Investment Is A Journey

This is the post excerpt.

Investment is a journey.  Sometimes it is better to leave a track record of what you think when you made the investment and come back later to learn from my own journey.  If it is a mistake, check what I have missed.  If it is a success, check why I haven’t made a bigger investment (I remember that Charles Akre once said that the biggest mistake he has made is that he didn’t make a winner a bigger position).



How to Screen Moat Stocks

Recently, I finished reading the book “Quantitative Value” by Wesley Gray.  I think the book is well written with a lot of quotes from famous value investors.  It is interesting that he provides the quantitative way to screen moat stocks, which includes

  • CFOA = SUM(8-Yr FCF) / Total Assets (FCF = Net Income + D&A – WC Change – Capex)
  • 8-Yr geometric ROA
  • 8-Yr geometric ROC (ROC = EBIT/(Net PP&E + Net WC))

The reason to use 8 years is that you want to include the whole business cycle.  The reason to use geometric is to exclude very cyclical businesses that could have extremely high return in a few years but negative return in down years.  In the end, you can rank each screen and assign the value to it.  The sum of three screens will provide the best moat stocks.

The most powerful screen that the author found out is actually just EBIT/EV from magic formula, which beats all other factors.  The book was published in Dec, 2012.  The author actually launched a quantitative fund (ticker: QVAL) based on his research.  QVAL was launched in Nov 2014.  Unfortunately, the performance of QVAL was not good, with annualized return of -0.2% vs. S&P’s 8.8% over the same period.

Despite that, I think these 3 quantitative screens are still useful tools to screen moat stocks.


Is OneMain a Buy from here?

Below is something I have written in Dec 2016 when the stock was around $20-23.  With 4Q result out, the stock appreciated to $28.3/sh.  Recently it pulled back to $26.5/sh.  Is it still a buy from here?  I think so as not much has changed from my write-up.

OneMain Financial


OneMain Financial (OMF $20.5 on 11/23/2016) represents a compelling opportunity to invest in a nationwide branch-based near-prime consumer lending franchise.  Due to slower growth and higher provision caused by acquisition integration, OneMain stock currently trades at 5.3x P/E and 0.87x P/B with ROE of 18%.  Stock is now at its IPO price in 2013 with $5.5bn receivables and 10x P/E vs. current $13.6bn receivables and 5x P/E.  In a stress scenario, NCO could rise 50% as it did in GFC.  In this case, EPS will drop from $4.06/sh to $2.50/sh in 2017.  Stock would trade at 8.2x of this trough earnings.  It is easy to foresee 50-100% return in the next two years.


OneMain is the leading consumer finance company in the sub to near-prime space.  It was previously known as American General Finance (part of AIG).  Fortress Investment Group purchased the company in August 2010.  It went IPO at $17.00 per share in Oct 2013.  Later it merged with OneMain subsequent to its spinoff out of Citigroup in March 2015.  Fortress is still a major shareholder with 54.5% stake.  The company changed its name to OneMain after the merger.


OneMain originates personal loans that are fully amortizing, fixed rate, non-revolving, secured by titled personal property, consumer goods, or other personal belongings.  OneMain currently holds a $13.5bn portfolio of direct, unsecured or hard secured loans.  There is a $210m legacy real estate portfolio in runoff mode, which we will not focus in this write-up.  Last but not least, OneMain cross-sells credit, life insurance, disability insurance, involuntary unemployment insurance and property and casualty insurance together with its loans.

Img 0.png


OneMain has a distinct competitive advantage in the near prime space against both banks and other payday / title lenders.  Loans have an average yield of 24-25%, which is lower than competitors Regional Management (RM) and World Acceptance Corp (WRLD) as can been seen above but is clearly above bank lending rates.  Generally payday / title lenders will lend small amounts at 50%+ rates, to FICO borrowers 660 FICO at 10%-20% rates for longer terms.  OneMain falls in the middle of these two segments lending at rates between 12%-36%, at FICO scores <700 for $3,000-$50,000 and up to 5 years.  Due to Basel III, banks cannot compete in this space. Under the A-IRB approach, a subprime loan with a 8% probability of default receives a risk-weight of 161%, or roughly $20 in equity for each $100 in subprime loans given current regulatory capital requirements north of 12%.

Img 1.png


OneMain acquisition is transformational for the company by increasing loan portfolio by 3.5 times.  After the merger, OneMain became the leading subprime/near-prime consumer lender with national footprint.  OneMain has total 2.2m customers, with total 10m customers with whom OneMain has relationship.  Total number of subprime/non-prime consumers is 75-100m in the country, so the potential market is quite large.  Also can be seen from above, OneMain and Springleaf customers are very similar.

Img 2

OneMain and Springleaf share a lot in common in terms of age, income, financial health, etc.  OneMain is tilted more to higher FICO score.  As such, loans per branch are higher at old OneMain branches ($7m vs. $5m) as higher FICO score typically supports larger loan size.  The other differences are Springleaf is growing much faster than old OneMain (15-20% vs. 5%) due to several new initiatives and has much lower NCO due to high percentage collateralized loans (55% vs. 17%).  However, old OneMain does have better expense ratio (9% vs. 13% of receivables).  As such, the rationale of merger is quite logic by combining the best practices of both companies.

Img 3

OneMain loans are fully amortizing with equal monthly payments (terms range from 36 – 60 months).  There are two main categories: auto loans with average size of $13,400 and APR 20% and personal loan with average size of $6,500 and APR 27%.  The main usage of loans are for debt consolidation and home/auto bill payments.

Lending standard is based on NDI (Net Disposable Income).  OneMain will verify income/residency/other info and deduct rent, living expenses, all existing debt servicing costs and new OneMain loan costs from NDI while still making sure that customers have some income cushion left.  Unlike online lending which typically focuses on prime customers, underwriting in this segment is more risky and OneMain has the advantage of being in the local community and familiar with each customer’s individual situation.

Unlike other consumer lending, OneMain’s NPL is based on 60-day delinquency (vs. typical 30 days in the banking industry) as customers in this segment do have some earning volatility (such as overtime pay).  NPLs are charged off after 180 days (same as the banking industry).  Loan loss reserve setup is based on forward 9 months coverage.  Collection is initially handled by the branches as branch people know customers personally.  After 60-90 days, the collection is handled by the central collection offices (not outsourced).

Img 4.png

Historical NCO can be seen from above.  Springleaf (old OneMain) has a better track record due to higher percentage in collaterized loan underwriting.  Over the last 20 years, Springleaf’s average NCO is 5.5%.  During the GFC, we could see that NCO doubled from historical low point in 2006, which actually shows lower volatility than prime credit card lending that NCO tripled during the same period.

Img 5.png

Funding is done through a mixed of ABS structure notes (58% of total) and unsecured debt (42%).  Total cost of debt is 5.5%.  OneMain is more liability sensitive.  Corporate debt is fixed rate but ABS is based on LIBOR plus some spread.  As interest rate rises, the fund cost will also increase (mainly from LIBOR).  However, OneMain does have strong liquidity of $8-9bn through a mix of cash on B/S, undrawn conduit and unencumbered consumer loans.

Last but not least, OneMain also cross-sells credit, life insurance, disability insurance, property and casualty insurance and involuntary unemployment insurance together with its loans. These are underwritten by 4 insurance subsidiaries (2 P&C and 2 life, 2 belong to Springleaf and 2 belong to OneMain).  Insurance penetration is ~50% and most of them are single premium wrapped with loans.  There is very low risk of PPI issues that plagued British banks as these policies are expressively explained and customers could opt out after 30 days without any penalties.  Insurance premiums account for ~20% of revenue.

What happened to the stock

Img 6.png
CEO Jay Levine is the president and CEO of OneMain and he came from Springleaf side.  Previously Jay was the president and CEO of RBS Securities.  After GFC, he was the president and CEO of Bluestem Group, an online retailer of a broad selection of name-brand and private label general merchandise serving low- to middle-income consumers in the U.S.  After Fortress took a major stake in Springleaf, Jay was instituted to run the company and later to run the combined Springleaf and OneMain.Springleaf went IPO at $17/sh in Oct 2013 and the stock quickly rose to +$20/sh afterwards.  In March 2015, Springleaf announced acquisition of OneMain and the stock quickly rose to $45-50/sh as management guided strong core EPS of $6.20-6.70/sh in 2017 (see below).

Img 7.png

In Feb 2016, the whole market was sold off, especially financial sector.  OneMain didn’t escape the bruise.  Moreover, investors started to worry about its coming $1.9bn debt due in 2017 (see above) and fear there could be equity issuance coming.

Img 8.png

To address investors’ concerns, management sold SpringCastle (a JV between Springleaf and Fortress) early and cut core EPS to $5.60-6.10 in 2017.  However, in 3Q16, management has to cut core EPS guidance again from $5.60-6.10 to $3.75-4.00.  Investors gave up and the stock was sold off.  The main reasons for EPS cut were due to integrations that slowed the growth assumed and caused more NPL (OneMain branches spent more time on system integration and less on new business and collection effort).


Jay owns 3.1m shares of OMF, which puts his stake at $60m at current share price.  After stock’s nosedive after 3Q result, he did purchase 50k stocks or $1m additional stocks.  If the stock could just go back to $30/sh prior to 3Q result, Jay will add $30m wealth and investors will make 50% return.  It is fair to assume that Jay has strong incentive to make integration work and grow the company from there.

Field Research

Img 9.png

I went to the nearest OneMain branch.  This is the old OneMain branch and there are typically 4 loan officers with one branch manager.  Most customers I saw do match the customer profile that the company provides (50 years old).  It does seem most officers know customers and customers know each other (local community root).  Loan officers are compensated by fixed salaries and bonus are tied to loan production.  Branch manager’s bonus is more tied to loan production and branch profitability.  Higher up would be regional managers whose bonuses are even more tilted to these two metrics.   Branch managers confirmed that both companies share the same culture and underwriting practices.  The systems are very similar to each other except that Springleaf system uses net income (thus better) vs. gross income used by old OneMain.  Branch managers also confirmed that business is a bit of slow due to integration and they are going to complete integration in 1Q17.  Overall, the field trip does validate a lot of points that management made on the conference call.

Model & Valuation

The stock currently trades at 5.3x P/E and 0.87x P/B with ROE of 18%.  Stock is very cheap considering that it went IPO at $20/sh with $5.5bn receivables and 10x P/E.  Three years later, we have the only nationwide non-prime consumer lending company trading at the same price but with $13.6bn receivables and 5x P/E.

Management guided $275-300m cost savings with $100m already achieved in 2016.  2017 will see another $100m cost taking out of the system.  Operating expenses % receivable will drop from 13% to 8% by 2018.  Receivable growth is guided to 5-10%, which seems more achievable than previous +15% guidance.  Meanwhile, normalized NCO seems to be 6.75% (see below) due to 45% of secured lending penetration.

Img 10.png

Long-term guidance is 10% unlevered return with +20% ROE and 5-7x leverage.  Return on receivable is expected to be 4.5-5.0%.  With the mid-point of guidance, we have $5.10/sh on 2017 estimates.

Img 11

Looking at chart above again, Springleaf’s 20-year average NCO is 5.5% and it went to 8.0% during GFC, a 50% jump.  Applying same math to the normalized NCO of 6.75%, we have a relatively stress case of NCO at 10%.  By applying 10% provision to my 2017 estimate, EPS will drop from $4.06/sh to $2.50/sh.  With stock at $20.50, the stock trades at 8.2x trough earnings.

Img 12.png


Regulation: The federal Consumer Financial Protection Bureau (CFPB) unveiled proposed rules in June that take aim at short-term payday loans charging triple-digit annual percentage rates.  These rules would also cover installment loans with larger prepayment periods that charge rates higher than 36%.  With a hard cap on interest rates at 36% and an average yield of receivables of 24-25%, this puts OneMain in a favorable position from a regulatory standpoint.  CFPB also proposed new rules regarding loan collection practices, especially for those loans outsourced to third parties.  OneMain is in fully compliance by preforming collection in-house.  Moreover, OneMain hires former regulators from the CFPB to ensure best-in class regulatory compliance.

Complicated Reporting: company’s financial reporting is messier than others with non-core/run off portfolio and GAAP earnings which are impacted to purchase accounting and merger one-time costs.  Fortunately, the difference between GAAP EPS and adj. EPS will greatly shrink after 2017.

Leverage: Springleaf acquired OneMain which is 3.5x larger with only 20% equity issuance.  A/E leverage seems fine at 6x but equity includes large goodwill ($1.44bn) and intangibles ($0.56bn).  Adj. debt/Adj. Tangible was 18.5x after acquisition.  Due to the sale of SpringCastle and high cash generation of the business, this leverage is now down to 10.7x in 3Q16 and management guides 7x in 2018.  Due to this reason, the company can’t pay dividend or initiate stock buyback until it reaches its target leverage point until 2018.

Funding: OneMain is now an established issuer in the securitization market, issuing at rates that are competitive with its peers.  However, we must recognize that the company relies on wholesale funding (the securitization and debt markets) to fund the business.  Moreover, OneMain is more liability sensitive so rising interest rate is a minor negative (mainly through LIBOR).

Fortress Sells Down Stake: Fortress still owns 54% of OneMain and there is no lockup for this stake.  However, the fund that owns the stake has a contractual expiration in Feb 2018.  As such, it is foreseeable that Fortress will make sure the management improving the operation to have a higher stock price before the fund is closed down.