Arnold Van Den Berg owns 768,519 shares of Automatic Data Processing (ADP, Financial). While the stock is only his 13th largest position ā itās one of the highest quality companies in his portfolio. Thatās why Arnold Van Den Berg chose to write about ADP in his most recent letter to investors.
And thatās why Iāll be talking about it today.
Since Arnold Van Den Berg says ADP is a great company ā and I said Microsoft (MSFT, Financial) was a great company yesterday ā letās compare the two stocks. Arnold Van Den Berg owns both. Microsoft is his largest position. For more information, see my article on why Arnold Van Den Berg loves Microsoft.
ADP vs. Microsoft ā Vital Signs
ADPās Z-Score of 387.44 is not a typo. Remember: a Z-Score above 3 is considered safe.
ADP is a AAA rated company. One of only about half a dozen left. You could make the argument that ADP is the best business credit on planet Earth. I wonāt. Weāre talking about stocks. Not bonds. But ADP is a super safe debtor.
Iāll let Arnold Van Den Berg explain why:
āADPās annuityālike business is a great model. For starters, it can bundle many of its services. This has helped ADP maintain fairly consistent afterātax profit margins in the 12% to 15% range for more than 10 years. Furthermore, this company has a reasonably predictable recurring revenue stream, as well as a very āstickyā business. For example, ADP has a 90% client retention rate; and the average client relationship is more than 10 years old. This is a sign of a great company.ā
Retention rates are very important in terms of both profitability and safety. Usually companies with high retention rates ā low churn ā earn higher returns on capital. And they are almost always safer from a bond buyerās perspective. ADPās customers are super sticky. The companyās ultra-low free cash flow and return on capital variation coefficients ā 0.13 and 0.19 respectively ā support the idea that ADP is an incredibly reliable business.
ADP holds billions of dollars in client funds. This unique aspect of the business ā a lot like an insurance company ā excites Arnold Van Den Berg:
āADP also has the opportunity to earn money on what is referred to as āfloat.ā This is the money that ADP collects for payroll taxes from its clients each pay period. It then holds this money on behalf of its clients until it is time for them to send these payroll taxes into the governmentā¦until this money is due to the government, ADP basically escrows this money for its clientsā¦The investment income gained from this float currently provides approximately $600 million in pretax income to ADP. Eventually when interest rates go up, so will its profits on this float. This company will likely do well in an inflationary environment.ā
More on this later.
Right now, I want to temper Arnold Van Den Bergās hot talk about ADPās great business ā all of it true ā with some cold reality. The last 10 years havenāt been kind to ADP. Low interest rates and the mother of all recessions is a bad combination for a company that depends on low unemployment and high interest rates to juice its profits.
Letās compare ADPās free cash flow growth with another ā alleged ā high quality, no growth business: Microsoft.
ADP vs. Microsoft ā Real Free Cash Flow Per Share
Microsoftās average real free cash flow from 2000-2002 was $1.93 a share. From 2007-2009 it was $2.23. Microsoftās free cash flow grew by 2.1% ā over inflation ā during those 7 years.
ADPās average real free cash flow from 2000-2002 was $3.09 a share. From 2007-2009 it was $2.92. ADPās free cash flow shrank ā after inflation ā during those same 7 years.
Why?
Because the economy sucks.
The economy wasnāt great in 2000-2002. But it was a lot worse in 2007-2009. This influences the growth comparison more for an employment dependent company like ADP than for a tech company like Microsoft. Although, it should hurt Microsoft too. Yet the 2.1% a year growth in real free cash flow makes it clear the 2000s werenāt a lost decade for Microsoft. They were for ADP.
Iām going to deal with the interest rate issue in-depth a little later ā itās a bit math heavy. Always best to put the math at the end of the article. Right now, letās talk about ADPās possible future growth apart from rising interest rates.
Arnold Van Den Berg is upbeat about ADPās growth prospects:
āCurrently, with only 20% of sales coming from outside the U.S., ADP has plenty of room to grow for many years to comeā¦In our analysis, we are growing its earnings at the annualized rate of 9.7% over the next five years. Even during the past three years in a challenging economy with high unemployment, ADP grew its earnings at an annualized rate of 5.7%. We believe our growth rate is reasonable, especially when you factor in its rapidly expanding international business and the ability to bundle its various products and services. Under this scenario, we believe ADP has future earning power of $3.75.ā
I disagree.
Arnold Van Den Berg has an amazing track record. So, you should probably go with him. But I have a hard time coming up with reasonably conservative future growth numbers that get ADP past $3.50 a share in normal earning power. I definitely wouldnāt invest with the idea youāre going to see more than $3.50 a share in earnings from ADP anytime soon. It could happen. But itās not something you want to risk your retirement savings on. ADP needs to justify its stock price without assuming earnings above $3.50 a share. At least in my opinion.
Again, Arnold Van Den Berg doesnāt agree. He sees earnings of $3.75 a share in ADPās future.
Thatās the minor disagreement between Arnold Van Den Berg and me.
Hereās the major disagreement:
āIf we were to use the fiveāyear median Value Line P/E of 17.1 and add a 20% premium since ADP is much better than the average company, the P/E we would use is 20.5. If we multiply this P/E times the fiveāyear earnings potential of $3.75, it would suggest a price of $76.90 for a potential fiveāyear annualized return of 12.9% from todayās price. However, if you add the 3.2% dividend yield to this return, there is potential for a 16% total return over the next five yearsā¦Now, if we assume ADP gets back to its last fiveā year average P/E of 22.39, it suggests ADP could sell for approximately $84 per share. This means that from todayās price of $42, this stock has the potential for a fiveāyear annualized return of 15%. This does not include the 3.2% dividend yield that is paid out every year. If we include the dividend, the potential total annualized return would be 18.2% over the next five years.ā
Although Arnold Van Den Berg and I both consider ourselves followers of Ben Graham ā heās okay with assuming a P/E premium and Iām not. I just canāt apply a future P/E of 20 to a stock Iām buying. Maybe Iām old fashioned. Maybe Iāve got ā deservedly ā less faith in my prophetic powers than Arnold Van Den Berg.
But I just canāt assume a stock I buy will ever trade at a P/E of 20.
That difference in process explains the difference in outcome. Arnold Van Den Berg is willing to assume ADP will one day trade at a P/E of 20. Iām not. So Arnold Van Den Berg is willing to buy ADP stock. And Iām not.
Finally, I promised a discussion of ADPās float and the interest it earns on that float. For those readers brave enough to stick with meā¦
Hereās a comparison of ADPās inflation adjusted float and interest on that float. As always, Iām showing you per share numbers that use the current number of shares outstanding. Thatās all that matters to new investors. Iāve also included the āyield on floatā which is just interest divided by float.
Interest rates havenāt helped ADP. Theyāve hurt. But theyāve only hurt to the tune of about 11 cents per share. That only explains two-thirds of ADPās real free cash flow shrinkage. The rest is explained by high unemployment.
ADP has made competitive progress in the last 10 years. None of that shows up in the inflation adjusted numbers. Thatās because the U.S. economy is so bad right now that any progress ADP made is more than offset by where we are in the economic cycle.
Does that mean ADP will bounce back with the economy?
Lower unemployment and higher interest rates will both benefit ADP. As a result, the 12 times EBIT and 16 times free cash flow multiples you see for ADP may be overstating how expensive the stock is right now. Itās possible the 10 year numbers are not normal, because interest rates were low for much of the last 10 years.
Depending on what you consider a normal interest rate, itās possible to conclude ADPās normal earnings power is anywhere from 35 cents to 70 cents higher than what is was over the last 10 years. I think 35 cents of improvement due to higher interest rates is possible. 70 cents of improvement looks possible, when you think in terms of nominal rates. But inflation hasnāt been high these last 10 years. So thatās part of the explanation for low interest rates. An adjustment to normal real interest rates is likely to increase the yield on ADPās float by 1% rather than 2%.
Still, if we moved to a normal real interest rate ā as judged by averages going back 50, 100, or 150 years ā weād probably see ADPās yield on its float rise by 1 full percentage point. Earnings would rise by 35 cents per share. That interest rate increase alone would warrant a $5.25 rise in the stock price ($0.35 * 15 = $5.25).
Is that what Arnold Van Den Berg is looking for? Higher interest rates. Probably not. Higher interest rates alone will only do so much for ADP. But the combination of higher interest rates and lower unemployment ā basically a stronger economy ā will do wonders for ADP.
My best guess ā and this is only a guess ā is that ADPās normal earning power in the sense of what it would earn with ānormalā interest rates and ānormalā unemployment given its competitive position is $3.50 a share. That implies a normal stock price of $52.50 a share.
Thatās not a best case scenario. Iām really just assuming ADP would match its real free cash flow from the early 2000s and then add another 35 cents in additional interest from float.
Itās hard to imagine ADP stock trading below $52.50 in normal economic times. But Iām not an economist. Iām an investor. And 13% upside ā which is all the increase to $52.50 from $46 amounts to ā just isnāt enough.
Like Iāve said before, I need at least a 30% upside in any stock I buy. And I donāt want to have to assume the stock trades at a premium P/E. Iām always looking at something like net current assets, tangible, book value, EBIT, or ā most often of all ā 10-year average real free cash flow.
With ADP, we only get really impressive annual return numbers when we assume the stock will trade at more than 15 times free cash flow.
Thatās risky. And uncertain.
I like to find stocks that will outperform even while the company underperforms. ADPās stock will only outperform if the company outperforms. Or, if ADP regains its historical premium over the market wide P/E.
ADP is a great company. It deserves its AAA rating. And for someone looking for high-quality stocks like Arnold Van Den Berg this investment makes sense.
I just donāt see the upside.
The margin of safety here is all qualitative. Youāve got no quantitative protection. The stock is basically trading at fair value for an average company. The only thing protecting you is the idea ADP is an above average company. I agree with that assessment. But I donāt like investing based on my qualitative assessments.
I need some protection from the numbers. And ADP doesnāt offer any.
I wouldnāt recommend buying the stock at $46 a share.
There are cheaper stocks out there.
ADP shares have a 3.1% dividend yield. The company has a history of increasing the dividend while simultaneously decreasing the number of shares outstanding. Itās a wonderful company that does a wonderful job allocating capital. In all likelihood ā like Microsoft ā ADP stock will do as well or better than the S&P 500.
But it just isnāt cheap enough. Unless youāre limited to owning 20 plus big cap stocks ā and Iām not sure why any individual investor would be ā you shouldnāt buy ADP at $46 a share.
You should wait for a margin of safety.
And thatās why Iāll be talking about it today.
Since Arnold Van Den Berg says ADP is a great company ā and I said Microsoft (MSFT, Financial) was a great company yesterday ā letās compare the two stocks. Arnold Van Den Berg owns both. Microsoft is his largest position. For more information, see my article on why Arnold Van Den Berg loves Microsoft.
ADP vs. Microsoft ā Vital Signs
ADP | Microsoft | |
Z-Score | 387.44 | 14.87 |
F-Score | 6 | 7 |
FCF Margin | 15.65% | 37.23% |
Return on Capital | 9.67% | 53.29% |
FCF Margin Variation | 0.13 | 0.21 |
Return on Capital Variation | 0.19 | 0.22 |
Price / 10-Year Real FCF | 16.50 | 14.06 |
EV / 10-Year Real EBIT | 11.83 | 11.73 |
Price / NCAV | 41.16 | 15.55 |
Price / Tangible Book | 10.04 | 6.98 |
ADPās Z-Score of 387.44 is not a typo. Remember: a Z-Score above 3 is considered safe.
ADP is a AAA rated company. One of only about half a dozen left. You could make the argument that ADP is the best business credit on planet Earth. I wonāt. Weāre talking about stocks. Not bonds. But ADP is a super safe debtor.
Iāll let Arnold Van Den Berg explain why:
āADPās annuityālike business is a great model. For starters, it can bundle many of its services. This has helped ADP maintain fairly consistent afterātax profit margins in the 12% to 15% range for more than 10 years. Furthermore, this company has a reasonably predictable recurring revenue stream, as well as a very āstickyā business. For example, ADP has a 90% client retention rate; and the average client relationship is more than 10 years old. This is a sign of a great company.ā
Retention rates are very important in terms of both profitability and safety. Usually companies with high retention rates ā low churn ā earn higher returns on capital. And they are almost always safer from a bond buyerās perspective. ADPās customers are super sticky. The companyās ultra-low free cash flow and return on capital variation coefficients ā 0.13 and 0.19 respectively ā support the idea that ADP is an incredibly reliable business.
ADP holds billions of dollars in client funds. This unique aspect of the business ā a lot like an insurance company ā excites Arnold Van Den Berg:
āADP also has the opportunity to earn money on what is referred to as āfloat.ā This is the money that ADP collects for payroll taxes from its clients each pay period. It then holds this money on behalf of its clients until it is time for them to send these payroll taxes into the governmentā¦until this money is due to the government, ADP basically escrows this money for its clientsā¦The investment income gained from this float currently provides approximately $600 million in pretax income to ADP. Eventually when interest rates go up, so will its profits on this float. This company will likely do well in an inflationary environment.ā
Right now, I want to temper Arnold Van Den Bergās hot talk about ADPās great business ā all of it true ā with some cold reality. The last 10 years havenāt been kind to ADP. Low interest rates and the mother of all recessions is a bad combination for a company that depends on low unemployment and high interest rates to juice its profits.
Letās compare ADPās free cash flow growth with another ā alleged ā high quality, no growth business: Microsoft.
ADP vs. Microsoft ā Real Free Cash Flow Per Share
ADP | Microsoft | |
2009 | $2.95 | $2.58 |
2008 | $2.68 | $1.88 |
2007 | $3.13 | $2.23 |
2006 | $2.10 | $1.93 |
2005 | $3.12 | $1.65 |
2004 | $2.59 | $2.09 |
2003 | $2.46 | $1.84 |
2002 | $3.11 | $2.08 |
2001 | $3.13 | $1.94 |
2000 | $3.03 | $1.78 |
Microsoftās average real free cash flow from 2000-2002 was $1.93 a share. From 2007-2009 it was $2.23. Microsoftās free cash flow grew by 2.1% ā over inflation ā during those 7 years.
ADPās average real free cash flow from 2000-2002 was $3.09 a share. From 2007-2009 it was $2.92. ADPās free cash flow shrank ā after inflation ā during those same 7 years.
Why?
Because the economy sucks.
The economy wasnāt great in 2000-2002. But it was a lot worse in 2007-2009. This influences the growth comparison more for an employment dependent company like ADP than for a tech company like Microsoft. Although, it should hurt Microsoft too. Yet the 2.1% a year growth in real free cash flow makes it clear the 2000s werenāt a lost decade for Microsoft. They were for ADP.
Iām going to deal with the interest rate issue in-depth a little later ā itās a bit math heavy. Always best to put the math at the end of the article. Right now, letās talk about ADPās possible future growth apart from rising interest rates.
Arnold Van Den Berg is upbeat about ADPās growth prospects:
āCurrently, with only 20% of sales coming from outside the U.S., ADP has plenty of room to grow for many years to comeā¦In our analysis, we are growing its earnings at the annualized rate of 9.7% over the next five years. Even during the past three years in a challenging economy with high unemployment, ADP grew its earnings at an annualized rate of 5.7%. We believe our growth rate is reasonable, especially when you factor in its rapidly expanding international business and the ability to bundle its various products and services. Under this scenario, we believe ADP has future earning power of $3.75.ā
I disagree.
Arnold Van Den Berg has an amazing track record. So, you should probably go with him. But I have a hard time coming up with reasonably conservative future growth numbers that get ADP past $3.50 a share in normal earning power. I definitely wouldnāt invest with the idea youāre going to see more than $3.50 a share in earnings from ADP anytime soon. It could happen. But itās not something you want to risk your retirement savings on. ADP needs to justify its stock price without assuming earnings above $3.50 a share. At least in my opinion.
Again, Arnold Van Den Berg doesnāt agree. He sees earnings of $3.75 a share in ADPās future.
Thatās the minor disagreement between Arnold Van Den Berg and me.
Hereās the major disagreement:
āIf we were to use the fiveāyear median Value Line P/E of 17.1 and add a 20% premium since ADP is much better than the average company, the P/E we would use is 20.5. If we multiply this P/E times the fiveāyear earnings potential of $3.75, it would suggest a price of $76.90 for a potential fiveāyear annualized return of 12.9% from todayās price. However, if you add the 3.2% dividend yield to this return, there is potential for a 16% total return over the next five yearsā¦Now, if we assume ADP gets back to its last fiveā year average P/E of 22.39, it suggests ADP could sell for approximately $84 per share. This means that from todayās price of $42, this stock has the potential for a fiveāyear annualized return of 15%. This does not include the 3.2% dividend yield that is paid out every year. If we include the dividend, the potential total annualized return would be 18.2% over the next five years.ā
Although Arnold Van Den Berg and I both consider ourselves followers of Ben Graham ā heās okay with assuming a P/E premium and Iām not. I just canāt apply a future P/E of 20 to a stock Iām buying. Maybe Iām old fashioned. Maybe Iāve got ā deservedly ā less faith in my prophetic powers than Arnold Van Den Berg.
But I just canāt assume a stock I buy will ever trade at a P/E of 20.
That difference in process explains the difference in outcome. Arnold Van Den Berg is willing to assume ADP will one day trade at a P/E of 20. Iām not. So Arnold Van Den Berg is willing to buy ADP stock. And Iām not.
Finally, I promised a discussion of ADPās float and the interest it earns on that float. For those readers brave enough to stick with meā¦
Hereās a comparison of ADPās inflation adjusted float and interest on that float. As always, Iām showing you per share numbers that use the current number of shares outstanding. Thatās all that matters to new investors. Iāve also included the āyield on floatā which is just interest divided by float.
Real Interest Per Share | Real Float Per Share | Yield on Float | ||
2009 | $1.18 | $38.28 | 3.08% | |
2008 | $1.29 | $33.77 | 3.83% | |
2007 | $1.31 | $32.39 | 4.04% | |
2006 | $1.45 | $39.95 | 3.63% | |
2005 | $1.21 | $39.01 | 3.11% | |
2004 | $1.13 | $41.26 | 2.74% | |
2003 | $0.88 | $30.59 | 2.88% | |
2002 | $0.92 | $28.60 | 3.20% | |
2001 | $1.06 | $27.63 | 3.82% | |
2000 | $1.36 | $28.54 | 4.77% |
Interest rates havenāt helped ADP. Theyāve hurt. But theyāve only hurt to the tune of about 11 cents per share. That only explains two-thirds of ADPās real free cash flow shrinkage. The rest is explained by high unemployment.
ADP has made competitive progress in the last 10 years. None of that shows up in the inflation adjusted numbers. Thatās because the U.S. economy is so bad right now that any progress ADP made is more than offset by where we are in the economic cycle.
Does that mean ADP will bounce back with the economy?
Lower unemployment and higher interest rates will both benefit ADP. As a result, the 12 times EBIT and 16 times free cash flow multiples you see for ADP may be overstating how expensive the stock is right now. Itās possible the 10 year numbers are not normal, because interest rates were low for much of the last 10 years.
Depending on what you consider a normal interest rate, itās possible to conclude ADPās normal earnings power is anywhere from 35 cents to 70 cents higher than what is was over the last 10 years. I think 35 cents of improvement due to higher interest rates is possible. 70 cents of improvement looks possible, when you think in terms of nominal rates. But inflation hasnāt been high these last 10 years. So thatās part of the explanation for low interest rates. An adjustment to normal real interest rates is likely to increase the yield on ADPās float by 1% rather than 2%.
Still, if we moved to a normal real interest rate ā as judged by averages going back 50, 100, or 150 years ā weād probably see ADPās yield on its float rise by 1 full percentage point. Earnings would rise by 35 cents per share. That interest rate increase alone would warrant a $5.25 rise in the stock price ($0.35 * 15 = $5.25).
Is that what Arnold Van Den Berg is looking for? Higher interest rates. Probably not. Higher interest rates alone will only do so much for ADP. But the combination of higher interest rates and lower unemployment ā basically a stronger economy ā will do wonders for ADP.
My best guess ā and this is only a guess ā is that ADPās normal earning power in the sense of what it would earn with ānormalā interest rates and ānormalā unemployment given its competitive position is $3.50 a share. That implies a normal stock price of $52.50 a share.
Thatās not a best case scenario. Iām really just assuming ADP would match its real free cash flow from the early 2000s and then add another 35 cents in additional interest from float.
Itās hard to imagine ADP stock trading below $52.50 in normal economic times. But Iām not an economist. Iām an investor. And 13% upside ā which is all the increase to $52.50 from $46 amounts to ā just isnāt enough.
Like Iāve said before, I need at least a 30% upside in any stock I buy. And I donāt want to have to assume the stock trades at a premium P/E. Iām always looking at something like net current assets, tangible, book value, EBIT, or ā most often of all ā 10-year average real free cash flow.
With ADP, we only get really impressive annual return numbers when we assume the stock will trade at more than 15 times free cash flow.
Thatās risky. And uncertain.
I like to find stocks that will outperform even while the company underperforms. ADPās stock will only outperform if the company outperforms. Or, if ADP regains its historical premium over the market wide P/E.
ADP is a great company. It deserves its AAA rating. And for someone looking for high-quality stocks like Arnold Van Den Berg this investment makes sense.
I just donāt see the upside.
The margin of safety here is all qualitative. Youāve got no quantitative protection. The stock is basically trading at fair value for an average company. The only thing protecting you is the idea ADP is an above average company. I agree with that assessment. But I donāt like investing based on my qualitative assessments.
I need some protection from the numbers. And ADP doesnāt offer any.
I wouldnāt recommend buying the stock at $46 a share.
There are cheaper stocks out there.
ADP shares have a 3.1% dividend yield. The company has a history of increasing the dividend while simultaneously decreasing the number of shares outstanding. Itās a wonderful company that does a wonderful job allocating capital. In all likelihood ā like Microsoft ā ADP stock will do as well or better than the S&P 500.
But it just isnāt cheap enough. Unless youāre limited to owning 20 plus big cap stocks ā and Iām not sure why any individual investor would be ā you shouldnāt buy ADP at $46 a share.
You should wait for a margin of safety.