Is convertible debt with a price cap really the best financing structure?

January 9, 2011

It’s been a while since I’ve written a substantive post.  (I started writing this post in September 2010, but only got around to finishing it.)

Convertible debt with a price cap seems to be the preferred structure for early-stage financings

Over the last 12 months, I’ve noticed a trend where early-stage startup companies raise seed financings of between $250K and $1M using a convertible note with a price cap.  In fact, it seems very rare to see a convertible note without a price cap these days, although it was fairly common to raise convertible debt without a price cap a couple of years ago.

Paul Graham has sparked some discussion on the topic in August 2010 when he tweeted “Convertible notes have won. Every investment so far in this YC batch (and there have been a lot) has been done on a convertible note.”

Seth Levine asks, “Has convertible debt won?“  Seth points out that “the convert cap reflects a premium to the current fair market value of the business. One super-angel I talked to told me that while a year ago these caps “approximated what I’d pay in equity” that they’re now “33% higher than what I’d normally agree to pay now.”

I have also heard that many of the companies in the YC batch that Paul Graham was referring to have been dictating convertible debt terms with price caps in the high single digit million range when pitching to angel investors.  This leads me to believe that there is a mini-bubble in the early stage financing universe.

Is a priced Series A financing a valid alternative?

A logical alternative to convertible debt is a priced Series A preferred stock financing. Mark Suster does a good job analyzing whether convertible debt is preferable to equity, and concludes that convertible debt is better. I’ve pondered the issue of convertible debt vs equity and come to the same conclusion.  Most commentators generally seem to be concluding that convertible debt with a price cap is better than a priced Series A round.

However, Ted Wang points out that Series Seed documents are better than capped convertible notes.  (I’ve used the Series Seed documents twice in the last six months and have managed to keep company-side legal fees under $15K on both of the deals, although I suspect that the documents for the “real” Series A financing will end up being more painful to draft as the Series Seed documents don’t scale well in later rounds.)  Fred Wilson, while he doesn’t endorse the Series Seed documents, says that Union Square Ventures has never participated in a convertible note deal.  Jason Mendelson points out that the use of debt fundamentally changes the fiduciary duties of managers and board members of the company, as executives and directors of “insolvent” companies face potential suits from various creditors.

Tweaking convertible debt so that common stock (instead of preferred stock) is issued for the conversion discount in order to limit liquidation preference overhang

Given that convertible debt with a price cap seems to be firmly entrenched these days as a typical structure for early-stage financings, I’ve pondered some methods to make the structure more favorable to companies.

One issue that has bothered me is that the conversion discounts and price caps result in a company creating more liquidation preference than the amount of money that investors have invested.  For example, if the amount of money raised in the convertible debt financing is large (say $1M), and the conversion discount is 50% (or the price cap is 50% of the Series A valuation), then the company will end up issuing $2M (+ interest) of Series A preferred stock to investors that only paid $1M.  This Series A preferred stock will typically have a 1x liquidation preference, so $2M of liquidation preference overhang will be created as a result of the conversion discount/price cap.

Thus, a company is in a worse situation by using convertible debt with a price cap than if the company simply did a Series A financing at the price cap valuation instead of the convertible debt.

Please keep in mind that there are two principal economic features of preferred stock:  the liquidation preference, and the equity on an as-converted to common stock basis.  Convertible debt with a price cap preserves the investor’s “equity” ownership, but gives the investor extra liquidation preference.

In order to solve this problem, I have resorted to modifying the conversion discount formula in certain deals so that the conversion discount is paid in common stock.  In other words, the principal and interest on the convertible debt will convert into preferred stock with no discount, and the discount portion will converted into common stock valued at the preferred stock price. This is an attempt to make convertible debt with a price cap economically equivalent to a priced Series A financing.

One might note that this is not the exact economic equivalent, as the application of the anti-dilution formula on the preferred stock issued upon conversion of the debt will not be as favorable as if the convertible debt financing was originally done as preferred stock. However, many seed equity financings don’t provide for anti-dilution protection, as the valuations are low to begin with.

[Note: A better way to make convertible debt identical to a seed financing is to have the convertible debt convert into its own series of preferred stock.  In other words, the new investors would purchase Series A and the convertible debt would convert into Series A-1 with an aggregate liquidation preference equal to the principal and interest of the debt (and a liquidation preference per share lower than the Series A price per share).  However, most people don't like the messiness of creating an extra series of preferred stock.]

[Please also note: It probably doesn’t make sense to entirely focus on the equivalent theme, as there are plenty of other ways that convertible debt and equity are different. For example, convertible debt has no voting rights, and typically don’t have protective provisions. However, seed investors don’t seem to mind the lack of these features.]

An example of a term sheet provision to provide for the conversion discount to be paid in common stock and the math involved is below.

Term sheet excerpt:

CONVERTIBILITY:        In the event the Company consummates, prior to the Maturity Date, an equity financing pursuant to which it sells shares of its preferred stock, which are expected to be Series A Preferred Stock (the “Preferred Stock”), with an aggregate sales price of not less than $[1,000,000], excluding any and all indebtedness under the Notes that is converted into Preferred Stock, and with the principal purpose of raising capital (a “Qualified Financing”), then all principal, together with all accrued but unpaid interest under the Notes, shall automatically convert into shares of the Preferred Stock and the Company’s Common Stock at the lesser of (i) 80% of the price per share paid by the other purchasers of Preferred Stock in the Qualified Financing and (ii) the price obtained by dividing $5,000,000 by the Company’s fully-diluted capitalization immediately prior to the Qualified Financing (the “Discounted Purchase Price”).  The total number of shares of stock that a holder of a Note shall be entitled to upon conversion of such holder’s Note shall be equal to the number obtained by dividing (i) all principal and accrued but unpaid interest under such Note by (ii) the Discounted Purchase Price (the “Total Number of Shares”).  The Total Number of Shares shall consist of Preferred Stock and Common Stock as follows (see also example on Exhibit A):

The number of shares of Preferred Stock shall be equal to the quotient obtained by dividing (i) all principal and unpaid interest under the Note by (ii) the same price per share paid by the other purchasers of the Preferred Stock in the Qualified Financing (such price, the “Undiscounted Purchase Price,” and such number of shares, the “Number of Preferred Stock”).

The number of shares of Common Stock shall be equal to (i) the Total Number of Shares minus (ii) the Number of Preferred Stock.

Example:

Assumptions:

Principal + Interest    $1,000,000
Discounted Purchase Price    $1.00
Undiscounted Purchase Price    $2.00

Total Number of Shares:

Principal + Interest ($1,000,000)        =    Total Number of
Discounted Purchase Price ($1.00)        Shares (1,000,000 shares)

Number of Preferred Stock:

Principal + Interest ($1,000,000)        =    Number of Preferred
Undiscounted Purchase Price ($2.00)        Stock (500,000 shares)

Number of Common Stock:

Total Number of Shares             -    Number of Preferred          =    Number of Common
Shares (1,000,000 shares)            Stock (500,000 shares)    Stock (500,000 shares)

Comments

  • http://twitter.com/chrisco chrisco

    Nice addition to the “convert with cap” thread. Here http://news.ycombinator.com/it… Paul G at YC said he was going to publish a model doc, but that was a few months ago. Hopefully coming soon. Maybe needs friendly reminder. Would be a great kickoff to 2011 discussion and evolution, especially if mixed in with your fix/nuance. Cheers!

  • http://twitter.com/ken_hillyer Ken Hillyer

    It has been too long since your last post. Sage (and entrepreneur friendly) advice as always.
    Might have to these terms a try…

  • http://twitter.com/tydanco Ty Danco

    Saw this reposted on VentureHacks. Nice addition to the internet commentary around on converts, had not appreciated some of the subtleties you bring up.

  • Johne

    Nice article…..your discussion referenced preferred shares only. What about common stock for the Angel round? I have a company financed by the founders over the last 4 years. They are looking for $1M from Angels and then looking at a Series A round of $5M-$10 with VCs 12 to 18 months down the road. The VCs are obviously going to ask for first position if a liquidation were to occur so why go down the path of preferred shares forthe Angels?

    I look forward to your reply.

  • http://www.startupcompanylawyer.com Yokum

    @Johne – Issuing common stock to investors at a high price screws up your ability to issue options to employees at a low price. In addition, most sophisticated angel investors will not agree to a common stock financing.

  • http://www.mattbartus.com/ Matt Bartus

    Yokum, do you think the issuance of common stock at the preferred price has an pricing impact on the common, or is that such an outlier event that the 409A evaluators will ignore it. Also, why not fix the problem above by just using warrant coverage instead of a discount? I think all things being equal, if I was going to take the dilution, I'd rather have the investor have to put in more money to purchase the Series A shares on a warrant instead of getting free common stock. Thoughts?

  • http://www.startupcompanylawyer.com Yokum

    @Matt – I suspect the 409A folks would probably ignore the common stock issued upon conversion. Certainly, a warrant might be better, but I think the reality is that we're stuck in an environment where the seed financing market has moved to convertible debt with discounts and price caps, as opposed to warrant coverage.

  • http://twitter.com/mbartus Matt Bartus

    I agree on that point. Interestingly, I think one of the reasons the discount has become market is because it has been pushed by lawyers as easier and less paper intensive (and resulting in a cleaner cap table) then using warrants.

  • http://bottomlinelawgroup.com/ Antone Johnson

    Great discussion here. Yokum, your main point about liquidation preference overhang is an astute observation; it's something that had never occurred to me before. Have you seen much pushback from investors when proposing as Company counsel that they take common stock for the conversion discount?

    @Matt I share your concern about 409A effects. Valuation is so much more an art than a science; I always worry about any event that might inadvertently jack up the FMV of Common, as well as changes in accounting rules that can be as unpredictable as the weather. Cheap equity has such a powerful incentive, recruiting and retention effect for early stage startups, I hate to see clients do anything that could undermine it. That said, I'm not a big fan of warrant coverage either.

    This really is an interesting conundrum where doing things the most theoretically “perfect” way (issuing a Series A-1 or warrants) inevitably produces more complexity — inherently undesirable in my view. For similar reasons, even though they might benefit founders, I'm not a fan of the Series Seed approach (per my blog post last year at http://bll.la/55), or Series FF for that matter (as clever an innovation as it was). Most founders I work with seem to favor the KISS approach over the pursuit of theoretical perfection.

  • Pingback: K9 Ventures » Blog Archive » Thoughts on Convertible Notes