1. Name of the Registrant:
2. Name of persons relying on exemption:
3. Address of persons relying on exemption:
4. Written materials. Attach written material required to be submitted
pursuant to Rule 14a-6(g)(1)
Misaligned Executive Compensation Structure
Has Rewarded Leadership While Stockholders Have Suffered
Entrenched Board and Unfriendly Governance
Structure Leaves Stockholders Only One Avenue to be Heard: Withholding Votes
Board’s Refusal to Include Proposal
to De-Stagger Underscores Importance of Holding
Board Accountable NOW
Directors Standing For Reelection This Year
Are the Board’s Leaders Most Responsible for its Failures
Voce Plans to Vote AGAINST All Three OnDeck
Directors at Upcoming Annual Meeting and Urges Fellow Stockholders to Do the Same
San Francisco,
CA (April 17, 2020) – Voce Capital Management LLC (“Voce”) is the beneficial
owner of more than 1.3 million shares of OnDeck Capital, Inc. (NYSE: ONDK) (“OnDeck” or the “Company”),
making Voce one of the Company’s largest stockholders. Today, Voce announced that it is planning to vote all of its shares
AGAINST the reelection of the three “Class III” Directors at the May 7, 2020 Annual Meeting of Stockholders (the “Annual
Meeting,”) and is strongly urging its fellow stockholders to do so as well.
Voce issued
the following letter to OnDeck stockholders in connection with the announcement:
The Pyramid | 600 Montgomery
Street
San Francisco, CA 94111
(415) 489-2600 tel
(415) 489-2610 fax
www.vocecapital.com
|
|
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 2 of 15
|
|
|
April 17, 2020
Dear Fellow Stockholders of OnDeck Capital, Inc.:
Voce Capital
Management LLC (“Voce”) is the beneficial owner of more than 1.3 million shares of OnDeck Capital, Inc. (NYSE: ONDK)
(“OnDeck” or the “Company”), making us one of the Company’s largest stockholders.
As
investors in OnDeck for the past year, we’ve actively and regularly engaged with management, consistently providing
recommendations for how OnDeck could improve stockholder value. Those discussions have intensified in recent weeks and, at
our request, have broadened to include dialogue with several members of OnDeck’s Board of Directors (the “Board”). We have
appreciated the professional and constructive tenor of these interactions.
But soothing
words alone are not enough, particularly for a company with issues as urgent, and chronic, as OnDeck’s. Stockholders have
suffered from years of strategic drift and expense profligacy, resulting in the loss of 93% of the stock’s value since its
IPO just over five years ago. As far back as August of 2019, we urged management to tighten
its focus and enhance profitability:
“We believe OnDeck’s
core value proposition is strong, but significant work needs to be done to reestablish credibility with investors. OnDeck is clearly
stuck in the ‘dugout’ at the present time. A series of disappointments and surprises has left investors ‘dazed
and confused.’ Too many new initiatives, shifting priorities and mixed signals [have left] investors uncertain whether OnDeck’s
opportunity set is growing or shrinking.”1
We fully appreciate
the challenges that the unprecedented COVID-19 pandemic poses to OnDeck, and many other companies that we own, and have been mindful
of them in our dealings with the Company. Yet OnDeck’s slow response to current events has only magnified the issues that
now confront it. We met with management over a month ago as the crisis erupted, and pleaded with them to initiate immediate expense
reductions in order to protect the Company’s ability to continue to serve its core customer base of small US businesses;
yet so far, no action has been taken, further imperiling the Company.
_____________________
1 “OnDeck:
An Owner’s Perspective” (August 29, 2019), p.2. We understand our presentation was subsequently shared with the Board.
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 3 of 15
|
|
|
We also believe
that OnDeck’s weak corporate governance has enabled and prolonged these long-running problems, and that lasting progress
cannot be made until this issue is addressed. However, out of further sensitivity to the operating environment, we committed that
we would fully support the Board at this year’s Annual Meeting if it took only one step: include and recommend a proposal
to de-stagger the Board immediately so that all Directors would be subject to reelection every year, starting at the 2021 Annual
Meeting, thereby deferring some of these issues for a year while preserving stockholders’ ability to hold the Board accountable
for its performance. We viewed this proposal as measured and proportional, taking all things into account.
Incredibly,
the Board refused.
In light
of this most recent example of the Company’s failure to align what it does with what it says, we felt we
could not simply stand idly by for another year. At a certain point, even a health crisis like the current one cannot be used
to shield a Board from any accountability or responsibility. We therefore intend to vote all of our shares AGAINST the
reelection of the three “Class III” Directors at the May 7, 2020 Annual Meeting of Stockholders (the
“Annual Meeting”) and are strongly urging our fellow stockholders to do so as well.
Batter, batter, what’s the
matter?
OnDeck’s
problems began long before the tragic and unforeseeable global pandemic erupted.2 It debuted in a splashy December 2014
IPO, with a $1.3 billion initial valuation, that seemed to validate the hopes and dreams of the venture capitalists that had funded
and promoted OnDeck as a fin-tech unicorn.
But it never worked out that way, at least
not for OnDeck’s public investors. In less than two years, OnDeck plunged from $20 per share to $4. In six calendar years
as a public company, it has had only two years with positive returns for stockholders, one of which was negligible. The stock’s
losses during this time were staggering and consistent:
________________________________
2 Unless otherwise noted, all prices are as of April 15, 2020. We
generally consider February 24, 2020 as the day the markets began to react to the crisis, and use February 21 the “unaffected”
price.
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 4 of 15
|
|
|
Other than
a brief period when the stock responded favorably to expense reductions forced by public pressure from
another dissatisfied stockholder, OnDeck has languished around that price (the VWAP from 1/1/17 through 2/21/20 was $5.25).
In 2019, when the Russell 2000 leapt 25%, and OnDeck’s peers returned 18%, OnDeck fell 30%.3 The total
return of its stock since going public makes it one of the worst performing IPOs of its generation.
While
COVID-19 has clearly amplified OnDeck’s challenges, even during this period OnDeck has underperformed its peers. Year-to-date, OnDeck is down 67%, compared to peers down 52% during this time. At its lows in March, it had lost more than 80% of its
value just since the onset of the virus panic on February 24. Now a penny stock – it actually has closed below $1.00 per
share several times recently, trading like an option rather than a stock – its market capitalization (well less than $100
million) and technical profile further confirm Wall Street’s lack of respect for the stock, if not the Company. It also
illustrates just how difficult it will be for OnDeck to recover unless radical and urgent action is taken now.
The boo-birds are out
Underpinning
the collapse of OnDeck’s stock has also been the implosion of its valuation, reflecting investors’ increasing loss
of faith, and sheer frustration, with its performance and direction.
Despite the
top-line growth and the maturity of its business model, OnDeck has proven increasingly unable to forecast results or manage investor
expectations. In the past five quarters (4Q18-4Q19), OnDeck met or exceeded both its revenue and adjusted net income guidance just
three times, and fulfilled market expectations on those metrics just twice. In four of those five quarters, it badly whiffed on
one or both measures, or disappointed the market with its guidance. The immediate (one-day) stock price reactions to these earnings
reports tell us everything we need to know: The stock fell in four out of the five periods (with three double-digit drops).
_____________________
3 Peers:
CIT Group, Elevate Credit, Enova, LendingClub, Marlin Business Services, OneMain, Regional Management, World Acceptance
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 5 of 15
|
|
|
OnDeck’s
inability to deliver consistent results has kept all but the most courageous of analysts in the bullpen. Only two out
of more than ten covering firms have the stomach to recommend the stock. For most, OnDeck is a perpetual “wait and see”
story and most don’t seem to follow the Company closely or to even care much; below is a representative excerpt from a Stephens
report on 2/12/20:
“2020 Another Wait and
See Year . . . 2019 closed out the year with a 41% y/y decline to EPS. While 2020 may be better, it still doesn’t return
OnDeck to 2018 levels, and we do not see any particular 2020 catalysts. Achieving 2020 guidance would send the stock higher, but
there has been enough volatility in execution . . . to discount guidance until it is achieved.”4
Over the prior
year (from 12/31/18 until 2/21/20), OnDeck’s multiple of Price to Book Value (“P/BV”) shriveled from 1.6x to
0.9x. This compares to an average P/BV of its peers of 1.3x, which stayed constant during that time. If earnings were reliable
enough to drive valuation, we would have likely seen an even more dramatic compression of its P/E multiple.
Rolling forward
to today, OnDeck’s P/BV is now 0.3x, meaning that investors have so little confidence they believe the Company’s
assets are worth only thirty cents on the dollar. As we discuss in the next section, OnDeck’s expense load is so high that
small swings in credit performance drive significant losses, making the market fully unable to trust its book value – hence
the massive haircut.
No runs, no hits . . . lots of
errors
The reasons
for OnDeck’s many strike-outs over the years aren’t hard to identify, starting with its lack of cost discipline. In
recent years, OnDeck’s expenses have grown much more rapidly than its revenues, rarely a great line-up for success.5
From 2017-19, OnDeck’s total revenues grew at a CAGR of 12.6%, yet its core operating expenses (technology and analytics;
processing and servicing and G&A) swelled 17.2% percent per year during this time, adding $42 million of expense unrelated
to sales and marketing over the two years.6 These operating expenses in total as a percent of revenue – a key
efficiency ratio – surged from 32.4% in 2017 to
_______________________________
4
Several others have recently made similar comments. KBW: “[W]e are Market Perform [hold] until . . . greater confidence in
the path toward higher levels of sustained profitability”; JMP Securities: “Compared to our estimates, operating expenses
came in ~8% higher, driving much of the earnings downside.”
5
We focus our analysis on the three-year period from 2017-19. Prior to 2017, a large part of OnDeck’s revenue came from its
“marketplace” lending platform, with a different business model and expense structure than today’s balance sheet
lending approach.
6
The sales and marketing expenses for OnDeck’s model are variable and grow with originations as new customers are acquired
(but should become more efficient over time as repeat customers increase). As such, we exclude them from our analysis of operating
expense efficiency.
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 6 of 15
|
|
|
35.1% in 2019.
Negative operating leverage, in the context of meaningful revenue growth and decent scale (almost $450 million in revenue),
suggests disorder. It’s also at odds with the commitments management repeatedly made to stockholders. “We are
also intensely focused on driving operating leverage,” promised Chairman and CEO Noah Breslow in May 2017 (1Q17
earnings call), and many other times, but it obviously hasn’t happened.7
A major culprit
of these runaway expenses can be seen in the rapid growth of OnDeck’s headcount. The 475 employees in 2017 quickly became
742 by the end of 2019, a CAGR of 25%. As a result, OnDeck’s revenue per head went from $738,000 in 2017, when the build-up
began, to $599,000 in 2019, and will likely fall much further in 2020. While these numbers are slightly better than those of specialty
lenders that maintain brick-and-mortar locations (averaging $430,000 per head), they’re significantly below those of tech-enabled,
pure-play online lenders (more than $1 million per head). It is the latter to which OnDeck must be compared, given its business
model.8
By our analysis,
which we have shared and reviewed with management, OnDeck could reduce its expenses by more than $50 million, just by bringing
its operating expenses as a percent of revenue in line with its online lending peers and returning its headcount to prior levels.
To date, OnDeck has announced no expense reductions whatsoever, one of the few companies we own to have failed
to take any action.
________________________________
7 Operating
leverage and expense discipline have long been aspirational, but elusive goals, for OnDeck. “Operating leverage is a differentiating
factor inherent within OnDeck’s technology-enabled model.” (4Q16 earnings call); “Hope is, as we go 2 to 3 years
out, we’re able to . . . increase our efficiency ratio.” (Conference presentation 11/3/16); and “We have made
the strategic decision to shift the company’s near-term focus from growing loans under management to achieving profitability.”
(1Q17 earnings call).
8 Specialty
lenders with brick-and-mortar infrastructure: America's Car-Mart, CIT Group, Consumer Portfolio Services, OneMain, Regional Management,
World Acceptance; Pure-play online lenders: Enova and Elevate.
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 7 of 15
|
|
|
Who’s on first?
Just what are
all of these employees working on? The answer leads us to OnDeck’s companion failure: its lack of strategic focus. As Chairman
and CEO Breslow wrote in his March 28, 2020 stockholder letter:
“We entered 2019 with
an ambitious agenda focused on building on the success of our US lending franchise, investing in growth adjacencies and innovating
on our core risk, technology and funding strengths. We also announced two new strategic priorities for the Company in 2019 –
increasing capital efficiency and pursuing a bank charter – and advanced each of these. . . . [W]e are well positioned in
2020 to further expand our core US lending business, scale our international operations to achieve profitability, drive value in
our ODX platform [and], advance our effort to obtain a bank charter . . . .”
Ambitious
indeed, particularly for a relatively small company. OnDeck is the market leader in online lending to small business, an
enormous market ($231 billion in loans less than $250,000), with extensive experience and a differentiated product. This
is why we invested in OnDeck in the first place. But what we cannot fathom is why OnDeck seems unwilling or unable to
bear down on exploiting this large domestic market opportunity by letting go of its many strategic distractions. In our
presentation to management last August, we cautioned that OnDeck suffered from “too many new initiatives, shifting
priorities and mixed signals” and urged that “OnDeck should focus on the ‘vital few’ initiatives
rather than the ‘useful many.’” Since then, it has not pruned, nor even pared, a single item on its
wish-list. Instead, it has continued to fritter away time and money in pursuit of “adjacent” markets, special
projects and distant geographies.
By our analysis,
OnDeck’s Australian foray has cost stockholders approximately $13 million of losses, plus an unknown sum of capital invested
that we estimate approaches $20 million. Harder to measure but costly nonetheless is the distraction of managing a business
literally on the other side of the world.9 Admittedly Canada is closer to home, but the Great White North has also
been a multi-year financial drain – the magnitude of which we do not know due to hazy disclosure, but is likely similar
to Oz, consuming profits, precious capital and valuable time. Management has never satisfied us with its rationales for these
adventures, and half-jokingly once commented that at least both countries “speak English.” Sadly, that may be their
only nexus to the core US business. OnDeck should immediately exit both of these units, eliminating substantial costs and freeing
up needed capital for its domestic business.
________________________________
9 Anyone who has ever tried
to coordinate a conference call between New York or San Francisco and Sydney quickly realizes how far away it is. We were disappointed
to hear that Mr. Breslow goes Down Under once per year, a journey that consumes more than a full calendar day each way just in
transit.
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 8 of 15
|
|
|
The justifications
for investments in other areas, such as the “ODX” platform and obtaining a bank charter, seem slightly stronger, but
result in a frustratingly similar pattern: consumption of what little profit is available through expense additions, rather than
offsetting reductions elsewhere to fund them. ODX is OnDeck’s white-label offering to help banks that struggle to serve small
business customers efficiently, but OnDeck struggles to provide ODX efficiently as well. Formerly known as “OnDeck-as-a-Service,”
the platform consumed between $5 million and $10 million of total investment prior to 2016 (before any major customer was signed);
management identified another $5 to $10 million of annual incremental spend on ODX during 2018 and 2019, but ODX likely contributed
to much more of the $42 million expense increase noted earlier than did any other initiative. Much of the expenditure on ODX over
the last two years was dedicated to customizing the platform for the use of OnDeck’s then-marquee client, J.P. Morgan (via
its Chase Business Bank). But Chase, which has the IT budget to bring such an effort in-house, last year decided to do exactly
that, and terminated ODX, which is now in run-off. OnDeck should have been efficiently marketing its ODX offering to smaller regional
banks and SBA lenders that are struggling to keep up with behemoths like Chase. They would likely have required less customization.
The pursuit
of a bank charter, announced last year, represents the latest of OnDeck’s big projects. Without access to information about
the cost, timing or feasibility of this pursuit, it’s difficult for us to analyze its merits. Our fundamental question would
be why such a purportedly important initiative cannot be executed with existing resources? We know of many talented individuals
in the Company, particularly in legal and government relations areas (and have spoken to a few), yet each year millions of dollars
of additional costs are incurred. How many more “investment years” must stockholders endure until the operating leverage
we’ve long been promised is actually realized?
Inside baseball
In the same
way that OnDeck has been unable to make the transition from its salad days as a tech start-up where profitability didn’t
matter, so too its corporate governance has not kept pace with its evolution into a publicly-traded company, with real fiduciary
duties owed to the stockholders who now own the Company. In particular, we see many indicia that suggest the Board is neither truly
independent nor properly aligned with the interests of the stockholders it is supposed to represent.
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 9 of 15
|
|
|
If one were
to make a list of the ways in which a Board could structurally insulate itself from accountability or foil any attempt at external
influence, OnDeck would almost merit a perfect score. The table (see Figure 1) speaks for itself.
First,
OnDeck’s decision to combine the CEO and Board Chairman roles is wrong as a matter of policy and is
particularly unwarranted under the circumstances. Best practices are for public company boards to have an independent
Chairman, and almost all do. A strong independent Chairman doesn’t just gavel the meetings to order or call the roll.
Rather, a Chairman performs an essential role by defining and maintaining the appropriate independence of the Board’s
functions from those of management; setting the Board’s agenda and then leading it; and overseeing, on behalf of the
entire Board, management and the CEO. In today’s uncharted environment, enterprise risk management is more important
than ever, and can only be overseen effectively by a Board with a strong independent Chairman at the helm.
We are aware
that OnDeck does have a “Lead Independent Director,” Mr. Henson, with whom we have interacted. However, we concur with
the views of proxy advisory firm Glass Lewis that “[w]hile many companies have an independent lead or presiding director
who performs many of the same functions of an independent chair (e.g., setting the board meeting agenda), we do not believe this
alternate form of independent board leadership provides as robust protection for shareholders as an independent chair.”10
________________________________
10 See
Glass Lewis’ 2020 Proxy Paper Guidelines [link].
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 10 of 15
|
|
|
The decision
to vest all of this authority with the CEO, Mr. Breslow, is also questionable. Rather than “earning it,” Mr. Breslow
was appointed Chairman at the same time he became CEO (at the age of 36), suggesting the Board awarded him the added title as a
perquisite rather than due to his experience. While bright and engaging, Mr. Breslow has never sat on another public company
board, let alone led one. Nothing indicates that he brings distinguished corporate governance expertise or is somehow uniquely
qualified to serve as the Board’s leader to the degree necessary to justify granting him this extraordinary authority.
We worry that Mr. Breslow’s outsized role may help explain why the Board has indulged management’s strategic lack of
focus, and tolerated its poor performance, for so long.
Second,
the Company’s staggered Board offends best governance practices. Directors should stand for election
every year. But here’s the irony: In every conversation we’ve had with the Board, it has agreed with us on this
point. In fact, we were told that the Board has long pondered de-staggering the Board and considers doing so every year. But
that raises a very simple question: Why hasn’t it acted? If it held strong views as to the merits of three-year terms
– as misguided as that perspective would be – at least the Board’s resolute commitment to the
status quo would be understandable. But to acknowledge a need to fix the issue and yet for several years to have considered
it repeatedly but never done it suggests either a lack of candor with us on this point, or utter ineffectiveness in doing its
job. Neither answer places the Board in a flattering light.
Third, OnDeck’s
advance notification provision for stockholders wishing to nominate Directors (the “Nomination Deadline”)
is unreasonably, and uncommonly, onerous. Under Section 2.4(a) of OnDeck’s Amended and Restated Bylaws, effective as of October
31, 2018 (the “ByLaws”), to nominate Directors, stockholders must do so “120 calendar days before the one-year
anniversary of the date on which the corporation first mailed its proxy materials.” OnDeck mailed its 2019 Proxy on April
1, 2019, meaning that stockholders wishing to nominate Directors at the 2020 Annual Meeting would have been required to do so by
December 3, 2019. How can stockholders be expected to determine whether to seek Board changes prior to receiving the financial
report of the previous year; indeed, in OnDeck’s case, the prior year wasn’t even concluded, let alone reported, to
stockholders at that time.
120 days prior
to mailing of the proxy – itself typically six weeks before the meeting date – effectively gives the Board and management
somewhere between five and six months of free reign with little worry that dissatisfied stockholders might react to adverse developments
by deciding to nominate Directors to oppose the incumbents. Such an extended period of time is atypical and unacceptable.
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 11 of 15
|
|
|
But don’t
just take our word for it: Look no further than OnDeck’s own past practice, which used to have a Nomination Deadline of 45
days, much more in-line with industry norms. After facing pressure from another vocal, dissatisfied stockholder in 2017, the Board
proactively moved to further entrench itself by nearly tripling the length of the notification period. Thus, in October 2018
the Board amended the By-laws to extend the Nomination Deadline from 45 days to the current 120. As a result, the Nomination
Deadline was pushed from February 16, 2020 all the way back to December 3, 2019. Such deliberate choices speak volumes about the
Board’s true alignment with stockholders and its willingness to be held accountable for its actions. At a minimum, the Board
should have allowed stockholders to vote on this ByLaw amendment, rather than acting unilaterally, given the adverse impact it
had on stockholder rights.
When taken
together – the many structural impediments to stockholder rights, the concentration of power and authority in the CEO and
the affirmative entrenchment efforts by the Board – we have serious concerns about the independence, alignment and accountability
of OnDeck’s Board to stockholders.
Take me out to the ballgame
The Board’s compensation
practices also raise further concerns about its independence, with far too much emphasis on short-term metrics and cash.
For example, the short-term incentive plan (“STI”) contains a highly unusual design that is crystalized
semi-annually rather than annually. This is way too brief a period to measure or incentivize performance and presents the potential
for abuse, as management could game the system through the selection of goals it already knows that it can easily meet. It could
also lead to anomalous results where management satisfies one of the half-year periods, barely, and misses the second one, dramatically
– and still receives half of the STI, even though the overall annual shortfall is much greater and would have resulted in
a lower STI if calculated annually. Yet that’s exactly what happened in 2019.11
The long-term
incentive plan (“LTI”), comprised of a variety of restricted stock units, is also excessively short-term in its approach.
We were surprised to discover in the 2019 Proxy that the Compensation Committee changed the LTI last year so that awards are earned
after only twelve-months, which then vest over the following three years. Prior to 2019, the LTI had three 12-month performance
periods, and management had to meet targets during each
of those periods in order to be compensated. This unexplained change places undue emphasis on the short-term (only one year,
odd for what is supposed to be a long-term incentive plan) and explicitly values “retention” over
performance in its vesting design. Interestingly enough, despite calling a portion of this compensation “Performance
Units” the amounts are paid out in cash (with no fluctuation in value correlated to stock price) when earned.
________________________________
11 Using
Chairman and CEO Breslow as an example, 1H19 performance missed two of three operating targets and resulted in a 75% STI payout,
while 2H19 performance met all three operating targets and resulted in a 100% STI payout; if calculated annually, the magnitude
of the 1H19 miss would have driven full year misses for two of the three targets.
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 12 of 15
|
|
|
We also question the performance
targets selected by the Compensation Committee. Each of the plans have a significant component of “Adjusted Pre-Tax
Income,” which excludes all stock-based compensation. Why exempt management from the cost of stock it uses to pay employees?12
The LTI is comprised of 40% adjusted gross revenue, with no measure of the quality of it such as credit profile or yield
(at least the STI takes into account reserve ratios). What’s also revealing is what’s absent from all of the plans:
There’s no component of return on equity, growth of book value nor any per share measure. As such, management has no incentive
nor responsibility for the good stewardship of the Company’s equity.
This isn’t
the first time we’ve addressed this with OnDeck. Last August, we advised OnDeck that from the perspective of an investor
“return on equity is the preferred valuation metric” because “ROE is the most comprehensive and elegant measure
of efficiency and financial success, [i]ntegrat[ing] income and balance sheet analysis.” We explicitly recommend it “establish
specific GAAP ROE targets for its business then publicly communicate, and commit, to achieving them.” Yet the Compensation
Plans ignore ROE altogether.
OnDeck’s
plans also fail to align management’s incentives with stockholder interests by relying far too heavily on time-based
vesting. The RSUs, when awarded, have no additional performance hurdles; the executive need only remain employed at the
Company to ultimately vest. In the case of Mr. Breslow, he gets accelerated vesting on 50% of his units even if he’s terminated
without cause. Compounding this misalignment, another concerning change in 2019 was the Compensation Committee’s decision
to eliminate stock options altogether and move exclusively to stock-based awards. While a healthy debate can always be had about
stock versus options, it matters greatly how they fit into the overall design of the plan. OnDeck’s LTI restricted stock
units vest exclusively based on time; no subsequent performance is required thereafter. This purely time-based vesting means that
all the executive must do is remain employed at the Company to vest his or her awards. At least with a healthy dose of options
in the mix, some inherent performance hurdle accompanies the awards (the stock must at least appreciate above the strike price
of the option when granted to have any value).
________________________________
12 We
repeatedly emphasized this to management in our August presentation. For example: “GAAP ROE is our preferred measure. Excluding
SBC from in-period ROE means these costs are completely unaccounted for.” (p.28)
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 13 of 15
|
|
|
Finally, the Compensation
Committee’s undue generosity has resulted in too rich a mixture of cash in the overall compensation scheme.
While the absolute dollars are not outrageous, the steady growth of Mr. Breslow’s salary seems anomalous given the performance
of the stock under his tenure. Since the IPO, the Board has nearly doubled his base salary. As a result,
over the past three years approximately 54% of Mr. Breslow’s total compensation has been paid in cash.13 Once
again, we believe this fails to incentivize long-term performance and does not align executive incentives with those of stockholders.
A schwing and a . . . miss
While the Board
has suggested that our requests for change are untimely and inconvenient, the facts don’t support that narrative. As a long-time
stockholder we have been constructively engaged with OnDeck for many months over steps we believe that it could take which would
protect and unlock stockholder value. As OnDeck began to stumble earlier this year, we escalated this dialogue. On March 13, we
presented management with detailed proposals for immediate steps, especially significant cost reductions and a culling of strategic
priorities; over a month later, we’ve seen no announcements indicating that any of them have been acted upon. We also requested
at that time a meeting with the Board to discuss our strategic and corporate governance concerns, which didn’t
occur until ten days later (March 23). On that call, we requested to speak with the Chairs of the Compensation and Nominating and
Governance Committees to get answers to questions that we were told could best be obtained from them, which took another two weeks
to occur (April 3).
After listening
carefully to the Board’s assurances that it had heard our concerns and found many of our ideas to be sensible and already
reflected in its thinking, and taking into account what are admittedly challenging circumstances for all involved, the following
Monday (April 6) we made one simple proposal: We asked that “the Board include and recommend a proposal at the 2020 Annual
Meeting of Stockholders to de-stagger the entire Board immediately so that all Directors will be subject to reelection every year,
starting at the 2021 Annual Meeting.” As we explained, our Proposal was tailored to give the “Board and management
time and space to act while providing stockholders with an appropriate accountability mechanism, should one become necessary.”
In other words, trust but verify. We asked for nothing else to secure our full support at this year’s Annual Meeting.
________________________________
13 Cash
includes his salary, non-equity plan compensation and “all other” compensation categories.
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 14 of 15
|
|
|
Consistent
with its past practice of considering a Board de-stagger and finding it meritorious while mysteriously failing to
actually act upon it, a week later the Board rejected our lone proposal, but not before telling us that it wasn’t
opposed to it and might do it next year. With all due respect, the letter of talking points we received likely took longer to
draft than would have the de-stagger proposal itself. We also reject the suggestion that we dawdled or waited too long to
raise our concerns with the Board. OnDeck didn’t even file its proxy until March 18. Many of the concerns we raised in
our conversations with the Board, and are reflected in this letter, arose from our review of that document, which had been
out for a mere two weeks when we made our proposal. Even today, an entire month remains until the Annual Meeting.
Three strikes
Which
brings us full circle. As a result of the Board’s decision to set the advance notification deadline so far in the past,
it was impractical for any stockholder to nominate Directors at this year’s Annual Meeting, notwithstanding the
strategic blunders and financial crisis engulfing the Company since the start of the year (and pre-COVID). Moreover, the
Board’s dogged adherence to its staggered structure where at most three of its members must stand for election
each year forces stockholders to not only decide whether to nominate Directors, but when to do so; the years
actually matter, because if the terms of the Directors most responsible for the Board’s shortcomings happen to expire
in a given year, they will likely buy themselves another three years of protection if they remain unopposed.
Looked at in
this way, the 2020 Annual Meeting is indeed a moment in time. The “Class III” Directors who are standing for reelection
this year are those that are most directly implicated by the nexus of our concerns about OnDeck’s lack of strategic focus,
financial laxity and poor corporate governance:
|
§
|
Noah Breslow, age 44, Chairman and CEO, and a Director since 2012;
|
|
§
|
Jane J. Thompson, age 68, Chairman of Corporate Governance and Nominating Committee, and a Director
since 2014; and
|
|
§
|
Ronald F. Verni, age 71, Chairman of Compensation Committee (and a member of Corporate Governance
and Nominating), and a Director since 2012.
|
All three
of these individuals have been on the Board since prior to the Company’s IPO, and they share a common track record of
terrible returns for stockholders as represented by the TSR analysis earlier (all three were also present when the Board
voted to unilaterally amend the ByLaws to triple the length of the advance notification deadline after the last run-in with
stockholders). They are three of the longest-tenured Directors, but most importantly,
they each hold key leadership roles on the Board – as CEO/Chairman and Chair of the two Committees that have
most failed stockholders – making them uniquely deserving of responsibility for the Company’s myriad
shortcomings. Withholding votes in favor of each of their election this year will be stockholders’ only way of
holding them accountable until 2023, given the staggered Board and three-year terms.
Some may ask,
“what good does voting ‘no’ do without a competing slate of nominees to replace the incumbents?” Having
conducted many proxy contests in the past, Voce inarguably has the wherewithal to nominate and elect directors when given the
opportunity. As noted, however, the Board’s unreasonable advance notification deadline has made that impossible this year
and the Board has banned all other forms of stockholder action, such as calling a special meeting or acting by written consent.
Fellow Stockholders of OnDeck Capital, Inc.
April 17, 2020
Page 15 of 15
|
|
|
But one of the
few positive aspects of OnDeck’s corporate governance structure is that it contains a “majority voting” standard
for uncontested elections such as the one this year (at least for now – one wonders whether the Board will “fix”
this provision too with another ByLaw amendment). Even when running unopposed, Directors must receive more votes in favor of their
election than withheld or voted against. Directors who do not receive a majority of the votes cast must tender their resignation.
In this year’s Proxy, the Board toots its own governance horn for having “a majority vote policy in uncontested director
elections and maintain[ing] a director resignation policy.” We fully expect the Board to honor that policy by accepting the
resignation of any of the Class III Director who does not obtain a majority of the votes cast. We will seek to hold accountable
any continuing Directors that fail to heed the will of stockholders.
Shutout
We
therefore urge our fellow stockholders to join us in voting AGAINST all three Class III Directors by withholding votes for
them on the Company’s proxy card.
Respectfully yours,
VOCE
CAPITAL management LLC
By:
|
/s/
|
|
|
J. Daniel Plants
|
|
|
Chief Investment Officer
|
|
About Voce Capital Management LLC
Voce Capital Management LLC is a fundamental
value-oriented, research-driven
investment adviser founded in 2011 by J. Daniel Plants. The San Francisco-based firm is
100% employee-owned.
Media Contact:
Sloane & Company
Dan Zacchei / Joe Germani
dzacchei@sloanepr.com / jgermani@sloanepr.com