Buffett Tests Summary: FY2016 perspective
The four tests have been completed and the results are as follows:
Buffett Test 1 (post 1422): 'Top Three' position in chosen market. Result: Pass
Buffett Test 2 (post 1417): Increasing 'eps' Trend. Result: Pass
Buffett Test 3 (post 1400): Return on Equity > 15% for five years. Result: Fail
Buffett Test 4 (post 1401): Ability to increase Net Profit margin faster than inflation: Result: Pass
On the surface, passing three of these four tests looks good. However, in order to use 'company generated earnings' to feed into a Buffett growth model, a pass of all four tests is needed. The key failure point is reprised below.
|
FY2012 |
FY2013 |
FY2014 |
FY2015 |
FY2016 |
Return On Equity |
4.2% |
6.0% |
5.0% |
10.5% |
12.0% |
|
Returns like this are not enough to ensure that Turners will be earning more than their cost of capital across the business cycle. So the Buffett growth model cannot be used in this instance to provide a reliable valuation. This doesn't mean that TRA is necessarily a bad investment. In just means we have to find another method to evaluate the company.
SNOOPY
.
Capitalised Dividend Valuation Model (FY2017 Perspective): Preamble 1
Quote:
Originally Posted by
Snoopy
This doesn't mean that TRA is necessarily a bad investment. In just means we have to find another method to evaluate the company.
I am going to keep working with the earnings and dividend figures from the group as though things were already combined by FY2012 (refer post 1399) 'Merging DPC and TUA early'.
To summarize what has happened so far:
1/ I have assumed that DPC/TNR/TRA (I'll call it TRA from now on as that is the current company name) took over TUA prior to FY2012.
2/ I have assumed that the capital structure in terms of 'bank interest due' (and implied interest rates charged) from the party that completed the takeover (TRA), was superimposed over the old TUA structure in years FY2012, FY2013 and FY2014 (no need to assume this after EOFY 2015 as the takeover really had happened by then).
3/ The TRA earnings for FY2014 (year ending 31st March) have been mixed with the the TUA full year annual earnings for the period ending 30th June 2014. The predator (TRA) and prey (TUA) have different balance dates. So it is not possible to strictly combine earnings over a 'like for like' period from published company accounts. Earnings for FY2013 and FY2012 have been combined in a similar way.
The first question to answer is, what happens to the earnings from TUA, when the higher cost TRA borrowing structure is imposed upon them? All figures in the table relate to TUA, unless otherwise stated.
|
FY2011 |
FY2012 |
FY2013 |
FY2014 |
Bank Liabilities |
$31.457m |
$33.272m |
$36.423m |
$45.654m |
Interest Charges |
|
$0.895m+$0.917m |
$0.913m+$0.952m |
$1.148m+$1.013m |
Implied Interest Rate (TRA): (*) |
|
12.0% |
9.8% |
15.1% |
Implied Interest Rate (TUA) |
|
5.6% |
5.4% |
5.3% |
Implied Interest Rate Difference |
|
6.4% |
4.4% |
9.8% |
Implied Interest Charge Difference (Extra Interest) |
|
$2.059m |
$1.520m |
$3.996m |
(*) from post 1398 "DPC/TNR/TRA and implied borrowing Interest Rates"
SNOOPY
Capitalised Dividend Valuation Model (FY2017 Perspective): Preamble 2
The second question to answer is, how would the higher cost TRA borrowing structure have affected TUA profits?
Turners Auctions (TUA) |
FY2011 |
FY2012 |
FY2013 |
FY2014 |
EBIT |
|
$7.342m |
$7.948m |
$9.117m |
less Interest Charges |
|
$1.812m |
$1.865m |
$2.161m |
less Implied Interest Charge Difference (Extra Interest) |
|
$2.059m |
$1.520m |
$3.996m |
equals EBT |
|
$3.481m |
$4.563m |
$2.960m |
less Tax at 28% |
|
$0.975m |
$1.278m |
$0.829mm |
equals Extra Interest Adjusted NPAT |
|
$2.506m |
$3.285m |
$2.131m |
For comparative purposes, let's look at the underlying profit from TUA over my redefined FY2014, without those extra interest charges.
NPAT = (EBIT - I -T) =($9.117m - $2.161m - $1.948m) = $5.008m
As you can see it makes a huge difference. Net profit after tax has more than doubled. It is from these much higher profits that the old TUA was able to pay out such a good dividend stream to shareholders. So how much in normalized dividends did that old TUA actually pay?
Turners Auctions (TUA) |
FY2011 |
FY2012 |
FY2013 |
FY2014 |
NPAT |
|
$3.982m |
$4.380m |
$5.008m |
Actual Dividend Paid |
|
$3.012m |
$4.106m |
$4.380m |
Extra Interest Adjusted NPAT |
|
$2.506m |
$3.285m |
$2.131m |
You can see that the NPAT more than covered the actual dividend paid. But look one line further down on the table. This shows that my modelled adjusted profits no longer cover the dividend. And that leads me to a conundrum.
With a 'Capitalised Dividend Model' I am usually quite strict on using actual dividend figures in my valuations. Why so? Because actual dividends represent actual cash deposited in a shareholders bank account. Measuring actual cash in a bank account is a measure of investment return that no-one can really dispute, because 'it is there' and as a shareholder you can spend it. However, looking at the table above with the TRA funding structure -imposed retrospectively on the TUA group-, there is no longer enough money to pay that dividend. Long term, a company can only distribute the amount of imputed dividend that actually existed as tax paid net profit. And that means that under my alternative scenario, that dividend could not have been paid, because the underlying profit was not there to support it.
For modelling purposes this means I must change my approach. The modelled dividends paid will now be equivalent to all of my adjusted profits, but no more than those figures.
SNOOPY
Capitalised Dividend Valuation Model (FY2017 Perspective)
Turners Auctions (TUA) + Turners Limited (TNR/TRA) |
|
FY2012 |
FY2013 |
FY2014 |
FY2015 |
FY2016 |
FY2017 |
Modelled Dividend Paid {A} |
|
$2.506m |
$3.285m |
$2.131m |
No. Shares on Issue (TNR/TRA) {B} (*) |
|
24.057m |
27.395m |
55.966m |
63.077m |
63.433m |
74.524m |
Modelled Dividend Paid (cps) {A}/{B} |
|
10.42c |
12.00c |
3.81c |
Actual Dividend Paid (cps) (**) |
|
|
|
|
5c + 4c |
6c + 6c |
7c + 3c +3c |
(*) The number of TNR shares on isssue at the end of the financial year has been adjusted retrospectively for the 10:1 share consolidation. To see how the number of TRA shares on issue was derived refer to my post 1414 "Buffett Test 2: Increasing 'eps' Trend (FY2016 perspective): Preamble Part 2.
(**) The actual dividends paid by TNR/TRA over FY2015 and FY2016 were unimputed. This was because of prior losses incurred under the DPC/TNR/TRA structure. However, in my modelling the TUA group was already combined with DPC/TNR/TRA. Previous year TUA profits wiped out those previous year equivalent DPC/TNR/TRA losses. Under this modelled scenario, those FY2015 and FY2016 dividends would have been fully imputed. That's because looking at the combined picture, those prior offsetting DPC/TNR/TRA losses never happened. Further note that all dividends have been adjusted retrospectively to account for the 23rd March 2016 10:1 share consolidation.
From the above table the 'six year average' dividend payout was:
(10.42c + 12.00c + 3.81c + 9c + 12c + 13c)/ 6 = 10.04c (net)
Average Gross Dividend Yield (based on a 28% tax rate) is therefore:
10.04/(1-0.28) = 13.94c
Using a capitalized value gross interest rate of 7.5% (see thread An Investment Story - Geneva/Turners/Heartland, post 40), this translates to a fair value share price of:
13.94/ 0.075 = $1.86
That makes for sobering reading, when the last price paid in the market on Friday was $3.75!
SNOOPY
Capitalised Dividend Valuation Model (FY2017 Perspective): Epilogue
Quote:
Originally Posted by
Snoopy
<snip>
Average Gross Dividend Yield (based on a 28% tax rate) is therefore:
10.04/(1-0.28) = 13.94c
Using a capitalized value gross interest rate of 7.5% (see thread An Investment Story - Geneva/Turners/Heartland, post 40), this translates to a fair value share price of:
13.94/ 0.075 = $1.86
That makes for sobering reading, when the last price paid in the market on Friday was $3.75!
I dare say that this $1.86 valuation will come as a bit of a shock to shareholders. It did to me! But is it right? When you are constructing a scenario, by combining two companies (TUA and TRA) before they actually were combined, then whether it is 'correct' or not is possibly not the way to think of it. I think what I have done is consistent and based on actual dividends paid.
If TRA management today had been in charge of TUA back in 2012, 2013 and 2014, then it is unlikely that any of those three years of dividends would have been paid out. At the time TRA, or DPC as it was then, were busy raising new capital from shareholders, not paying out money to them!
I toyed with the idea of superimposing the current TRA dividend policy of paying out 50%-55% of NPAT back onto years 2012, 2013 and 2014. But that would go against what actually happened. And the current policy of "paying out 50%-55% of net earnings as dividends" could change in the future. So I figured it is best to stick to what actually happened and allow readers to apply their own 'fudge factor' to this valuation calculation as they see fit. I see the $1.86 price as a 'fair valuation', assuming no growth across the business cycle. Turners management have already told us that they see themselves growing strongly. But what happens if the market for new/used cars stops growing? And defaults make finance sector earnings at lot more difficult? $1.86 would look a much more likely share price target under those circumstances!
I think given current 'interest rate' and 'economic' settings, that Turners shares will grow towards a much higher 'fair valuation' figure. But this puts me in the category of 'one of the Turners faithful'. And whether my faith in holding Turners shares that I could sell on the market last week at $3.75 means the share is ultimately worth $3.75 is something that will play out over the next few months. Just because management say they have a "growth plan", does not mean that they will definitely achieve it. I wonder if the 'smart ones' in this situation are really the TRAHB bond holders?
SNOOPY