Despite the stock’s substantial appreciation in 2007, MFW remains extremely undervalued with significant downside protection. The market is currently ascribing little value to this poorly-understood provider of services to financial institutions (Harland Clarke) while the stability of the licorice business (Mafco Worldwide) and the current valuation of 5.9x EV/EBITDA and a 19% levered free cash flow yield represents downside protection.

 

Both businesses have over 100-year histories, long-standing customer relationships, high levels of consistent cash flow, and experienced management teams with properly-aligned incentives.  The debt position on the operating companies gives the equity significant upside potential (100%), while the stability of cash flows and asset value minimizes the downside.

 

MFW has been posted several times over the past eight years, but since the business has changed so much, a refreshed view is necessary to understand the current situation and dispel several myths associated with the company.

 

As a company without analyst coverage and without an official investor relations department, MFW takes some time to figure out.  While I will go through more complicated (and complete) valuation metrics below, it’s worth starting with a simpler demonstration:

 

9-months ended 9/30/2007
CFFO	$133.3
Capex	$32.8
Free Cash Flow	$100.5
Annualized FCF	$134.0
Levered FCF Yield	15%

 

A 15% free cash flow yield is nice, but what if I told you that the 9-month figures above only included 5 months of John Harland, a larger business than pre-acquisition MFW? Once we add in the missing 4 months and easily-attainable cost synergies, we know that FCF yield will be much larger than 15%.

Now that I have your attention, allow me to describe the businesses.

 

Harland Clarke is comprised of three businesses:  (a) the largest check printer in the United States serving clients like Bank of America, PNC, Wells Fargo, Intuit, and USAA; (b) Harland Financial Solutions (HFS), which is a leading supplier of software and services to financial institutions; and (c) Scantron, which provides data collection products and services, as you’re probably familiar with from high school standardized tests. 

 

Harland Clarke Revenue and Profit Segmentation
LTM Period Ended 9/30/2007
	Revenues	Adjusted EBITDA (a)	Adjusted EBITDA %
Printed Products	$1,295	$323	25%
HFS	$325	$71	22%
Scantron	$80	$26	32%
Corporate		-$34
Total	$1,700	$386	23%
Source: Harland Clarke presentation from Credit Suisse Credit Conference, 11/15/2007
(a) Adjusted for acquisition-related expenses, restructuring expenses, and purchase accounting adjustments (excluding amortization of upfront contract payments)

 

Harland Clarke was created through the acquisition of the #3 check printer, John Harland, by the #2 check printer, MFW’s Clarke American in May 2007. Together they have roughly 60% market share in check printing, which is now a duopoly with Deluxe (NYSE: DLX).  While the check printing industry is in a long-term decline, Harland Clarke has high-quality operations and pricing power that will allow it to succeed even in a declining market as well as a strategy to expand beyond the check printing market to other value-added services for its existing financial services clients.

 

 

Yes, fewer people are writing checks.  The check printing industry is in a long-term decline due to the conversion from checks to electronic forms of payments.  A December 2007 study by the Federal Reserve reported that check-based transactions declined 4.1% per year from 2003 to 2006.  However, since checks are ordered at other times (name and address changes, principally), actual check volume declines may be somewhat slower than that pace.  Nontheless, there were still 30.6 billion checks written in 2006.

 

But checks actually generate revenues for banks.  In addition to selling checks directly to consumers, the check printers partner with financial institutions in a revenue-sharing arrangement (around 50%/50%).  Check printing requires large-scale operations to ensure the quality, consistency, and security that banks require.  The industry has consolidated over the past 25 years to mirror the consolidation of their financial institution clients.  Clients change their printers based on price, customer service, and quality, but typically sign up for 3-5 year contracts.

 

Check printers are adapting.  Due to declining check usage, the check printers have expanded into other lines of business.  Deluxe has focused on small business services, which has increased from 16% of revenues in 2002 to 58% currently.  Harland Clarke is focusing on expanding its relationships with the financial institutions in order to provide custom marketing campaigns and other value-added services.

 

 

Harland Clarke has the background and expertise to be the lowest-cost operator in the industry while maintaining the high quality and security that banks require.  In 2001, Clarke American won the prestigious Malcolm Baldridge Award, a national quality award given by the National Institute of Standards and Technology to organizations that have the highest achievements in quality standards.  It has been awarded to only 68 organizations out of 1,063 applicants since 1988.  While Clarke American won the award 7 years ago, the quality principles that allowed it to win the award are still alive and well in the organization, demonstrated by the increasing EBITDA margins (from 21% in 2002 to 23.4% in the LTM period prior to the JH acquisition) in a declining industry and the reduction of printing facilities from 14 to 7 from 2002 to 2006 while increasing share from 25% to 30%.

 

As seen in the table below, Clarke American was the class of the industry from a margin perspective.  Keep in mind that Clarke American did not have the benefit of highly profitable other segments like JH and DLX did.

 

Comparable Income Statements, Q2-Q3 2006 ($ in millions)
(last 2 reported quarters prior to JH acquisition announcement)
	Clarke American	John Harland	Deluxe
Revenue	$312	$516	$801
Adjusted EBITDA	$71	$105	$156
EBITDA Margin	23%	20%	19%

 

Clarke American has been a case study for high-quality operations in numerous books and publications. Most notably, I recommend reading chapter 6 of the aptly titled book, How to Grow When Markets Don’t which describes Clarke American’s transformation from just another check printer to a value-added business partner to its financial institution clients.

 

 

Harland Clarke Printed Products Segment Trends
(Pro Forma for Clarke American and John Harland combination)
	2004	2005	2006	LTM 9/30/07	CAGR
Revenues	$1,055	$1,234	$1,274	$1,295	8%
Adjusted EBITDA	$222	$271	$294	$323	15%
Margin	21.0%	22.0%	23.1%	24.9%

 

In this supposedly declining industry, Harland Clarke’s revenues have increased 8% annually since 2004, and their adjusted EBITDA margins have improved from 21% to 25%. In addition, capex is low: approximately 10-15% of EBITDA.  How have they done that?

 

As How to Grow When Markets Don’t describes, “by expanding its offerings from simple check printing into a broad array of bank services, [Clarke American] has enlarged its potential market space from some $1.8 billion to $14 billion - an eightfold increase.”  In 2006, approximately 35% of Harland Clarke pro forma revenues came from non-check products, including $187m from marketing and contact center services affiliated with their check products.

 

Through its call centers, Harland Clarke handles customer service calls (stop payments, account inquiries, etc.) and lead generation (direct deposit, insurance products, etc.) for banks.  Since Harland Clarke is already a trusted partner of their financial institution clients, it can more easily convince a bank that is ready to outsource those services to Harland Clarke rather than to some other call center.  In addition, since Harland Clarke’s core competency is mass personalization of printed products, they create customized direct marketing campaigns for their financial institution partners.

 

The growth strategy is working.  As seen in the table below, Clarke American itself has recently accelerated its growth over the past few quarters.  As a reminder, the Clarke American management team that produced these results is now running Harland Clarke.

 

Clarke American Standalone Revenue Trends
	Q1 2006	Q2 2006	Q3 2006	Q4 2006	Q1 2007	Q2 2007	Q3 2007
Revenues	$163	$156	$155	$150	$165	$165	$169
Y/Y Growth Rate	6%	2%	-2%	-2%	1%	6%	9%

 

 

Though check usage is declining, this segment is very stable.  Harland Clarke sells to 15,000 financial and commercial institution clients with 85% of revenues under long-term contracts.  The top 20 clients represent only 42% of revenues, with Bank of America representing just under 10% of revenues.  The BofA contract renews after 2010 and is sole-sourced to Harland Clarke (Clarke American won it all a few years ago).  Also, as you would expect from an exceptional operator, retention rates are high.  Before the John Harland acquisition, Clarke American had lost two accounts of significant size in the prior 11 years.  Incidentally, the BofA relationship is over 30 years old, which pales in comparison to its oldest relationship, a 129-year old relationship with Frost Bank.

 

Harland Clarke’s stability also comes from its pricing power.  Most people don’t notice what they spend on checks, since 80% of personal check users still purchase checks through their banks, which debit their accounts with the purchase.  As a result, most check buyers don’t even know the cost.  In addition, checks are sold in “box” quantities, which allow check providers to increase their revenues/check by decreasing the number of checks in each box.  While subtle, this is a form of pricing power often used by packaged food companies (especially Frito-Lay).

 

 

Harland Financial Services (HFS) Segment Trends
	2004	2005	2006	LTM 9/30/07	CAGR
Revenues	$235	$287	$322	$325	13%
Adjusted EBITDA	$45	$59	$61	$71	18%
Margin	19.1%	20.6%	18.9%	21.8%

 

HFS targets financial institutions with assets between $1b and $10b.  While large banks will build their own systems, smaller banks will only buy.  The replacement cycle on their software is 5-10 years, though the cycle for core processing systems is sometimes even longer due to the complexity and risk associated with system conversions.  Some of HFS software products include:

 The HFS business is highly diversified and stable.  The top 20 customers represent less than 6% of revenues, and 75% of their revenues are recurring.  They compete with Jack Henry (9.6x LTM EBITDA), Fiserv (9.2x LTM EBITDA), Metavante (6.6x LTM EBITDA), Visual Insight, and other smaller companies.

 

While Harland Clarke has not hinted that they would sell HFS, the financial institution software industry has consolidated over the years, and HFS would be an attractive asset to many competitors.

 

 

Scantron Segment Trends
	2004	2005	2006	LTM 9/30/07	CAGR
Revenues	$75	$74	$76	$80	2%
Adjusted EBITDA	$26	$23	$25	$26	-1%
Margin	34.9%	31.4%	33.2%	32.3%

 

We all know Scantron for their school-based testing forms for a good reason: Scantron products are in 80% of all public schools in the U.S. They have 100,000 accounts and the largest represents approximately 2% of revenues.

 

Scantron’s future growth comes from increasingly mandatory school-based testing and enterprise-wide testing and surveying.  The penetration into markets beyond schools is relatively recent, but Scantron has already penetrated corporate clients such as JPMorgan, Wells Fargo, Wachovia, and Fedex.  Scantron is led by Jeff Heggedahl, who used to run John Harland’s check printing business ($650m revenues), so he brings great relationships to financial institutions.

 

 

As the quarters progress, Harland Clarke should pay down its debt faster than expected, which is what occurred at Clarke American prior to the John Harland transaction.  Harland Clarke’s term loan facility requires that 50% of excess cash flow is used to pay down debt.

 

Clarke American acquired John Harland on May 1, 2007, in a cash deal that was financed with almost 100% debt.  The terms of the debt are very attractive and manageable since it was raised at a very good time in the credit markets (for borrowers, at least).  In fact, Wachovia Capital Market’s analysis of Harland Clarke’s debt was positive on the business fundamentals but negative on the “loose bond covenants,” and especially a provision that would “allow Harland Clarke to dividend out asset sale proceeds from the sale of any property or assets other than Printed Products (i.e.: Scantron and HFS) as long as pro forma leverage either (i) does not worsen, or (ii) does not exceed 4x.”  As an equity holder, that provision is fine with me, as long as they can cover their fixed charges, which should not be a problem, as shown in the table below.

 

One of the reasons the MFW stock performed so well in 2007 was a result of the estimated $112.6m annual synergies ($5.18/share) between Clarke American and Harland Clarke.  The company has stated as recently as November 15 that they are “on target to achieve” those synergies.  Through five months of integration (through 9/30), Harland Clarke has actually achieved (on a run-rate basis) over 50% of the total expected 24-month synergy target.

 

Incidentally, there is no cross-default or cross-collateralization between the Harland Clarke and Mafco Worldwide debt.  Harland Clarke also owns many of its properties, including a 21-acre wooded campus outside Atlanta.

 

Harland Clarke Capital Structure and Liquidity
($ in millions)
	LTM	Q3 Run-Rate (b)
Reported Adjusted EBITDA	$385.5	$454.0
Synergies Yet to be Achieved	$96.0	$57.4
Total Synergy-Adjusted EBITDA	$481.5	$511.4
Long-Term Debt (9/30/07)	$2,394.0
Cash (9/30/07)	$187.3
Net Debt	$2,206.7
Net Debt/EBITDA (no future synergies)	5.7x	4.9x
Net Debt/EBITDA (with future synergies)	4.6x	4.3x
Net Interest Expense (Q3 annualized)		$202.4
Mandatory Debt Repayment (1% quarterly till 2014)		$72.0
Capex (Q3 annualized)		$28.4
Next 12-Month Fixed Charges (a)		$302.8
EBITDA (w/o synergies) / Fixed Charges		1.5x
EBITDA (w/ synergies) / Fixed Charges		1.7x
(a) "Fixed Charges" per Harland Clarke debt documents only include Interest Expense and cash dividends.  The interpretation above is more strict than the debt documents.
(b) Difference between LTM and Q3 Run-Rate is due to full-year effect of John Harland and synergies already implemented and achieved

 

 

Mafco Worldwide is the dominant producer of licorice extract and other licorice derivatives in the world.  Its licorice extract is primarily used as a flavoring of tobacco products (71% of revenues), and it also sells licorice extract to confectionary markets and a sweetener substitute for sugar or saccharin.  Mafco has exceeded $30m of free cash (EBITDA less capex) each year for the past 10 years (as long as data are available) and will likely do so again in 2008.  Mafco likely has 80-90% market share in the global licorice extract industry. 

 

 

Licorice extract has been a component in cigarettes for over 100 years, used as one component of many to give a particular cigarette its distinctive taste.  Due to the extreme customer loyalty (aka: addiction) to a particular flavor, cigarette manufacturers do not change their cigarette recipes.  Licorice extract represents probably less than a penny of the cost of a cigarette.

 

Licorice extract comes from licorice root, which grows only in the Eastern Hemisphere between the 34^th and 38^th parallel.  People have tried unsuccessfully to farm it elsewhere.  Licorice root is essentially a self-propagating weed whose root can be dug up at any point – there is no specific harvest season. 

 

China is the largest producer, but a large percentage of Chinese herbal remedies and medicines contains licorice root and extracts, and the supply of licorice root is controlled at the central government level, with almost all consumed internally.  Afghanistan used to be a big supplier, but now the Afghanis do not want to be bothered with the back-breaking labor involved in collecting licorice root (poppy’s a better ROI).  Iraq is in the growing zone, and US presence in Iraq could encourage trade with Iraq, but progress is slow.  Italy, Spain, Turkey and Greece used to have larger amounts of licorice root, but it has largely given way to development over the years.  Syria used to be a bigger producer, but there are obvious political issues there now, and most of their licorice root goes to Egypt.

 

 

As a result of limited supply from other countries, most of the commercially available licorice root currently comes from three countries: Azerbaijan, Turkmenistan, Uzbekistan.   Since most of the supply comes from these three countries and limited suppliers, Mafco plays these countries/ suppliers against each other to keep them in check, as well as using its large inventory of licorice root (three years) to wait out any disruptions or attempts at pricing increases.  Since Mafco needs its suppliers, though, it does not squeeze them to a great extent even if given the chance, even though Mafco represents 85-95% of the licorice purchases from Azerbaijan, Turkmenistan, and Uzbekistan.

 

The key to understanding the success of Mafco is that the company has over 100 years of experience in dealing with this unstable supply chain.  They have dealt with cargo ships sunk by German U-boats, expropriation of their equipment by the Russian Revolution, sanctions against Iraq, sanctions against Iran, unpredictability from Syria, and wars all over the place.  Their suppliers sometimes put metal objects including grenades, mortar shells, and metal hand tools in order to increase the weight of the bales of licorice root.  This is not an industry for newcomers.

 

 

Despite declining tobacco consumption and customer concentration (top 10 represent ~69% of revenues), Mafco Worldwide has remarkably stable revenues and cash flows.  The stability is a result of their dominant market share, consistent product, and the relative low cost of licorice extract in their customers’ cost structure.  Mafco has been successful in selling its sweetener product, Magnasweet, to offset declines in the tobacco segment.  If you have tried Stride or Trident Splash chewing gum, you have tasted Magnasweet, which makes the sweetness last longer.

 

At 1.1x Debt/EBITDA, there is even more cushion at Mafco than at Harland, and an even less restrictive debt amortization schedule – only $1.1m is mandatory each year.  Despite the lower amortization requirements, Mafco consistently pays down their debt faster than required.  Like the Harland Clarke debt, the Mafco debt contains no cross-defaults or cross-collateralization.

 

In addition to the consistent financials, Mafco also owns three real estate properties that likely have significant value that is not accounted for on the company’s books, considering their age.  They own facilities in Gardanne, France; Richmond, Virgina; and Camden, NJ.  The Camden site, located on almost 9 acres right on the Delaware River across from Philadelphia, is likely the most valuable, and likely on the books at minimal value, since it has been the principal site of the Mafco operations for over 100 years.

 

Mafco Worldwide ($ in millions)
	1997	1998	1999	2000	2001	2002	2003	2004	2005	2006	YTD RR 2007
Revenues	$100.4	$99.8	$97.3	$93.1	$98.4	$96.9	$95.7	$93.4	$97.3	$98.1	$102.7
Growth Rate		-0.6%	-2.5%	-4.3%	5.7%	-1.5%	-1.2%	-2.4%	4.2%	0.8%	4.7%
Gross Profit	$46.4	$48.4	$46.1	$43.9	$46.8	$49.6	$49.4	$48.3	$48.2	$47.3	$47.9
Gross Margin	46%	48%	47%	47%	48%	51%	52%	52%	50%	48%	47%
EBITDA (a)	$31.4	$38.5	$37.5	$38.3	$35.9	$43.9	$40.5	$41.0	$39.7	$38.7	$37.3
EBITDA %	31%	39%	39%	41%	36%	45%	42%	44%	41%	39%	36%
Capex	$2.3	$3.4	$1.1	$1.1	$1.1	$0.8	$2.1	$1.6	$1.1	$0.8	$1.6
Free Cash Flow (b)	$29.1	$35.1	$36.4	$37.2	$34.8	$43.1	$38.4	$39.4	$38.6	$37.9	$35.7
Q3 Debt	$73.2
Q3 Cash (c)	$31.4
Net Debt	$41.8
(a) YTD RR based on YTD 9/30/07 financials
(b) EBITDA less Capex
(c) Represents cash on Mafco or corporate (MFW) balance sheet as of 9/30/07

 

 

As mentioned at the beginning of the write-up, MFW had an annualized FCF yield of 15% through the first 9 months of 2007, which only accounts for 5/9 of John Harland performance.  So what is the normalized free cash flow yield?  It is 19% including all projected synergies (so far).  Recall that in just 5 months they already achieved over 50% of their 24-month synergy target, so it is a question of when, not if, the synergies will be achieved.

 

Normalized Free Cash Flow and Valuation Analysis (a)
Harland Clarke Full Quarter Adjusted EBITDA	$113.5
Mafco EBITDA	$8.6
Recurring corporate costs	-$5.4
Q3 EBITDA (includes some net synergies)	$116.7
Less Net Synergies Achieved in Q3	-$13.8
Less Interest Expense	-$52.9
Less Capex	-$8.0
Less Normalized Cash Taxes (b)	-$10.8
Levered FCF (no synergies)	$31.2
Annualized FCF (no synergies)	$124.9
Annualized FCF Yield	14%
Total Targeted Synergies	$112.6
Effective Tax-Rate on Synergies	36%
Annualized FCF Yield (full synergies)	19%
Fully-diluted shares outstanding	21.4
(a) MFW reported Q3 free cash flow is much higher due to tax refunds and other transaction-related items.  This analysis attempts to normalize FCF.
(b) Harland Clarke received a refund in Q3 related to tax-deductible transaction expenses.  Assumes 36% tax and $20m of tax-deductible depreciation

 

Based on the stability of the business, world-class management team, and the fact that Harland Clarke has already implemented over 50% of the synergies on a run-rate basis, there is little downside in MFW stock, unless you think it will trade at a 25% free cash flow yield. As for upside, if you think that they will achieve 100% of the targeted synergies and it will trade at a 11% free cash flow yield, MFW is an $84 stock (100% higher than the 1/15/08 price).  This ignores, of course, the fact that MFW has actually grown over the past few years and will likely continue to grow as they pass on price increases and expand their margins.  This also ignores the probability that they will exceed their initial synergy targets.

 

		% of Potential Synergies ($112.6m total)
		(In Q3, they achieved $13.8m net synergies.  On a run-rate basis, that's over 50%)
		50%	60%	70%	80%	90%	100%
Levered Free Cash Flow Yield	8%	$94	$98	$102	$107	$111	$115
	9%	$84	$87	$91	$95	$99	$102
	10%	$75	$79	$82	$85	$89	$92
	11%	$68	$71	$74	$78	$81	$84
	12%	$63	$65	$68	$71	$74	$77
	13%	$58	$60	$63	$66	$68	$71
	14%	$54	$56	$59	$61	$63	$66
	15%	$50	$52	$55	$57	$59	$61
	16%	$47	$49	$51	$53	$55	$58
	17%	$44	$46	$48	$50	$52	$54
	18%	$42	$44	$46	$47	$49	$51
	19%	$40	$41	$43	$45	$47	$48
	20%	$38	$39	$41	$43	$44	$46

 

 

The employment agreements of Chuck Dawson, CEO of Harland Clarke, and Stephen Taub, CEO of Mafco, read like private equity employment agreements.  Dawson, who has been CEO since 2005 and has been with Clarke American since 1993, must achieve 90% of his EBITDA target to earn any bonus, and can earn up to 150% of his base salary ($850k) upon beating his target by 45%.  Taub, who has been CEO or COO of Mafco since 1993 (and with Mafco since 1973), receives 60% of his base salary ($965k) upon achieving 80% of EBITDA target, and up to 150% of base salary upon achieving 115%.  Based on an analysis of proxy statements, it appears that Mafco has hit at least 95% of its target EBITDA since 1995.

 

Beyond the Harland Clarke CEO, in January 2008, MFW will put in place a long-term incentive plan (LTIP) for the Harland Clarke senior management team.  While the details have not been disclosed, it will likely look like the Clarke American LTIP, which pays out 20% of the 3-year cumulative consolidated EBITDA achieved in excess of the target 3-year cumulative consolidated EBITDA.

 

 

1)  Ron Perelman screws minority shareholders.  MFW is controlled by Ron Perelman, whose poor reputation with minority investors/Wall Street is the single biggest overhang to the stock.  The source of this reputation is deserved, as in 2001, he used MFW to acquire a struggling investment (Panavision) at 4x its publicly-traded value.  This maneuver was later challenged in court, and Perelman was forced to unwind the transaction.  I find it unlikely that in a post-Enron, SOX era, Ron Perelman could try such a maneuver again.

2)  Check printing is a declining business.  Yes, check printing is a declining business, but approximately 30 billion checks are still written each year, and they will not go away overnight.  Harland Clarke constantly finds ways to increase their revenues/check with a number of different levers including (a) price increases, (b) reducing number of checks in each box, (c) making checks more appealing with licensed characters/sports teams, (d) upselling additional check accessories, and (e) including the financial institution’s marketing materials in the check box.

As far as a recessionary environment goes, Harland Clarke doesn’t care if a check is written for $1,000, $100, or $10.  As long as people spend money for goods and services, many will be using checks – and they’re certainly not concerned about the nickel cost of the check when they’re paying.

3)  There’s too much debt and MFW is subject to the whims of the credit market.  As mentioned above MFW can easily handle its debt load which has easy amortization terms (less than 4% per year).  Their senior facility and notes are not due until 2014 and 2015, respectively.  I think the credit market will recover by then.

 

4)  HFS revenues are tied to discretionary financial institution IT budgets and the mortgage markets.  The software that HFS sells is sold under long-term contracts and it is not discretionary.  Banks need HFS core processing systems to simply operate, and some recent surveys have shown that banks do not expect to cut back spending on core processing.  While their mortgage origination software might suffer from some lost clients, it only represented about 4% of that segment’s revenues, or less than 1% of all of Harland Clarke.

 

1)  Debt paydown:  Strong cash flows will result in the shift of value from debt to equity.

 

2)  Continued strong performance and improved “screens”:  Currently, MFW screens on CapitalIQ or Bloomberg at 11.2x LTM EBITDA as a result of the timing of the Harland acquisition.  Assuming Q4 and Q1 are similar to Q3 (which would exclude additional synergies), it would screen at 7.4x and start to pop up on more “value screens.”

 

3)  Harland Clarke becomes more open with investors:  Management at Harland Clarke has hinted that they may renew conference calls sometime in 2008, and they may even accept some analyst coverage.

 

4)  Perelman buys more shares or takes the whole business private.  He’s has been a buyer at higher prices, buying 200K shares at $60 last April and 196K shares last December around $51.  If the stock stays at such depressed levels, he may just take the remaining 60% private.