YumChina (NYSE:YUMC)

[Snoopy]

There is an old phrase in computation. 'GIGO' which means 'Garbage In, Garbage Out'. The practical translation of this glib phrase is that, no matter how sophisticated your mathematical model, if you put unreliable numbers in a sophisticated front end, then the output will be a sophisticated but unreliable result.

Return on Equity

In the 'Buffett Growth Model', it is 'Return on Equity' that is the most important factor in determining earnings for the year. I am happy with assuming a return on equity for Yum China of 18.8%. In absolute terms, this will be a high number to roll over on itself for the next ten years. Yet the actual ROE over the last three years (the time since YUMC has been listed as a separate entity) have been noticeably higher than this. I am not expecting the high ROE figures from the last three years to continue. Profits have been growing a lot faster than sales. And I expect some re-balancing of costs upwards. Indeed, over FY2018, the 'Net Profit Margin' was, apparently, already shrinking.

Dividend Payments

YUMC was demerged without the promise of paying dividends. Yet by the last quarter of FY2017 the first of what have become quarterly dividends (10cps in this case) was declared and paid. Total dividends declared over FY2018 amounted to 42cps. The share price on 31st March 2018 (the tax year boundary in NZ) was $41.50. But this price was inflated as the result of a now withdrawn takeover offer. So I am going to use an indicative normalised price of 0.85 x $41.50 = $35.28. Based on that price, the dividend yield for FY2018 looking forwards was:

$0.42 / $35.28 = 1.19%

This is not great. But it is close to covering the FIF tax liability of 1.5%, which is a good thing for shareholders. An overseas shareholding that does not cover the FIF tax liability via dividend yield means a negative cashflow liability for New Zealand holders indefinitely into the future, just to maintain the existing FIF holding you have.

The ability to pay a dividend for a company which operates in the Chinese market is restricted. I quote from Item 5 (p54 AR2017).

"The laws rules and regulations applicable to our Chinese subsidiaries permit payments of dividends only out of their accumulated profits, if any, determined in accordance with applicable Chinese accounting standards and regulations. Under Chinese law, an enterprise incorporated in China is required to set aside at least 10% of its after tax profits each year, after making up the previous year's accumulated losses, if any, to fund certain statutory reserve funds, until the aggregate of such a fund reaches 50% of its reserve capital." As a result, our Chinese subsidiaries are restricted in their ability to transfer their assets to us in the form of dividends,"

Since YUMC is now paying dividends, it follows that at least a proportion of these Chinese subsidiary companies have accumulated an aggregate fund reaching 50% of its reserve capital. But did that happen only at the start of Q4 FY2017, when the first YUMC dividend was paid? Or did it happen before that? And what is 'Reserve Capital' anyway?

From " http://www.businessdictionary.com/de...e-capital.html "

"Reserve Capital" is the authorized capital of a firm that has not been called up and is, therefore, available for drawing in case of a need.

So it looks to me as though 'reserve capital' is probably a moving target. This would make sense as the bigger the company gets, the more capital/ retained earnings are needed to keep it running. At EOFY2017/SOFY2018, the capital position of YUMC, just after the first dividend was paid, was like this:

	Total Equity	50% of Total Equity
SOFY2017	$2,443m	$1.222m
SOFY2018	$2,859m	$1,430m
SOFY2019	$2,976m	$1,488m

Note that for comparative purposes, I have included the one year either side figures as well. Can we then say that $2,443m was insufficient capital to pay dividends, while $2,859m was enough? Or is this whole dividend thing an 'operating free cash flow' question? I pose that last question because it is 'cash flow' that eventually builds 'shareholder equity'.

	FY2017	FY2018
Net Cashflow from Operating Activities	$884m	$1,333m
less Stay in business Capital Expenditure (1)	$415m	$470m
add Net Interest Gains	$25m	$36m
equals Operating Free Cashflow	$444m	$899m
Compare with
Dividends Paid during Year	$60m	$197m
add Repurchase of Shares over Year	$128m	$307m
equals Total Discretionary Payments to Shareholders over Year	$188m	$504m

(1) I have equated 'Stay in Business Capital Expenditure' with 'All Capital Expenditure'. Capital expenditure each year is substantial in total. But this is directed towards a large number of small sites, in general. Most of these new stores will last for ten years before any lease renegotiations. But payback, in the case of KFC at least, can be as short as two years. Capital expended in constructing new restaurants is expected to continue at current levels or accelerate as part of the YUMC business plan going forwards. So I think it is fair to consider the capital used in developing new additional restaurants as 'stay in business capital.'

My interpretation of the above two tables goes like this:

1/ If $2,859m (balance SOFY2018) is a sufficient capital base from which it is legal to consider paying dividends AND
2/ the operating free cashflow over the FY2018 year generated an additional $899m of discretionary capital AND
3/ $504m in discretionary payments were made to shareholders over the FY2018 year THEN
4/ There remains $395m (=$899m-$504m) of new capital that could have been paid out to shareholders on the books at YUMC

Why is this important? I have shown in the Buffett model an increasing stream of dividend payments over the next ten years. And the above calculation shows that even if earnings per share do not rise in the future, then total discretionary payments to shareholders could rise by $395m/$197m = +78% right now. This shows that my projected dividend increases, at least in the medium term, are doable.

New Shares Issued

The Buffett modelling shows a 'net increase' in YUMC shares over ten years, up from 392m to 400.178m. This is an 'net increase' of 8.178m. The 'net increase' is the difference between new shares issued as part of long term employment bonuses ( 10x 3.542m = 35.420m ) and those bought back on market. By simple subtraction the number of shares modelled to be bought back on market is therefore:

400.178m - (392m + 35.420m) = -27.242m

(Note: this result is a negative number, because buybacks remove shares from the total).

Over FY2017, 2.254m new share options were granted an average exercisable price of $26.56. With the current share price now 60% higher than that, it is likely that future share options issued will be lower in number to equate to a similar dollar value of compensation. Even so, if the number of shares issued over the next ten years was comparable in quantity to those issued during FY2017, then that would only equate to:

2.254m x 10 = 22.540m

of new shares being issued. This is rather less than the number of shares I have modelled being bought back. Yet we also need to consider that the balance outstanding of previously awarded share options was 21.595m at EOFY2017 (the latest figure available). And I am modelling 2.254m options to be added to the current 21.595m pre-existing total this year, and every future year. If this grand total gradually reduces over time (IOW relatively more options are exercised than are issued), that means my apparently excessive modelling of buybacks may indeed only just balance the options exercised.

I feel, on balance, it is likely that I have overestimated the number of new shares coming on stream in the next ten years as a result of employee remuneration packages. This is because with new options being listed at a much higher excise price, and the YUMC share price perhaps getting a little ahead of itself, there is a chance more newer options will expire unexercised. Fewer options exercised means less capital available to the company for future expansion. But it also means a lower level of on market buybacks will be required to neutralise the earnings dilutive impact of these new shares. And that it turn means more cashflow available for dividends.

Starting Share Price

The modelled share price on 1st January 2019 was $33.43 - the actual opening price on the day. The implied share price, based on, starting shareholder equity, projected earnings using an 18.8% averaged ROE rate, and an averaged 23.2 PE ratio is:

23.2 x [0.188 x $7.62] = $33.24

The fact that these two figures are similar is pleasing, but largely a co-incidence. Guessing with precision the PE ratio the market will assign to a share at any one particular time is a game of luck. Besides, the annual estimate of the share price in the calculation table is only there to estimate the FIF regime tax bill for the year. It does not affect the compounding calculation as regards future earnings and dividends as modelled.

Similarly, my modelled share price for starting the years 2020 and 2021 are lower than today's share price, even though the growth outlook for the company is good. However this is because the model is a mathematical construct based on averaged assumptions. For this model, the averaged assumptions are likely to get better the more years roll by. But to single out any particular year and declare that figure as a 'prediction' is something that is technically beyond the scope of this 'Buffett Growth Model'.

'ROE normalised' -as derived- applied to unnormalised unadjusted capital

On the surface, what I have done is derive an ROE figure from my adjusted profit data, then applied this figure to model the growth of 'something else'. This does not sound good. My adjustments are most distorting to the published result figures (or put another way the published result figures are most distorting to realistic operating results) over FY2018. But the way the 'Buffett growth model' is constructed, such a distortion will compound through ten years of future earnings data. So the overall effect may not be trivial!

The 'Wuxi KFC equity revaluation', which I have ignored for profit purposes, has resulted in a 'realistically realisable capital gain' of $98m coming onto the books. This is incorporated in the company's $2,873m of 'shareholder equity on hand' at the end of the FY2018 financial year. Consequently, when I apply an averaged normalised ROE figure to this 'inflated capital', in order to estimate FY2019 earnings, I am applying that ROE figure to more capital than would otherwise be on the books through normal operating trading. Is this a problem? Particularly when we realise that there is more capital on the books than there would be as a result of normal operating trading. The answer to that question depends on how 'capital constrained' YUMC is. IOW, if YUMC suddenly had a windfall of capital (which they did), could this capital be redeployed to open new restaurants at a faster rate than would otherwise be possible? I believe the answer to that question is 'yes'. So, in this instance, I am convinced that applying normal operating earnings rate to a larger than expected capital pile will not result in an overestimate of YUMC earnings going into the future.

The $98m of 'bonus capital' that I am talking about here represents:

$98m/$2,873m = 3.4% of the total company capital.

So what I am saying here is that an incremental restaurant build rate of 3.4% every year for ten years is sustainable. I am very comfortable in saying this.

SNOOPY

[Snoopy]

Return on Equity

In the 'Buffett Growth Model', it is 'Return on Equity' that is the most important factor in determining earnings for the year. I am happy with assuming a return on equity for Yum China of 18.8% (edit now updated to 18.5%) . In absolute terms, this will be a high number to roll over on itself for the next ten years. Yet the actual ROE over the last three years (the time since YUMC has been listed as a separate entity) have been noticeably higher than this. I am not expecting the high ROE figures from the last three years to continue. Profits have been growing a lot faster than sales. And I expect some re-balancing of costs upwards. Indeed, over FY2018, the 'Net Profit Margin' was, apparently, already shrinking.

I have mentioned before that YUMC came into being as a 'stock split' for the parent YUM corporation. This is a slight simplification of the truth. In fact at the time of the split, YUMC received an outside capital injection of $460m (AR2018 p115). This is as the result of two strategic investors being brought on board the share register:

1/ An affiliate of the 'Primavera Financial Group' called 'Pollis Investment L.P.' invested $410m.
2/ An affiliate of ''Zhejiang Ant Small and Micro Financial Services Co. Limited" called 'API Hong Kong Investment Limited" invested $50m.

The net effect of these transactions was to add 19.145m shares (along with the $460m) to the 363.758m shares that came into existence at the time of separation. These shares were added in the very last quarter of 2016 and so already appear on the FY2016 balance sheet information as presented in the table below.

	EOFY2016	Change	EOFY2018
Normalised Earnings {A}	$472m		$634m
No. of Shares {B}	383m		392m
eps {A}/{B}	$1.23c	+39c {D}	$1.62
Owner Equity {C}	$2,443m		$2,976m
Owner Equity per share {C}/{B}	$6.38	+$1.21 {E}	$7.59
Return on Incremental Equity / Share {D}/{E}		+32%

It is likely that the net effect of this earlier $460m investment was not felt immediately. So much of the profitability gain apparent from subsequent net capital injection into YUMC (mainly from senior employees cashing in their stock options) is 'piggy backing' on the earlier $460m cash injection not shown in the above table. Thus in my opinion a more meaningful comparison table is this second one:

	31/10/2016 (spin off date)	Change	EOFY2018
Normalised Earnings {A}	$472m (for all of FY2016)		$634m
No. of Shares {B}	364m		392m
eps {A}/{B}	$1.30c	+32c {D}	$1.62
Owner Equity {C}	($2,443m-$460m)		$2,976m
Owner Equity per share {C}/{B}	$5.45	+$2.14 {E}	$7.59
Return on Incremental Equity / Share {D}/{E}		+15%

Note that 15% is well below the overall ROE figure of 21.3% achieved over FY2018 and also below the ROE figure of 18.5% over the last five years. 15% return on 'incremental equity' is nevertheless a good figure, the kind of figure that a Warren Buffett would be happy with.

SNOOPY

[Snoopy]

I have mentioned before that YUMC came into being as a 'stock split' for the parent YUM corporation. This is a slight simplification of the truth. In fact at the time of the split, YUMC received an outside capital injection of $460m (AR2018 p115). This is as the result of two strategic investors being brought on board the share register:

1/ An affiliate of the 'Primavera Financial Group' called 'Pollis Investment L.P.' invested $410m.
2/ An affiliate of ''Zhejiang Ant Small and Micro Financial Services Co. Limited" called 'API Hong Kong Investment Limited" invested $50m.

The net effect of these transactions was to add 19.145m shares (along with the $460m) to the 363.758m shares that came into existence at the time of separation. These shares were added in the very last quarter of 2016 and so already appear on the FY2016 balance sheet information as presented in the table below.

	EOFY2016	Change	EOFY2018
Normalised Earnings {A}	$472m		$634m
No. of Shares {B}	383m		392m
eps {A}/{B}	$1.23c	+39c {D}	$1.62
Owner Equity {C}	$2,443m		$2,976m
Owner Equity per share {C}/{B}	$6.38	+$1.21 {E}	$7.59
Return on Incremental Equity / Share {D}/{E}		+32%

It is likely that the net effect of this earlier $460m investment was not felt immediately. So much of the profitability gain apparent from subsequent net capital injection into YUMC (mainly from senior employees cashing in their stock options) is 'piggy backing' on the earlier $460m cash injection not shown in the above table. Thus in my opinion a more meaningful comparison table is this second one:

	31/10/2016 (spin off date)	Change	EOFY2018
Normalised Earnings {A}	$472m (for all of FY2016)		$634m
No. of Shares {B}	364m		392m
eps {A}/{B}	$1.30c	+32c {D}	$1.62
Owner Equity {C}	($2,443m-$460m)		$2,976m
Owner Equity per share {C}/{B}	$5.45	+$2.14 {E}	$7.59
Return on Incremental Equity / Share {D}/{E}		+15%

Note that 15% is well below the overall ROE figure of 21.3% achieved over FY2018 and also below the ROE figure of 18.5% over the last five years. 15% return on 'incremental equity' is nevertheless a good figure, the kind of figure that a Warren Buffett would be happy with.

I have mentioned before that YUMC came into being as a 'stock split' for the parent YUM corporation. This is a slight simplification of the truth. In fact at the time of the split, YUMC received an outside capital injection of $460m (AR2018 p115). This is as the result of two strategic investors being brought on board the share register:

1/ An affiliate of the 'Primavera Financial Group' called 'Pollis Investment L.P.' invested $410m.
2/ An affiliate of ''Zhejiang Ant Small and Micro Financial Services Co. Limited" called 'API Hong Kong Investment Limited" invested $50m.

The net effect of these transactions was to add 19.145m shares (along with the $460m) to the 363.758m shares that came into existence at the time of separation. These shares were added in the very last quarter of 2016 and so already appear on the FY2016 balance sheet information as presented in the table below.

	EOFY2016	Change	EOFY2019
Normalised Earnings {A}	$472m		$687m
No. of Shares {B}	383m		395m
eps {A}/{B}	$1.23c	+51c {D}	$1.74
Owner Equity {C}	$2,443m		$3,175m
Owner Equity per share {C}/{B}	$6.38	+$1.66 {E}	$8,04
Return on Incremental Equity / Share {D}/{E}		+31%