The Problem of Anonymized Data

Over the past three or so years both OK Cupid and Facebook have published a great deal of aggregated, anonymized data for, let's say, the enlightenment and mirth of society at large. Facebook shared information on how "happy" users seemed on certain days and OK Cupid shared (as it has been doing) some interesting information about dating preferences of certain groups. 

The LA Times blog has been one of the few voices to point out the privacy implications of these data shares, saying, "Despite its silly name, the Gross National Happiness indicator is creepy. We're in there."

Several events - most notably the promotional (and seemingly innocuous) publication by Netflix of data regarding movie preferences of its customers - have revealed fundamental security problems with so-called anonymized data (Netflix was sued in 2009 in connection with this and settled out of court). By some accounts, the widely used concept of "personally identifiable information" on web site privacy policies - which by implication suggests the existence of information about you that is somehow not "personally identifiable" - is misleading. There's been a spate of studies finding that people with PHDs have the capacity to "re-identify" or disambiguate so-called anonymized data. The underlying principle is that ostensibly random data points (such as a birth date and a zip code) can be tied to a specific person if coupled with some other set of information (publicly available or readily mined, etc.) It's not clear that data anonymized in accordance with best practices is easy (cheap) to re-identify but it is clear that it can be done.

Thus, a quandry for businesses that want to sell or otherwise profit from anonymized data. U.S. laws and state laws with respect to consumer data privacy don't forbid it. In fact there are hardly any restrictions whatsoever on properly de-identified data, even if the data is being sourced from financial or health-related data or data concerning children (laws are more restrictive with respect to these categories of data). Further, as most privacy policies on web sites inform users that "anonymized" data may be shared with third parties, in most circumstances users have a reasonable expectation (due to a privacy policy or otherwise - provided they had proper notice and consent of the privacy policy) that their data might be released in such a form and hence would not have contractual cause to sue in the event it ever was. Relatedly, one would not expect the FTC (or analogous state agency) to prosecute a company for sharing anonymized data (on grounds of misleading consumers about data collection and sharing practices) so long it took precautions to ensure the data couldn't be easily re-identified.

But this is uncertain. The lack of case law at this point makes it difficult to clearly define what constitutes a) proper precautions (if any) for anonymizing data and b) reasonable consumer expectations as to the same . By way of example, The California Supreme Court recently held, for the first time, that a ZIP code - standing alone- qualifies as "personally identifiable information." See Pineda v. Williams-Sonoma Stores Inc., No. S178241 slip op. (Calif. Feb. 10, 2011).

Share Calculation on a "Fully-Diluted Basis"

Most seed or series financings calculate the preferred stock share price paid by the investor on a "fully-diluted basis."FN1 This sounds like math but really it's basic, macro-level accounting.

The underlying principle is this: the determination of how many shares the investor receives for the money he's investing is a proxy for something more crucial: the investor's percentage ownership of the company.

The investors have an incentive to capture as much percentage ownership as possible and thus have an incentive to capture as many shares as possible. "Fully diluted basis" means that the investor's proposed percentage ownership of a company be calculated against the absolute TOTAL number of shares outstanding in the company - the "basis".

The reason this matters, and why the terminology is often embedded in legal documents, is that often at the time of a seed or series financing there is outstanding stock that gets overlooked. This stock has been effectively but not technically issued - in the form of unexercised stock options, warrants, convertible debt and the like. The calculation of what's included in the basis determines who will assume the future diluting effect of any stock that is not exercised at the time of the financing (but almost surely will be at a later date).

If the basis calculation includes unexercised stock, then at the time of the financing the existing common stock holders (i.e., founders) - but not the investors - are, in effect, diluted. If the calculation doesn't, then when the stock is eventually exercised, all the existing stock holders, including investors, share pro rata in the dilution effect. For example:

Pre-financing Company has issued 5 shares of common stock but also has unexercised debt and stock options convertible into a total of 3 shares of common stock. Investor wants 50% ownership. If basis does not include the unexercised securities, investor gets 5 shares. If basis does so include, then investor gets 8 shares (for the same amount of money invested).

So, that's the concept. Mechanically, the calculation works on the level of the share: the larger the basis the less the investors will have to pay per share of the preferred stock. This is a function of the oft-cited formula: per share price = pre-money valuation / total outstanding shares.

FN1. “Fully-diluted” capitalization typically means (i) all issued (outstanding)  common stock, (ii) all issued (outstanding) preferred stock, (iii) any issued (outstanding) warrants, (iv) all issued (outstanding) options, (v) options reserved for future grant, and (vi) any other convertible securities. In essence, the only category of stock that is not counted in this fully-diluted capitalization number is the stock that is "authorized" in the Articles of Incorporation but as yet not issued.

The Crowd-Funding Problem

Crowd-funding - raising capital through a large network of individuals each providing a small investment that collectively adds up to a large sum - seems a natural evolution of the start-up financing model. The problem is that it's illegal. FN1

Securities laws generally require the "registration" of the sale of any securities, such as the common or preferred stock offered in seed or Series A financings, requiring technical compliance with a number of registration rules, including, among other things, the disclosure of vast quantities of information to purchasers and the preparation of audited financials. All of this is super expensive. The economics of crowd-funding don't allow it.

Of course, there are a number of exceptions to the registration rules, the most notable being the "private placement" exemption, originally curated under Section 4(2) of the Securities Act and later broadened under Regulation D. Most seed and series financings are conducted pursuant to Section 506 of Reg D, which permits sales of securities to "accredited" (i.e., wealthy) investors without costly disclosure documentation.

The whole idea of crowdfunding of course is that you shouldn't have to own a polo pony and a Porsche to make investments in start-ups, which moves us to Section 504 of Reg D, which does not impose an "accredited" investor requirement.FN2 The big problem here is that Section 504 still prohibits a "general solicitation" or advertisement of the offering to the public at large. 504 requires a pre-existing, substantive relationship between an issuer (or its broker-dealer) and the investor(s), and this is why 504 is relied on almost exclusively for investments from close family and friends (and not say, from half-known acquaintances you're connected to on linkedin).FN3

Accordingly, absent some special story, neither 504 nor any other federal exemption provides safe harbor for the crowd-funding structure. The only legal way (for now) to raise capital through crowd-funding would be to go through the full registration process at both the federal and state levels, which isn't economically tenable.FN4

FN1. This blog post presumes a conventional financing by a corporation and not a creative alternative (such as a revenue sharing plan, giving investors a flat payment or investors being involved in the day to day running of the business). Although be aware that Section 30(b) of the Exchange Act may prohibit transactions which are designed to evade the Exchange Act.

FN2. Note than 504 expresses a federal securities laws exemption. State securities laws impose their own requirements and exemptions. Although state securities laws tend to track federal requirements, in some cases, state laws are more restrictive.

FN3. Relatedly, for any entrepreneur interested in building a crowd-funding platform/web site, there is the broker-dealer problem. Section 15 of the Securities Exchange Act requires that anyone acting as a "broker-dealer" - which is broadly defined under the Securities Exchange Act of 1934 to mean “any person engaged in the business of effecting transactions in securities for the account of others" - must go through its own expensive and lengthy registration process. Whether a crowd-funding platform constitutes a "broker-dealer" is a technical question and the relevance of past precedent (comprised of almost exclusively of SEC no-action letters, etc.) is hardly clear. However, it's probably safe to say that the more any such platform/web site is promoting sales of securities (telling an investor about some company, trying to encourage investment) and the more the incentive the platform/web site has to make such sales happen (receiving a commission), the more the broker-dealer requirement is implicated. 

FN4. If history is any indication so long as there is significant market interest in a particular kind of transaction and no one is being taken advantage of, the SEC will make adjustments to the law to allow a market in such transactions to develop. Being a first mover in such a market involves without question legal risk but it would seem anyone who waits until the SEC makes compliance issues crystal clear will lose out. Such is the dilemma.

409A Constraints on Stock Option Grants

Hot tax code insight of the week!FN1 Section 409A of the Internal Revenue Code requires that companies issue stock options at a strike price that reasonable approximates FMV.FN2 This has one of two practical consequences:

              i) start-ups issue restricted stock instead (or an analogous exemption (any property (such as stock appreciation rights) subject to Section 83 is okay)), thereby bypassing 409A; or
              ii) start-ups go through the (somewhat expensive) process of establishing a FMV (in compliance with 409A) prior to issuing in stock options.FN3

Here's the background: under the previous regulatory framework (governing things during the early '00s), stock options could be issued fast and loose, often according to some vague formula (like 1/10 of the value of the last financing's share price). This allowed rich guys to defer huge portions of their rich guy salaries into the future.

So the IRS enacted 409A, which governs all deferred compensation plans and agreements entered into, or vesting after, January 1, 2005, and requires that i) stock options are issued at FMV and ii) the FMV must be determined using “reasonable application of a reasonable valuation method.” The IRS has provided guidance that the determination of reasonableness (an inherently circumstantial standard) will presumptively be satisfied by either of the two following approaches:

1) Independent Appraisal. An independent valuation by qualified experts using standard methods recognized under the IRS Code.

2) Illiquid Start-up Appraisal. Certain private companies (in existence less than 10 years and not anticipating an IPO in the next 6 months nor a merger in next 90 days (among other things)) can rely on valuation by a person (including an employee) with significant knowledge or training in performing such valuations. (Go here for a detailed discussion of the requirements).

The presumption of reasonableness is rebuttable only upon evidence that the method or the application of such method was "grossly unreasonable" (there's no official guidance on what this means).

Such valuations last for 12 months absent any intervening events that would materially and reasonable affect FMV. Failure to comply has a number of consequences, the most salient being that employees will be subject to taxation at each vesting milestone plus a 20 percent penalty, and potential interest.

Most companies engage in precisely the kind of appraisal required by 409A at each financing event but if such financing was more than a year in the past and the company is short on cash it is probably better off issuing restricted stock rather than stock options as incentive equity.

FN1. A character in David Foster Wallace's novel Pale King (set in an IRS office in the Midwest) describes tax compliance as "boredom beyond any boredom he’d ever felt."

FN2. More generally, 409A applies to any legally binding right to deferred compensation (any agreement, plan or arrangement that provides for a deferral of compensation, even if such compensation is subject to restrictions (such as vesting)).

FN3. The cost for valuation appraisal varies, but will probably come in between $5k and $25K.

83(b) Tax Elections

83(b) of the Internal Revenue Code is a niche tax-savings allowance for entrepreneurs and the tech geeks they hire (not the technical definition).

83(b) implicates the following: if you're a founder, an early employee or otherwise a recipient of RESTRICTED EQUITY ("equity" = stock or LLC units, etc. but NOT options (which aren't recognized as property by the IRS until exercise (but note that some options plans allow for early (prior to vesting) exercise and these options would be "restricted")); "restricted" = restrictions that lapse, typically due to vesting) in a start-up, then, presuming you'd rather pay less taxes than more, you should notify the IRS IMMEDIATELY upon receiving such equity (by filing an 83(b) notice).FN1

According to 83(b), if you voluntarily and timely (the 30 day(!) deadline is notoriously inflexible) notify the IRS that you've received such restricted equity, then you assume immediate income tax liability on the difference between the FMV of that equity (usually not much about zero (i.e., par value) in the inception stage of a start-up corporationFN2) and the amount you paid for that equity (usually par value or zero, as negotiated). At the future date when you later sell that equity, then you pay capital gains tax on the appreciation from the original date of purchase.FN3

If you fail to make that initial notification, you are taxed instead at each future vesting milestone (typically, at the year cliff and then monthly/annually for the next three years). This may have significant implications. For example:

Joe Founder is granted company stock at some nominal purchase (probably par value) price (say $0.01 per share) with a FMV of $0.001 per share. The stock has four year vesting with a one year cliff. Joe doesn't file an 83(b) election. At the end of the one year cliff, the stock having appreciated to $1.00/share, Joe recognizes and must taxes on $0.999/share of income for that year. As the remaining stock subsequently vests each year/month (whatever each vesting milestone is), Joe again recognizes and must pay ordinary income taxes on "income" (even though the stock is presumably illiquid and Joe can't sell it) equal to the difference between the (presumably rising) FMV of the newly vested portion and the original $0.001/share purchase price. Moreover, the company is required to pay the employer’s share of FICA tax on the income and to withhold federal, state and local income tax.
If Joe had made an 83(b) election, he would not recognize any income as the stock vests, because the 83(b) election forever freezes the income calculation as of the original grant date.

It almost always makes financial sense to file the 83(b), however, it does depend on the underlying value of the equity at the time of issuance and future prospects. In the odd circumstance that the equity at the time of grant has a material FMV AND there is a material risk that it won't increase in value (because, say, the start-up fails), you've accelerated your tax liability without receiving any benefit.

FN1. The mechanics of filing the 83(B) notice are surprisingly informal. While forms exist to facilitate the process, none are issued or required by the IRS. A handwritten note sent to your local IRS office (the full list of information that needs to be provided can be found here) would technically suffice. What really counts is the 30 day filing period. There is no extension available, or any simple cure for missing the deadline.

FN2. Note that Companies (via the Articles of Incorporation) typically assign a "par value" to stock (in some states its a requirement), which prohibits the Company from issuing stock to anyone at a price below that par value (i.e., no free stock).

FN3. The default Section 83 rule is that income (the difference between FMV and the price paid) on restricted stock is not recognized until the restrictions lapse. This rule is actually intended to benefit the taxpayer - the unique economics of start-ups undermine the intention.

Vesting: Single versus Double Trigger Acceleration

Most vesting provisions for restricted stock or stock options ("incentive equity") include acceleration provisions as insurance of sorts for employees and founders ("service providers").

The underlying concept is this: incentive equity typically vests on a four year schedule (as per market standard). Absent acceleration, if an event were to occur before the end of those four years - such as a sale of the all the assets of the company, a merger, or an IPO - that resulted in the service provider's termination (or resignation with good cause), then she'd lose the benefit of her expectation. She'd lose the right to the portion of the equity that hadn't vested - which happens to be precisely what certain interested parties (investors, new management/ownership, etc.) would like to see happen (because unvested restricted stock effectively vanishes and the rest of the shareholder base benefits proportionately from the reverse dilution) and why termination is such a real risk.FN1

Acceleration triggers guard against this. "Single" trigger refers to the automatic vesting of any unvested incentive equity upon the said event. "Double" trigger requires two events before the automatic vesting - not just the merger but the actual termination or early release of the service provider.

A common acceleration agreement averages the two triggers: combining 25% – 50% single trigger acceleration with 50% – 100% double trigger acceleration. Fairly convincing arguments are made that double trigger acceleration best balances the interests of service providers with interests of the company (entire shareholder base).FN2

It's all pretty straightforward stuff. One point to note: the trigger should run for a period of time before as well as after the transaction that constitutes the trigger event (or otherwise be constructed to to avoid any preemptive house cleaning before the transaction is done).

FN1.The assumptions behind the logic of what interested parties will want in the event of a merger/sale can get speculative but generally, regardless if the acquirer's interest is in company assets or people and /or if the equity at issue is unrestricted stock or stock options (Stock options, in contrast to unrestricted stock, "return" (to the extent they ever left) to the pool of stock reserved for employees (usually 5%-15% of the outstanding stock)), acceleration clauses will likely decrease the purchase price. To the extent there are provisions in place (such as unvested equity) that incentivize service providers to stick around the acquirer will see and presumably pay for that additional value. Although the acquirer could separately create (and pay for) an employee/management retention mechanism as part of the deal, that payout would (for the rational acquirer) carve-out from the overall deal value, reducing the consideration allocated to the target company stockholders. Note here the conflict of interest between VC investors and the service providers being bought out).

FN2. A variation suggested here is single trigger plus a minimum X (say 12) months of service before the out.

Warrants and Employee Stock Options

Employee stock options and warrants (both give the holder the right to purchase a security at a set price, usually referred to as the "exercise" or "strike" price) function in about the same way but have two basic structural differences (which explain why warrants tend to go to advisor/investor types while options go to employees).FN1

First, from the issuing company's perspective, warrants behave like a financing (albeit with no initial servicing costs like dividends or interest) whereas stock options behave like an employee incentive. This is because although both warrants and stock options are derivative instruments (the value is not in the thing itself but in its derived value from an actual security (i.e., stock)), the stock ultimately issued for a warrant is newly issued stock - prior to the issuance it did not exist - similar to the way new stock is issued to VCs when they make an investment in a start-up.

The stock issued for an exercised option, by contrast, is derived from the previously existing "stock option pool" (which usually comprises somewhere between 5-20% of the outstanding (i.e., already issued) stock of a company).FN2 Thus as a consequence, the exercise of warrants (like the issuance of preferred stock in a seed or series financing), necessarily dilutes existing shareholders (which is a significant event for the company). The exercise of stock options, by contrast, just consumes some portion of the shares set aside for the stock option pool.

Second, from the holders' perspective, a stock option is less valuable because it is subject to a set of restrictions.FN3 While in most cases a warrant implicates the right to purchase the underlying stock at any point in the future at the holder's reasonable discretion (and in some cases the right to transfer that right), a stock option i) can almost never be transferred, ii) is subject to a vesting period and iii) has a limited exercise period (meaning, the stock option, once fully vested (usually four years down the road), has to be exercised within a few months or a year after vesting).FN4

FN1. This discussion solely addresses the common use of employee stock options and warrants in a start-up context. Stock options and warrants can be manifested in myriad ways - practices outside the U.S. in particular add additional complexity.

FN2. If you get fancy, you can argue that if a company does not presume the existence of a employee stock option pool, the relevant distinction between warrants and employee stock options gets exceedingly small.

FN3. Usually. While employee stock options are highly standardized, the terms of warrants are highly customizable.

FN4. The tax implications of warrants and employee stock options depend on the circumstances. As a general matter, warrants are a taxable event upon issuance but options are not, provided, however, that the warrants were issued as part of a financing while the stock options were issued in exchange for future services. If the warrant is compensatory, taxability is deferred under section 83 until exercise.