SCT - Scott Tech

[Snoopy]

What is clear is that this change is bad for cashflow. The R&D expenditure must be paid for up front by the company up front before any tax relief is forthcoming.

Scotts are also receiving R&D assistance in Australia. Information on Australian governmental assistance may be found here:

https://www.business.gov.au/Grants-a...-Tax-Incentive

The Australian R&D tax incentive replaced the Australian R&D tax concession from 1 July 2011, and applies differently from the concession.

Australia, right now, has a lower minimum spend to qualify for assistance of $A20,000 (vs $NZ50,000 in NZ). There is also a possibility of claiming for overseas R&D expenditure ('Advance and Overseas Findings'). This is specifically ruled out in New Zealand.

As an R&D tax incentive, it goes without saying that a company must be liable for paying tax in Australia to qualify. All applications must be pre-registered. Scotts qualify here, largely through their Australian subsidiary MAR.

The tax offset for a private company that can be claimed is 43.5% (for aggregated R&D turnover less than $20m) or 38.5% (for aggregated R&D turnover greater than $20m). Since the Australian company federal income tax rate is 30%, it looks like you can claim back an amount greater than your tax bill. However, Australian businesses face other taxes as well, like payroll tax (an Australian state and territory tax, levied at about 5% in Victoria and NSW). So it is possible that in practice Australian companies would not be able to claim more tax back than they were due to pay after all. Incremental notional R&D deductions above $100m of R&D spending are claimable at a lower level (the Australian company tax rate).

The Australian R&D tax incentive (no dollar cap on tax rebate with between 100% to 145% of tax paid returnable) looks to be a lot more generous than the New Zealand system ($18m dollar cap on tax rebate with 54% of tax payable refundable). I wonder if that difference is enough for Scotts to abandon their R&D program in New Zealand and move the whole lot to Australia?

SNOOPY

[kiwidollabill]

Scotts are also receiving R&D assistance in Australia. Information on Australian governmental assistance may be found here:

https://www.business.gov.au/Grants-a...-Tax-Incentive

Australia has a lower minimum spend to qualify for assistance of $A20,000 (vs $NZ50,000 in NZ). There is also a possibility of claiming for over seas R&D expenditure ('Advance and Overseas Findings'). This is specifically ruled out in New Zealand.

As an R&D tax incentive, it goes without saying that a company must be liable for paying tax in Australia to qualify. Scotts qualify here, largely through their Australian subsidiary MAR.

Not only do you need to be liable for tax there you also have to conduct the R&D spend in Australia, cant get away with a transfer pricing it to the NZ entity...

If they have a UK business you can claim R&D credits from there as long as you show the R&D involves a UK entity and they dont care where in the world the R&D money is spent.

Or this is how our accountant explained it to me...

[Snoopy]

Not only do you need to be liable for tax there you also have to conduct the R&D spend in Australia, can't get away with a transfer pricing it to the NZ entity...

Couldn't Scotts do it the other way around though? By that I mean use transfer pricing to create most of their meat processing robotics profits in Australia and then do the R&D spending in Australia at MAR?

After all:

1/ Scotts biggest shareholder is Australian based (JBS Australia),
2/ The biggest medium term growth market for meat processing robotics is in Australia AND
3/ If they can get the automated beef boning room right there are potentially very large profits in Australia, and not just from JBS.

Of course if Scotts go this way, it is probably good-bye imputation credits for NZ shareholders!

If they have a UK business you can claim R&D credits from there as long as you show the R&D involves a UK entity and they dont care where in the world the R&D money is spent.

Or this is how our accountant explained it to me...

That would be fine if Scotts venture into the UK was profitable. Profits look hard to come by in UK/ Europe at the moment!

SNOOPY

[Snoopy]

Scotts are about to enter their tenth year of being on the R&D grant system in New Zealand. These R&D grants are distributed through Callaghan Innovation.

In dollar terms the 'old' tax grant scheme paid out 20% of the business’s eligible R&D spend up to $25m per annum

Scotts are also receiving R&D assistance in Australia.

The tax offset for a private company that can be claimed is 43.5% (for aggregated R&D turnover less than $20m) or 38.5% (for aggregated R&D turnover greater than $20m).

Time to tabulate all of this government support that Scott Technology have received over the last five years:

Financial Year	Government Grants NZ (GGNZ)	GGNZ/0.2 {A}	Australian Tax Credits (ATC)	ATC/0.435 {B}	{A}+{B}	Declared R&D Spend
2015	$0.673m	$3.365m	$0m	$0m	$3.365m	?
2016	$2.172m	$10.860m	$0m	$0m	$10.860m	?
2017	$0.926m	$4.630m	$0m	$0m	$4.630m	?
2018	$1.861m	$9.305m	$0.563m	$1.294m	$10.599m	$11.0m
2019	$2.026m	$10.130m	$1.112m	$2.556m	$12.686m	$14.0m

Notes

1/ FY2015 was the year in which MAR in Australia was brought into the Scott's fold.

2/ I haven't been able to find a declared R&D spend for FY2015, FY2016 and FY2017. For the two years in which I could find a declared R&D spend total (FY2018 and Fy2019), the declared total exceeds the 'sum of the parts total' that I have calculated. I would assume this is because not all R&D expenditure is recoverable via a grant or tax relief.

3/ For AR2019, the 'Government Grants related to Research and Development' are found in section A1. The 'Research and Development tax credited claimed (Australia) are found in section A2.

SNOOPY

[Snoopy]

The Research & Development (R&D) Expense to Revenue ratio ( {A}/{B} in the table below) measures the percentage of sales that is allocated to R&D expenditures. In shorthand I will call this ratio "R&D to T."

Financial Year	Declared or Estimated R&D Spend {A}	Turnover {B}	{A}/{B}
2015	$3.365m (e)	$72.298m	4.7%
2016	$10.860m (e)	$112.044m	9.7%
2017	$4.630m (e)	$132.631m	3.5%
2018	$11.0m (d)	$181.779m	6.1%
2019	$14.0m (d)	$225.093m	6.2%

The above is what 'R&D to T' looks like for Scott Technology. But what should we expect it to look like? 'R&D to T' is very industry dependent. So the only comparison that makes sense is to look at R&D amongst Scott's peer industry players.

New Zealand ranked 21st of OECD countries in GERD (Gross expenditure on R&D) in 2018, up from 28th five years earlier.

https://www.stats.govt.nz/reports/re...w-zealand-2018

Total R&D expenditure as a proportion of GDP for the whole country was 1.37 percent, still lower than the OECD average of 2.4%. The highest spending NZ sector for R&D in dollar terms was manufacturing.

A series of NZ companies are grouped together for statistical purposes as the as the "Technology Investment Network " (TIN)

https://www.callaghaninnovation.govt...siness-numbers

The Technology Investment Network consists of 450 export-focused New Zealand businesses operating in the high-tech manufacturing, ICT and biotechnology sectors. It analyses the results for 200 of the biggest (the “TIN200”). Over 2015-2016 the businesses’ expenditure on R&D represented 8.8% of total revenues. On the TIN 'R&D to T' comparative scale then, it looks like Scott's spend on R&D is toward the lower end of the TIN scale.

SNOOPY

[Snoopy]

Apart from the increase in cash signifying completion of projects, a large amount of cash -$2.75m- (including my rights issue subscription money, as U.S. points out!) is sitting on deposit as cash pending approval of the JBS Australia capital injection.

How things have changed since the rights issue was banked. Scotts have spent it all and debt is back on the balance sheet again. To get some idea of the debt holding burden going forwards, we need to look back at what the net debt balance was at the three documented points throughout the year (beginning, middle and end)

	EOFY2018	EOHY2019	EOFY2019
Cash & Cash Equivalents	$12.473m	$0m	$0m
Bank Overdraft	$0m	($5.678m)	($4.737m)
Current Portion of Term Loans	($3.321m)	($3.996m)	($4.217m)
Term Loans	($4.088m)	($2.904m)	($7.450m)
Total Net Bank Debt	$5.064m	($12.578m)	($16.404m)

From the published full and half year balance sheets, there is no way to know the distribution of net debt throughout the year. However, we can calculate a linear approximated average that gives us an indicative net debt figure for the year from the table above.

Indicative Net Debt Over FY2019 = ( $5.064m- $12.578m - $16.404m) / 3 = -$7.996m {B}

The net interest paid over FY2019 was: $0.020m - $1.715m = -$1.695m (A)

So the net indicative interest rate paid was {A}/{B}:

$1.695m / $7.996m = 21.2%

That seems very high. Have I made a mistake?

SNOOPY

[Snoopy]

Indicative Net Debt Over FY2019 = ( $5.064m- $12.578m - $16.404m) / 3 = -$7.996m {B}

The net interest paid over FY2019 was: $0.020m - $1.715m = -$1.695m (A)

So the net indicative interest rate paid was {A}/{B}:

$1.695m / $7.996m = 21.2%

That seems very high. Have I made a mistake?

To answer my own question - no mistake!

There are several possible explanations as to why the interest rate that I calculated above is so high.

1/ Bank Debt Changing Seasonally: Let me split the year into four parts and not two. Now let's average the indicative debt over five snapshot in time points that bookend these four periods of the year.:

( $5.064m - $12.578m - $12.578m - $16.404m - $16.404m) / 5 = -$10.580m

That average debt is still consistent with the debt figures printed is in the Annual Report and the Half Year Accounts. But the timing of the debt burden is different. In this case the indicative interest rate is:

$1.695m/ $10.580m = 16.0% (still high)

2/ One Off Fees: The 'Finance Costs' could include some kind of 'set up' fee in addition to the interest expense. This explanation seems less likely because there were term loans both at the start and the end of the year. But Scott's may have suddenly needed to raise funds and incurred penalty fees (see note 3). However, from p52 AR2019 there is a credit line of close to $NZ30m of which only about half is being utilised at balance date. So perhaps Scott's is paying handsomely to have this credit line available?

3/ Payments Withheld as a Result of Project Delays: There was a large jump in 'Contract Assets' of $8m over the year. 'Contract Assets' represent work performed for customers that has not been invoiced for contractual reasons. When work that is already done is invoiced, it turns from 'Contract Assets' to 'Trade Debtors' in the Balance Sheet. Scotts have told shareholders that three big projects have run into unexpected problems during FY2019.

"In the half year to 28 February 2019, the Board noted that several projects had a significant impact and as these projects continue they will also impact the second half of this fiscal year. These projects are expected to be fully resolved by the time we enter the 2020 fiscal year."

These were cutting edge projects and the associated cost overruns are effectively R&D that can be used in future high tech installation projects. But contract underperformance can also mean a delay in payment. Scotts may have had to suddenly increase its borrowings to account for the $8m deficit in the 'Contract Assets'.

4/ Incompetance: With the Chief Financial Officer leaving during the year, the company may have just stuck all their debts on the company Visa card at 20%+ interest rates ;-P (Edit: I put point 4 in as a semi-joke. However I see that from p52 AR2019 there is $0.512m on the Visa bill at balance date - ouch!)

5/ A combination of 1 to 4.

I am very uncomfortable about that 21% interest figure. The 16% calculated in this post is better. But the disruptive project elements that lead to that 16% may happen again. So I am sticking with that 16% figure as the indicative interest rate figure for FY2020.

What about the indicative level of debt for FY2020 that leads to these interest charges? Scott's have said that these rogue projects will sort themselves out by FY2020, I am interpreting this to mean that the $8m jump in contract assets will turn into invoiced debtor assets and be collected. If this happens then the indicative beginning of year debt figure will drop to:

$16.404m - $8m = $8.404m

Assuming this is an indicative debt figure for the whole year, the 'Net Financial Expense' for FY2020 is:

$8.404m x 0.16 = $1.345m

This is a decrease from FY2019 of:

$1.695m - $1.345m = $0.350m

SNOOPY

[Snoopy]

Normalised Profit Calculation FY2018 (Refer AR2018 p33)

Declared Net Profit for FY2018	$10.772m
add Loss on Property Plant and Equipment Sales	$0.021m
add Due Diligence Expenses	0.72 x $0.271m
add Unrealised Forex Losses	0.72 x $0.271m
add Fair Value Derivative Losses held as Fair Value Hedges	0.72 x $1.579m
add Unrealised Interest Rate Swap Contract losses	0.72 x $0.043m
less Foreign Exchange Gains	0.72 x ($1.627m)
less Fair Value Gains on Firm Commitments	0.72 x ($1.579m)
equals Normalised Net Profit	$10.043m

My previous forecast profit in the post above assumes a full years contribution from Alvey which didn't happen. Adjusting for that, the forecast was:

= $16.927m - [ (365-100)/365 ] x $8.470m = $10.777m

To calculate this I had assumed a drop in profit, due to a net interest annual income drop to $0.162m

$0.664m - $0.162m = $0.502m

Actual interest income received was a little more, at $0.369m.

I didn't get around to making a profit forecast for FY2019.

Fast forward to FY2020. The way Scott's now operate, segmented results are declared in geographical areas. This makes it difficult if you are used to thinking of results in terms of what is happening to different global industries. I think it best to think of Scott Technology as having a base of earnings that does not grow from year to year, then superimpose on this picture the earnings growth from certain fast growing business units.

Taking the base level profit of FY2019,

FY2019 Profit Normalisation:

Most of these adjustments may be found on p39 of AR2019. I have:

a/ Subtracted the gain on sale of property plant and equipment of $0.106m and $0.237m (assumed non taxable)
b/ Added back in the unrealised loss on foreign exchange derivatives ($1.334m) and fair value losses on derivatives used as hedges ($1.216m) and the unrealised fair value losses on interest rate swap contracts ($0.346m) .
c/ Subtracted foreign exchange gains ($0.008m) and fair value gains on firm commitments ($1.216m) .

$8.604m-($0.106+$0.237m)+ 0.72($1.334m+$1.216m+$0.346m-$0.008m-$1.216m)= $9.464m



How do I see things developing over the current year FY2020?

Amount	Description	Calculation
$9.464m	Baseline Adjusted Profit FY2019
+$0.252m	Interest saved from lower interest intellectualbill	$0.350m x 0.72
+$1.649m	Meat Industry Robotics (incremental)	$10.994m x 0.15 [See Note (A)]
+$0.705m	Transbotics (incremental)	0.15 x $4.7m [See Note (B)]
-$0.500m	Appliance Production Lines	My post 800
+$0.630m	Mining	2 x $0.315m [See Note (C)]
-$1.458m	Loss of NZ R&D Grant (scheme expired)	0.72 x $2.026m
$10.742m	Forecast NPAT Total

With 78.311m shares on issue, this equates to an earnings per share figure of:

$10.742m / 78.311m = 13.7cps

At $2.20,SCT is trading on a projected PE of 220/13.7 = 16.0

Notes

A/ Meat Industry Robotics Thanks to the expansion of the Scott Technology site at Kaikouri Valley in Dunedin, FY2020 represents the first year in which the meat industry robotics team have double the area to build their meat industry robotics projects. From my post 485, the aim is to supply 5-10 automatic robotic projects per year. The estimated value of a full production line is $12m to $13m, or some $6.5m for half automating a production line. Most growth looks to come from Pork (see Pork Primal System already drawn up on pages 27&28 of AGM2019 presentation) and Beef processing in FY2020, because the lamb market in Australasia is significantly automated already.

Seven half projects completed over FY2020 would produce revenues of: 7 x $6.5m = $45.5m. This represents an increase of:

$45.5m - $34.506m = $10.994m

From post 485, the margin on this increased turnover is approximately 15%. This allows us to calculate the NPAT incremental profit from the increase in meat industry turnover


B/ Automatic Guided Vehicles We know from slide 6 of the Forsyth Barr March 2019 Emerging Companies Conference that the EBITDA margin at Transbotics is 20%. Interest is not negligible and Depreciation and Amortisation at Technology companies can be high. So I am assuming a NPAT profit margin of 15%

Sales over the last five years at Transbotics have ranged from $US4.5m ($NZ6.9m) to $US11m ($NZ16.4m) over the years 2014 to 2108 (slide 30 same presentation (using exchange rate of $NZ1 = $US0.67). In the year that Transbotics came into the Scott fold (FY2018) the annualised sales rate worked out to be:

$NZ4m x (365/122) = $NZ12m

The AGV market has a target growth rate of 30%+ (slide 14, Scott Presents with Moelis November 2019 Slide Show). So I am estimating Tramsbotics revenue was 1.3 x $12m = $15.6m for FY2019 and 1.3 x $15.6m = $20.3m over FY2020, That equates to incremental revenue of $4.7m. The incremental increase in NPAT from that can be estimated as follows:

0.15 x $4.7m = $0.705m

C/ Mining

Hopkins noted in his AGM address that:

"we expect Mining to rebound significantly in 2020,"

"Opportunities for the Mining sector is primarily for Scott’s sample preparation systems but through recent developments, extends into field automation ,such as robotic refuel,robotic idler change and automated fire assay."

Multiple Robo-prep installations are planned for FY2020. The addressable market for these is judged to be $20m to $50m per annum.(Moelis November 2019 Presentation , Slide 13). The largest installation so far was an automated Sample Preparation System for Pernoles, the second largest mining company in Mexico. This build took 5 months, largely at Scott's Sydney engineering base. MAR, now "Scott Australia' had a turnover of about $7.7m when acquired in FY2015. With organic growth that could be $10m today. If half of MAR's resources were assigned to this robo-prep project for 5 months, that would account for:

0.5 x 5/12 x$10m = $2.1m of turnover.

If we assume a 15% profit margin, this one project could have contributed: 0.15 x $2.1m = $0.315m of the group profit after tax. We can use this figure to judge the NPAT effect of an incremental two extra Robo-prep system sales on net profit.

D/ 'Robotworx' profit assumed unchanged.

E/ 'Alvey' profit assumed unchanged due to the continued uncertainty and flow on fall out from Brexit.

F/ 'HTS-110' Superconductor Technology Subsidiary assumed to be no longer a meaningful contributor to the group (it wasn't mentioned in the Annual Report).

G/ I have not made any allowance for costs relating to the closure of "DC Ross". However this is a 'one off' event that is unrepresentative of the ongoing operation of the business.

SNOOPY

[percy]

On your figures eps will be 9.29 cents per share,and with the share price at $2.19 the PE ratio is 23.57,,
I note the dividend yield is a very modest 3.65%.

[Snoopy]

Some musings from the presentation on the current state of the appliance business.

A fridge selling for $300 fifteen years ago still sells for $300 today. So Appliance makers globally are under huge cost pressures. In Europe alone, three manufacturers have disappered completely over the last few years.

After the GFC, and the accompanying housing downturn, most appliance manufacturers had their lines turning over at only 30% capacity. In tight times, the first thing to be cut is capital expenditure. The only time capital expenditure is not cut in hard times, is when government legislation demands new environmental standards, necessitating new product design and manufacture. Big Appliance players normally work on 120 to 130 day payment terms, tough for smaller contracting companies like Scotts. Nevertheless, the 30-40% deposit on order is helpful to cashflow at the front end of any project.

However, a new order has been received just recently. So that means Appliance manufacturing lines in FY2016/2017 will likely be in better shape than in FY2015 position.

One of the most important predictors of profit for Scott Technology in any year is how well the 'Appliance Line Production' section of the business does. Appliance Line Production projects are usually large (up to $20m in gross value) and take from several months to two years to complete, once an order is received. The flow of these jobs is often lumpy from year to year. Most of this business is done in the United States, some in China (which often uses the $US as the functional currency of payment) and Europe. Once Scott's receive an order, they hedge their currency position in the functional currency of the job to control their costs. This occurs when the labour component of the job is paid in NZ dollars but the customer will be paying in a different currency. Scott's report their currency hedging in their annual report. So I thought it would be an interesting exercise to compare:

1/ The summed hedged currency position in US dollars and Euros, converted to NZ dollars, at balance WITH
2/ Sales in the appliance division in the ensuing year.

Can a predictive pattern of future Appliance Division sales be gleaned by looking at the hedge book?

	2019	2018	2017	2016	2015	2014
0-12 month hedging 'Sell USD'	$11.326m	$25.241m	$3.121m	$3.177m	$3.640m	$1.030m
0-12 month hedging 'Sell Euro'	$0m	$0.097m	$0.272m	$0.118m	$0.641m	$0.573m
0-12 month hedging Total	$11.236m	$25.338m	$3.393m	$3.295m	$4.281m	$1.603m
Appliance Division Turnover: 'Subsequent Year'	$N/A m	$45.069m	$41.069m	$26.308m	$20.181m	$13.606m

What in the way of useful information can be got from this table? When the actual sales come through they are always significantly greater than that implied by the hedged position. That's good. This could indicate that although these project completion times are long, most are completed within the twelve months from 'go to whoa'. The hedging on the balance of these completed jobs , if any, might go unreported. An obvious anomaly in this table is in 2017 when a relatively modest hedging position turned into a large dollop of Appliance Division sales. That position can be juxtaposed against the FY2018 result where a much larger of proportion of hedged sales was booked compared to actual sales (am anomaly going the other way). I have to conclude that my 'forecasting technique' may not be that useful!

Chairman McLauchlan claimed to have "a number of large projects in the final stages of negotiation." These are not necessarily Appliance Division projects.

McLauchlan goes on to say

"This is being driven by businesses wanting to remove labour from their processes due to the sharp reduction in labour force participation in most geographies as a result of an ageing work force."

Since most appliance manufacturing companies are already highly automated, this suggests to me that these projects are most likely in other areas such as meat industry automation. Consequently I am forecasting a decrease in the contribution made by the Appliance Line Manufacture business for the FY2020 year.

To see how a revenue drop in the Appliance Division might affect profit, we have to go right back to FY2013 and earlier when 'Automated Equipment' (effectively the Appliance Line Manufacturing Division) reported separately:

	2013	2012	2011	2010	2009	2008	2007	2006	2005	2004
Automated Equipment NPAT	$2.099m	$0.034m	$2.455m	$3.095m	$0.597m	($1.598m)	$3.042m	$0.242m	$0.315m	$3.716m
Automated Equipment Turnover	$32.329m	$29.499m	$32.150m	$30.800m	$22.141m	$15.843m	$29.186m	$27.510m	$40.263m	$35.789m

FY2008 was the year of the GFC. Two big projects were cancelled, but wages still had to be paid. That was an extreme downturn. The reference year I will use to forecast FY2020 is FY2011. In that year in a slowing economy, a $0.5m drop in profit was recorded.

SNOOPY