Skellerup (SKL) Fundamentals

[blackie]

cheers Snoop
thanks for the clarification ROCE Vs ROE.
I always enjoy and learn a lot from your detailed posts

[winner69]

My mate Snoops says he prefers ROE over ROCE because ROCE leaves out interest and taxes (which are real cash). Fair enough

I prefer to use an equivalent thing called ROIC (Return on Invested Capital) which measures the returns from what shareholders and lenders have invested in the business. Formula is (EBIT * (1 - Tax Rate) / (Equity + Borrowings)

You can see it counts tax so gives tax paid returns to investors (both shareholders and lenders) - shareholders gets divies and lenders get interest and some may be reinvested in the business.

I think its a better measure of how a company uses capital than ROE because it expresses returns on total capital used rather than just what shareholders have put in.

One always wants to see ROIC higher than the company's cost of capital - otherwise it is destroying economic value

In SKLs case ROIC is currently 13.4% which is pretty good. I reckon their cost of capital is about 7% (PWC reckon 4.5%) so SKL are making pretty good excessive profits (over the cost of capital employed)

Those excessive profits are about $14,5m (some call this EVA which is Economic Value Added)

There's another thing called MVA which is Market Value Added - this being the difference between the market cap of the company and shareholders equity. In SKL case the market is valuing SKL equity of $184m at $882m so MVA is just under $700m - cool eh

This MVA of $700m is in reality the present value of all the SKL excessive profits (over its cost of capital) in the future.

Working backwards it sort of implies that these are growing to grow at about 5% pa

Summary - SKL a value creating company that has been rewarded to the tune of $700m (today) for being so ...cool eh

Well, that occupied me with the rain coming down outside but facing up to taking the dog out in the rain




This is essentially where SWS was coming from. Much better than PE ratios etc which are in esence a cheats way of doing DCFs without realising it

[Snoopy]

cheers Snoop
thanks for the clarification ROCE Vs ROE.
I always enjoy and learn a lot from your detailed posts

Yes, and if I might offer a few more thoughts on the distinction.

ROCE = (Earnings Before Interest and Tax) / (Total Assets - Current Liabilities)

ROE = (Net Profit After Tax) / (Total Assets - Total Liabilities)

I think which of the two metrics you regard as 'better' depends on the angle at which you are approaching the company.

If you are looking at taking over the company for example, the ROCE is probably the better measure of how good the 'underlying bones' of any company is. This is because the divisor principally(*) using 'total assets', takes the amount of long term borrowing the company has 'out of the equation'. If you are taking over a company the amount of long term borrowing can be changed, simply by increasing or decreasing the amount of shareholder capital that any new owner wants to leave in there. As a new owner, I would be more interested in how good the 'underlying bones' are of what I want to buy, not the capital position that the previous owner put the company into.

By contrast as an investor, I am very interested in the amount of interest and tax that a company is paying today, because through my proportional ownership as a shareholder and our imputation credit system, it is I that is paying those interest bills and paying that tax. For me as an investor, that makes ROE the hands down winner of the two.

(*) I am not sure why 'current liabilities' are subtracted out. If what are normally part of a company's 'long term liabilities' happen to be maturing in the current year, this statistic will be functionally distorted by doing so. I guess most companies either have staggered debt maturities or reset their banking arrangements before the year those banking arrangement expiry dates fall due. So in practice this isn't a big issue. The other part of 'current liabilities' relates to how quickly you as a company pay your bills. If you decide not to pay your creditors, for example, then your 'current liabilities' will increase and so will your ROCE. Taking 'current liabilities' out of the denominator means that management that do not pay their bills on time are 'rewarded' with a higher ROCE rating. Is this statistic saying that business owners who keep their suppliers squashed down with late payments to the supplier begging bowl are better business people?

SNOOPY

[winner69]

The reward for being a wealth creator (ie increasing returns on invested capital) has been amazing over the last 5 years

Where as shareholder funds per share have increased from $0.81 in 2016 to about $0.97 Skellerups MVA per share has increased from $0,45 to $3.55

So since 2016 share price up $3.27 - of which $3.10 has come as that reward for doing so well

Truly staggering

One does have to think that maybe such 'market rewards' wont be as much over the next five years

[Snoopy]

ROCE = (Earnings Before Interest and Tax) / (Total Assets - Current Liabilities)

If I take the unadjusted earnings from FY2020 I calculate ROCE as follows:

($39.631m + $2.582m) / ($283.642m - $36.550m) = 17.1%

The 'Simply Wall Street' guys used slightly different numbers to get an ROCE of 20%. I cannot explain where SWS get their slightly different numbers from.

Sometimes I get so used to doing certain calculations 'my way' I forget that 'real analysts' do things differently. When you are doing a calculation over an 'asset base' it is more usual to try and represent that 'asset base' as an average value over the entire year. I just take the end of year figure as representative. If we take the start of the year and the end of the year as representative measuring points of the asset base over the year, the calculation of ROCE changes to this:

($39.631m + $2.582m) / 0.5[($283.642m - $36.550m) + ($257.059m - $28.913m)] = 17.8%

That still doesn't get us to the SWS figure of 20%. But I can do one better. We also have the half year balance sheet figures as found in the half year annual report. So we can create a 'triangulated' picture of assets that will give us a better annual view picture. Incorporate that picture and the ROCE calculation changes again to this:

($39.631m + $2.582m) / 0.333333[($283.642m - $36.550m) + ($275.347m - $29.906m) +($257.059m - $28.913m)] = 17.6%

We are still not close to that 20% figure and this is as detailed as I can make my calculation from the information that is in the public domain.

At this point I should mention that the reason I do not do this more extensive view of the asset base picture normally is that it is all historical information anyway. As an investor I am interested in future indicators. Generally the best 'future indicator' is the most recent information and that is the last published balance sheet. Finally I should add that not doing the asset averaging that I have done in this example is so much easier! If I can save work and make the calculation more relevant at the same time, why not do it?

SNOOPY

[Snoopy]

My mate Snoops says he prefers ROE over ROCE because ROCE leaves out interest and taxes (which are real cash). Fair enough

I prefer to use an equivalent thing called ROIC (Return on Invested Capital) which measures the returns from what shareholders and lenders have invested in the business. Formula is

(EBIT * (1 - Tax Rate) / (Equity + Borrowings)

You can see it counts tax so gives tax paid returns to investors (both shareholders and lenders) - shareholders gets divies and lenders get interest and some may be reinvested in the business.

I think its a better measure of how a company uses capital than ROE because it expresses returns on total capital used rather than just what shareholders have put in.

There is one other reason why I calculate ROE. I need that figure as part of my compounding earnings spreadsheet calculation for those companies that pass the 'Buffett Test'. But your post got me thinking Winner.......

I had considered that as a shareholder I owned the company equity (or a share of it at least) while the banks owned the debt. I was much more interested in the bit of the company that I owned and less interested in what the banks owned. Thus I was more interested in ROE (based on my equity) rather than ROIC (because that includes the banks money in the business that I have no claim to). But that thinking is not quite right is it?

As a part company owner, I have a contract with the bank to repay any money they lend to me. So although the bank loan money is not my money, I am still responsible for it, in the sense that my business is using it. That means that for each share I own, I own a portion of the company equity plus a share of the company responsibility in looking after and using any bank loans the company has. Thus ROIC ( EBIT * (1 - Tax Rate) / (Equity + Borrowings) ) is an entirely personal measure of return for shareholders and not a hybrid measure of return for shareholders and bankers as I had previously thought. (My incorrect logic in assuming ROIC was a hybrid measure of return is the reason I had not paid much attention to it up until now).

I don't know why I haven't figured all this out before now. I guess I am a bit slow at times :-( ........

To some extent though Winner, I think you have fallen into the same trap as I did. You say:

"ROIC (Return on Invested Capital) which measures the returns from what shareholders and lenders have invested in the business."

In fact ROIC is all about 'shareholder returns' only, because the shareholders dictate what capital the lenders put into the business. ROIC does not measure the returns on what lenders have invested in the business, as your sentence above implies. Lenders capital returns are the interest they are paid on their loan capital lent to the company are they not? - nothing directly to do with ROIC!

SNOOPY

[Snoopy]

My post 882 suggests an earnings per share growth rate of some 3% per year, verses the SWS figure of 8.45% per year. The justification for that 8.45% is that earnings have grown even faster than that over the previous five years (10.3% per year). My adjusted calculations (post 857) show a five year 'eps' growth rate of:

11.1(1+g)^5 - 14.8 => g = 5.9%, or much lower than SWS is telling us happened.

So based on historical record reality, I think the SWS projected growth rate is too high. It would be great if I was wrong and SWS turned out to be right. But I like to do my calculations based on earnings growth that can be demonstrated to have happened in the past, and I can't replicate the SWS calculations on growth.

There's another thing called MVA which is Market Value Added - this being the difference between the market cap of the company and shareholders equity. In SKL case the market is valuing SKL equity of $184m at $882m so MVA is just under $700m - cool eh

This MVA of $700m is in reality the present value of all the SKL excessive profits (over its cost of capital) in the future.

Working backwards it sort of implies that these are growing to grow at about 5% pa

I am going to have one last crack at sorting out these earnings growth figures. Projected earnings growth figures make an enormous difference to the value of a company. It bothers me intensely that a firm like Simply Wall Street (SWS) can have such a wildly different view of earnings growth than the views of Winner (on future growth) and myself (on historical growth).

My first thought is that perhaps SWS are talking about EBIT growth and not NPAT growth. After all ROCE is based on EBIT and not NPAT. We are looking the the five year period book ended by the end of FY2015 and the end of FY2020.

EBIT for FY2015 was $31.119m. Per share this amounts to: $31.119m / 192.806m = 16.1cps

EBIT for FY2020 was $42.486m. Per share this amounts to: $42.486m / 194.753m = 21.8cps

To calculate the five year growth rate 'g':

16.1c (1+g)^5 = 21.8c => g= 6.2%

That is a little higher than the 5.9% I previously calculated for NPAT eps growth. But still nowhere near the SWS figure. of 10.3%

My next thought is that Simply Wall Street is USA based, So maybe they are calculating their returns on a USD basis? The exchange rates on the relevant dates are as follows:

30-06-2020: NZD1 = USD0.6451

30-06-2015: NZD1 = USD0.6781

The NPAT growth in USD terms over five years is therefore:

(11.1/0.6451)(1+g)^5 = (14.8/0.6781) => g = 4.9%.

The USD growth is less than the NZD growth! I remain baffled as to how SWS calculate historical earnings growth over five years of 10.3% compounding.

SNOOPY

[forest]

But still nowhere near the SWS figure. of 10.3%.

I think the discrepancy might be that SWS has made an adjustment for the dividend SKL has paid over that period.

[winner69]

Snoops - equity or debt or owns or is responsible for either is interesting discussion but isn't it a bit meaningless.

ROE and ROCE (and as I prefer ROIC I talk about that) are essentially measures of how a company uses the capital it has.

That capital has a cost - cost of equity (can be a bit subjective) and cost of debt are real things and of course the aim should be to recover your 'costs' - otherwise one is not creating economic wealth. If cost of capital is x% you'd hope the company would make a greater return than x% on that capital eh.

As I said earlier I prefer ROIC as it measures returns on all capital used (and not just equity). A company can sometimes have an acceptable looking ROE but can actually not covering its costs of capital.

From a valuation perspective its often that the better the ROIC the higher multiples are given. Generating economic wealth is rewarded

[Snoopy]

But still nowhere near the SWS figure. of 10.3%.

I think the discrepancy might be that SWS has made an adjustment for the dividend SKL has paid over that period.

Excellent thought! I had been assuming all along that SWS were talking about company earnings. But perhaps they were talking about 'shareholder earnings' i.e. dividends?

Over FY2015, the dividend declared was 3.5c (interim) and 5.5c (final) for a total of 9.0cps.

Over FY2020, the dividend declared was 5.5c (interim) and 7.5c (final) for a total of 13.0cps,

This implies a five year growth rate of:

9.0c(1+g)^5=13.0c => g = 7.6%

That is greater than the 5.9% based on eps growth, but still short of 10.3%. There is the minor matter of there being full imputation credits attached to the FY2015 dividends and FY2020 dividends being only 50% imputed. But since I believe it is very likely the collection of SWS data is automated, I doubt they have taken into account such subtleties. And imputation credits are only useful to NZ shareholders anyway. If they had taken imputation credits into account the 'g' factor would have been lower than 7.6%

The other interpretation of your post is that SWS has made an adjustment for dividends received that has been added to earnings growth. That sounds like double counting though, as the dividends are a merely the cash manifestation of earnings in shareholder hands.

SNOOPY