[TYL – Tyler Technologies] Finding Common Ground in Government (Systems)

In the course of researching Tyler, I came across this synopsis of the company’s distant history.  If the laudatory retrospectives that we all know and love – the ones that start with outcomes and back into causes – were written in the late ’80s, Tyler Corp. might have been featured as an unexampled success story, offering critical insights for investors and operators.  In 1966, Tyler’s founder Joseph McKinney sold a collection of military suppliers that he had acquired the previous 6 years and embarked on a debt fueled buying spree – scooping up disparate industrial businesses involved in freight management, sewage pipe manufacturing (Tyler Pipe), electronic parts distribution (Hall-Mark Electronics), specialty chemicals (Reliance Universal), and industrial explosives production (Atlas Powder) – that would culminate, 20 years later, in a Fortune 500 company with soaring profits, $1bn+ of revenue, and leading market shares in most of its primary segments.  But with valuations soaring in the midst of the late ’80s LBO frenzy, McKinney spent the next several years divesting most of the businesses he had spent the prior 20 years acquiring, returning most of the sale proceeds to shareholders.

Sounds like an “outsider”, right?  But if you’re like me before I started digging into this company, I doubt you’ve ever heard of Joseph McKinney.  The media will gloss over a manager who deliberately downsizes to near extinction, even if doing so is likely the best way to preserve shareholder value.  [Henry Singleton, founder of Teledyne – another conglomerate whose halcyon days overlaps with those of Tyler’s – is a rare exception, with memory of his capital allocation feats reinvigorated by Buffett and subsequently preserved by Buffett acolytes].  But as importantly, McKinney did himself no favors by sticking around.  He should have pulled a Singleton and bailed at the top (Singleton stepped down as CEO in 1986 and retired from day-to-day management in 1989, just before Teledyne became embroiled in corruption scandals and lawsuits).  Instead, he continued to lead the company he had built over the next 8-9 years, during which time Tyler Pipe – one of the few remaining pieces of Tyler following the divestitures and once the centerpiece of McKinney’s industrial conglomerate – was besieged by environmental lawsuits.  Worse still, Tyler acquired two businesses in the early 90s – an auto parts retailer (Forest City Auto) and, um, a direct marketer of products sold for school fundraising programs (Institutional Financing Services) – that quickly turned out to be duds.  Sales declined, losses mounted, goodwill was impaired.  Tyler’s pipes and logistics businesses were sold in 1995; Forest City and IFS followed a few years later.  Along the way, McKinney unceremoniously left (was booted from?) the company.

So then in 1997 this guy Louis Waters, founder of waste management company Browning-Ferris Industries, invested $3.5mn in Tyler, taking 10% of the company.  With all the dross of Tyler’s past cleared away, Waters set an entirely new mandate for the company: to become an information services provider for local governments.  This description didn’t quite take, as Tyler tweaked itself in the subsequent years to be less about information management and record keeping and more about software…but anyways, the point is that in 1998 and 1999, Tyler rolled up a bunch of government IT/software companies – including Munis, the ERP division now responsible for nearly half of Tyler’s revenue – that would form the backbone of what the company is today.  By 2000, Tyler was a technology company with $93mn in revenue and a few million in EBITDA serving local governments.  With its identity firmly established, Tyler then transitioned from this roll-up, deal making phase to actually building a software company.  By 2003, the financial opportunists who cobbled these businesses together had moved on (Louis Waters retired as Chairman in 2002).  John Marr, who came from Munis and knew the business of software, took over as CEO in 2004.  Since then, most of Tyler’s acquisitions¹ have been small tuck-ins with $5mn-$15mn in revenue, used to fortify capabilities and cross-sell systems rather than to consolidate cost, though it has done a few big, one off deals in order to breach a new category or to complement an existing one.  Most of Tyler’s revenue growth over the last few decades (~13%-14%/year) has been organic (11%/year).

You will not be surprised to learn that local governments, staffed by risk averse bureaucrats allergic to change and with little incentive to optimize, do not scrutinize the ROI of their existing processes and systems to nearly the same degree as profit-seeking enterprises.  They will hold on to decades-old infrastructure to the point of absurdity, when it either ceases to function, doesn’t allow users to do basic, widely expected things like pay traffic tickets or download data into Excel, or is no longer maintained by outside vendors, who have moved on to supporting more modern software.  The average system that Tyler replaces is over 20 years old.  The court and property tax systems of Cook County, the second largest county in the country, were built in-house and run on 40-year old mainframes before Tyler replaced them in two $30mn deals.  There are hundreds of thousands of these systems used by over 36k US towns and cities, 2/3’s of which were sold by vendors who are no long competitive in the marketplace today, either because they are defunct or haven’t themselves upgraded their technology in the last few decades.  That’s not to say Tyler is wowing the world with its tech.  I suspect that the pace of innovation at Tyler reflects the needs of the customers it serves.  I doubt the company’s capabilities compare favorably to those of the hyper growth SaaS companies that investors fawn over.  In the demos I’ve seen, the user interface for Munis, its flagship application, needs work.  But so what?  Tyler’s technology is good enough for the market it is addressing.  Your bleeding edge cloud and machine learning capabilities will not compel a local government to act.  A public agency’s RFP cadence adheres to the decay rate of legacy technology, and its sales cycles, especially if we’re talking about a large government entity, can extend to 3 years.