at a net loss for a very long period of time, and therefore (i) aren’t worrying about taxes on net profits, and also (ii) are taking advantage of those losses at the corporate level in a way that may not be even use-able on the individual level. This, btw, is also why S-corps are usually not very helpful for tech startups either.

And to add an additional wrinkle: in an acquisition, corporations often have the ability to do tax-deferred stock swaps, whereas LLCs don’t. So, in short, the “LLCs save a lot of taxes” perspective, while generally correct, is usually misapplied to tech startups by people who simply don’t do enough startup/vc work to give sound advice. Yes, VCs often push companies to be C-Corps (read the background articles), but VCs are hardly the only reason why the C-Corp structure is used in tech. 

LLCs therefore make sense for tech startups that:
(i) expect substantial net profits very early on;
(ii) aren’t planning on raising institutional venture capital, and/or
(iii) aren’t planning on using equity to compensate a lot of people.

Lots of net profits early on (rare)? The single layer of tax may be worth it, and even institutional VCs sometimes are willing to accept the complexity of an LLC to take advantage of the tax savings.  Not planning on raising VC money any time soon? Other types of non-tech investors are usually more comfortable with LLCs than VCs are. Not planning on paying your employees with equity? Then you’ll avoid the tax nightmare of issuing LLC equity to dozens/hundreds of people.

Few tech startups fit the above scenario, and that’s why few are LLCs. The classic tech startups that operate (rationally) as LLCs are bootstrapped/self-funded software and app companies with no plans to scale very quickly with outside capital, and large “marketplace” startups for which the actual investment in the technology is minimal relative to the large amount of revenue/profit pushed through the marketplace. For almost everyone else, C-corps are king, and for good reason.

p.s. I am not your tax lawyer, and am not pretending to know the right answer for your specific company. The above is just general knowledge; not legal advice. If you rely exclusively on a blog post to determine your legal structure, without talking to a professional to understand your context, you’ve taken on the risk of screwing it up.

Legal Technical Debt

TL;DR: The entire point of contracts is that they are permanent, and can’t be fixed unilaterally. That makes legal mistakes far more costly to fix than coding mistakes. But similar to code, your early stage legal work sets the foundation upon which everything else gets built, and therefore the cost of fixing errors compounds over time; like technical debt. In short, legal work is the last area where any knowledgeable entrepreneur should cut corners, thinking they can just fix issues later.

“Siri, please amend my charter to authorize a Series AA round, prep me an offer letter for a CTO, issue options to 3 recent hires… oh and review/execute that stock purchase agreement with my accelerator. Keep the fees under $500.”  — Not too far off from how a (confused delusional) segment of the startup community thinks startup/vc law should work.

Imagine if advisors told startup founders that, in order to conserve cash, they should aim to spend as little as possible on developers. Find cheap ones. If the non-technical CEO can code something himself to get by, do so.  Just get it done. Don’t overpay.  In fact, if we can automate our development process, do it. Keep cash spend on ‘the business.’

Anyone with an ounce of experience in building successful tech companies would recognize this advice as absurd and dangerous, as if quality and accuracy are irrelevant. Yet every so often I hear about advisors giving this exact advice to founders, about legal spend. And while fewer may acknowledge it, such advice is equally as absurd.

Of course you’d say that, Jose. You’re a startup lawyer.

Well, maybe. But let’s process it a bit.

Why would minimizing your spend on software development (like legal services) be stupid and dangerous?

It can be explained in part with the term ‘technical debt.’ via Wikipedia:

“Technical debt… is a recent metaphor referring to the eventual consequences of any system design, software architecture or software development within a codebase. The debt can be thought of as work that needs to be done before a particular job can be considered complete or proper. If the debt is not repaid, then it will keep on accumulating interest, making it hard to implement changes later on. Unaddressed technical debt increases software entropy.”

While I’m not a developer, my general understanding of the term is that bad coding becomes more expensive to fix over time, in an almost compounding way. And there are even circumstances in which it is so bad that nothing short of a complete re-write will make it scalable and useable. In other words, going cheap on developers just means you are compounding your cleanup cost and headaches for the future, and even threatening a complete shut-down of the product.

Minimizing legal spend works exactly in this way, but magnified 10x.  I frequently write on SHL about the many parallels between complex contract drafting/VC law and top software developers. Both groups involve highly skilled people capable of analyzing, managing, and manipulating large amounts of complexity. Both implement changes for which the stakes on a company are very high. Both expect to be compensated well for their skill set.

Software developers produce the code base on which your product runs. Lawyers produce the code base on which your company, including its relationships with investors, board members, executives, and employees, runs. 

A crucial difference between software code and legal code is that bugs are far easier to find and fix in the former than in the latter. Software code is constantly being revised, with thousands or millions of users revealing bugs on a regular basis. Legal code (contracts) are executed and then put away, often to be reviewed only at high-stakes moments, when fixing them is extremely expensive or even impossible.

Unlike software code, you can’t unilaterally issue ‘updates’ to executed contracts. Any experienced lawyer has seen a deal cost 6-figures more than it should have, or even completely die, because of legal mistakes made earlier in the company’s history. So think of contract drafting for a scaled startup as high-stakes software development for which virtually any material bug is completely unacceptable once the code is shipped. Still want to ‘minimize’ legal spend?

Law is Code; Not Product.

In my experience dealing with many many sets of founders, a part of the startup community carries the very deep misconception that startup/vc law has been, or even can be, completely productized. Want to ‘just’ issue some stock, grant some options, close a seed round, etc? It’s been done hundreds of times before, so it must be all ‘standard’ by now. Just click a few buttons, fill in some names and numbers, and you’re done.

This is the attitude of someone using a product for which clean, standard, predictable, pre-defined features are already in place. “Just” issuing a service provider some stock should be like ‘just’ moving some files around on Dropbox, right? There’s a serious flaw in this thinking. The clean, standard, predictable company and contract history simply does not exist, and hence full automation is pure fiction. 

  • What state are you located in? Laws vary, even if you’re a standard DE corp.
  • Are you a C-corp? S-corp?
  • Are there protective provisions that need to be complied with?
  • Any anti-dilution protection?
  • Enough authorized shares in the charter?
  • Enough reserved equity in the equity plan?
  • A well-documented value of the equity?
  • Is there a written agreement explaining the consideration and complying with securities laws?
  • Is the recipient an individual or an entity?
  • Board approval?
    • Are we confident the composition of the Board is well-documented?
  • Is stockholder approval necessary?
    • Any specific thresholds?
  • Vesting?
    • 83b?
  • Acceleration? What kind?
  • Any other special provisions/requirements implemented by past investors?
  • etc. etc. etc.

Virtually every VC, angel group, accelerator, large company, etc. has its unique variances in the contracts it executes/negotiates. States have different requirements. Laws change. Reality: people are not standardized.  They have their idiosyncrasies, and people determine what does and doesn’t get signed; what gets added to the code base.

Even if companies share 90% of the same legal DNA, the 10% variance is a massive wrench that makes automation, or even any kind of significant simplification, impossible without taking on enormous legal technical debt. That statement is not coming from a luddite lawyer who hates technology, but from the CTO of a startup/vc law practice that I am 100% certain is on the cutting edge of legal technology (the kind that actually works) adoption.

Telling a VC lawyer that you ‘just’ want to issue some stock is not equivalent to ‘just’ using a pre-coded feature in a product. It is far more like telling a software developer that you ‘just’ want to add a feature to your existing, non-standard, unique code base.  Imagine telling that developer to do it as quickly and cheaply as possible. Imagine hiring the cheapest developer you can find to implement that feature.

The contract that actually issues the stock may be 99.8% standard, but it has to be implemented into the historical set of contracts/context without blowing anything up. Contracts and laws do not sit in little, isolated modules without any impact on the other. They’re all inter-connected, with a change in one potentially resulting in a cascade of effects in others. Hence the code base analogy.  The larger, more complex the code base (set of contracts, number of jurisdictions, people involved), the greater the skill and experience required to work with it safely. And having a well thought-out, well-designed architecture implemented from Day 1 dramatically impacts the scalability and resilience of that code base.

So when a client says they ‘just’ want to issue some stock, all they might think about is opening a word document, filling some names, and signing. Of course that can be automated.  What often isn’t considered is the lengthy, complicated list of steps and analysis needed to fit that template document within the Company’s existing legal history. That, not the template stock purchase agreement, is what lawyers do, and software cannot do.

De-Valuing Law, like De-velopers, is De-lusional.

Anyone who sells a product or service into a market learns that not every buyer is willing to pay the cost necessary to deliver that product or service in an efficient manner, within the bounds of physics/reality.  Some buyers simply can’t afford it. But many others just don’t value the product or service enough to pay even the lowest possible price. As a lawyer, I learned very early on in my career that this is the case with founders looking to engage lawyers.

If I’ve been sold the lie that startup/vc law is a completely commoditized, standard product, I am going to shop for lawyers, and assess cost, the way I would for any other commoditized, standard product. I “just” want to issue some stock. Like I “just” want some toilet paper. There are founders (a minority, but many) who understand very quickly why they need to pay good compensation for software developers, and yet will question every single invoice from lawyers.

While I’m always more than happy to walk through an invoice when it makes sense, E/N’s client intake process has been deliberately designed to filter out clients who, for whatever reason, de-value lawyers in this way.  Our website’s home page says “World Class Counsel, Brought Down to Earth.” Translation: top lawyers who are more efficient, responsive, and accessible than the large firms where they’ve historically been found. We compete with other firms who provide top-tier legal counsel to scaling tech companies; not with the unrealistic price expectations of people who, through inexperience or delusion, want Teslas at Kia prices.

The seed-stage period is the toughest time for startup legal budgeting. Things are starting to get more complex, but with only a few hundred thousand raised (let’s put aside California ‘seed’ rounds), every dollar paid hurts. Fixed fees, flexible payment arrangements, deferrals (but be careful), and good old-fashioned budgeting are the key to getting through that period with your lawyers. Any experienced set of startup/vc lawyers will know how to be flexible for seed-stage companies. Just always remember that flexibility (and efficiency) does not mean defying the laws of physics to get things as cheap as you’d like them to be. 

At the level of law that scaling companies require, technology will forever (or at least into the very distant future) remain a complement and not a replacement for lawyers. Yes, the legal industry as a whole is and will continue to undergo disruption as software eats up the more routine, commoditized parts of the profession. But VC-backed companies are not dealing with commoditized lawyers, and talented, creative VC lawyers are hardly, not even remotely, underemployed.  If anything, those of us who adopt new tools as they are developed have found our practices enhanced, not diminished, by technology.  It allows us to deliver more concentrated value with our time, which means a healthier attorney-client relationship overall.

If you engage your lawyers as the developers of an important foundation for your company – expecting effectiveness and efficiency, but staying realistic about the amount of complexity and value actually underlying their work, you’ll be surprised by the rewards.  For those who continue fantasizing about replacing lawyers entirely with apps, nothing will provide a better education than the moment the debt comes due.

Converting Your Startup to a Delaware Corporation, Correctly.

TL;DR Nutshell: If you’ve accepted that you need to convert your startup to a DE corp from a different entity type, then it’s also time to accept this: there is no off-the-shelf, “click a button and file” way to convert to a DE corp. It is highly contextual. The right lawyers can do it efficiently and correctly. The wrong ones will tell you it’s simple, screw it up, and require you to pay the right lawyers 5x more in the future to clean it all up.

Background Reading:

The purpose of this post is not to debate whether your startup should be a Delaware corporation. While we do work with a decent number of VC-ish backed Delaware LLCs (sometimes LLCs really do make sense), the vast majority of technology companies that raise venture capital either start or end up as Delaware corps. And the moment a lawyer starts playing contrarian with me, talking up why Delaware isn’t needed, or C-Corps are tax inefficient, I quickly end the exchange by asking how many VC-backed companies she’s actually worked with. We are talking about scaling tech companies and venture capital. Not small businesses or companies funded by local, non-institutional investors.

So for purposes of this post, we are going to take it as a given: you need to be a Delaware corporation, but you aren’t one right now. Converting is simple, right? Just file a form?

Converting from any kind of entity to a DE Corp is not “standard.” Ever. 

Properly forming a DE corp startup from scratch has, thanks to standardization and automation, become a relatively straightforward process.  The reason, of course, is that you’re starting from nothing, and nothing is the easiest condition to automate from; no messy context throwing a wrench in the system. Conversion, however, involves a history with any number of possible permutations, and that means all the shiny templates and technology must give way, partially, to human judgment.

  • What are your state’s rules around entity “conversions”; is a “statutory conversion” allowed, or will you need to do a merger or possibly even an asset sale? It depends. 
  • What approvals does your state’s rules require? It may be a majority of all equity, it may be 2/3, it may be unanimous. It depends.
  • What specific documentation (like a “Plan of Conversion”) and filings (often in both states, not just DE) do the rules require? It depends.
  • Are there any existing agreements in place that might require a separate consent to be obtained before the entity can convert? Ok you get the idea.
  • Does the company’s existing capital structure require a (hopefully quick) discussion with tax counsel regarding possible tax hits (phantom income) resulting from the conversion? This is a crucial issue to consider when converting an LLC to a Corp. 

Converting a Non-DE Corp to a DE Corp (Corp to Corp) is generally simpler than converting an LLC to a Corp. Converting any entity in a state that allows for statutory conversions (TX and CA do, NY does not) is generally simpler than having to do a statutory merger.  Whatever the context, you will screw it up trying to do it yourself. In fact, I’ve lost count of how many law firms have screwed it up, requiring founders to pay us 5-10x in cleanup costs than they would’ve paid if they had just hired competent counsel from Day 1.

The reason you will never just push a button to receive medical treatment is that every person is different, and tailoring high-stakes treatment to individual differences is precisely what professional judgment (supported by tech, of course) handles best.  Startup law is no different. Technology and tools absolutely cut down on waste, and yes there is a lot that is standardized even in conversions.  But in the end the institutional knowledge of the law firm you choose will determine whether it gets done efficiently and correctly, or whether you’re just deluding yourself into thinking that the guy promising a cheap, simple conversion actually knows what he’s doing.