AGL - Accordant Group. was previously AWF Madison, Allied...

[winner69]

Don’t make much money do they ....but the believers should be more than happy

Love the way they highlight that netprofit was $1.32m UP on the immediate preceding six month period but 36%DOWN on last year (even though the recent acquisition is producing results)

Don’t think those statements that’ll fool Snoopy

https://www.nzx.com/announcements/343202

[winner69]

When I saw an announcement from AWF headed ‘Media Release’ I thought hell surely the blowtorch man on the convention centre roof wasn’t an Allied worker

Phew - it wasn’t

[jg8512]

Don’t make much money do they ....but the believers should be more than happy

Love the way they highlight that netprofit was $1.32m UP on the immediate preceding six month period but 36%DOWN on last year (even though the recent acquisition is producing results)

Don’t think those statements that’ll fool Snoopy

https://www.nzx.com/announcements/343202

Winner, I'm not quite sure why the believers should be happy?

I thought that was a very weak result. they may be confident about the full year, but the 1H was very soft. Even with an on-expectations 4 month contribution from the new acquisition, the non-AWF segment profit was only flat year on year. And AWF's segment profit halved. FFS.

what bothers me a lot is the debt. AWF say $28.4m. But that excludes the $14.8m of capital lease commitments (which are debt! - they have to pay in cash) and $4m of contingent consideration (which if jackstonstone meets expectations they have to pay for too).

their full year pre-tax profit expectation is above $4.2m. given a 32% tax rate in 1h, that implies a NPAT of $3m (minimum). debt / NPAT of 7x at their stated debt figure. And much worse if you allow for the other amounts above. Sure, mgmt are confident they can do better than $4.2m pre-tax, but given the number of excuses they regularly trot out blaming sector regulation and policies - an inherent feature of their sector IMHO which they ought to manage better than anyone else - I'm not sure why investors would have much confidence in mgmt's assurances.

What price would this thing trade at if board and mgmt took the more prudent approach of cutting the dividend to improve the balance sheet?

Disc: I still hold (but not many left now)

[stoploss]

https://www.stuff.co.nz/business/119...p-agency-staff

Disc: Not held.

[Snoopy]

Financial Year	2015	2016	2017	2018	2019	2020e (low)	2020e (high)
EBITDA (Snoopy produced *) {B}	$12.729m	$11.945m	$12.751m	$11.751m	$7.679m	$10.397m	$12.797m
Finance Cost {C}	$2.109m	$1.333m	$1.193m	$1.297m	$1.380m	$1.232m	$1.232m
Interest Coverage {B}/{C} (target >3)	6.0	9.0	10.7	9.0	5.6	8.4	10.3
Net Bank Debt {D}	$18.608m	$21.870m	$32.383m	$29.731m	$26.643m	$23.643m	$23.643m
Leverage ratio {D}/{B} (target <3)	1.5	1.8	2.5	2.5	3.5	2.3	1.8

(* I calculated EBITDA from the income statement using the formula: EBITDA = NPBT + I + DA)

This means if Bennett's forecast figures for AWF in FY2020 can be believed and the white collar side of the business remains steady, then the banking covenants are 'back on track'. I wonder if we will get a first quarter update on FY2020 at the AGM on Wednesday?

The dividend reinvestment plan raised a useful $1.6m in new capital over FY2019. Trade and other receivables verses the equivalent payables were well controlled in FY2019.

	Trade Receivables {A}	Trade Payables {B}	Net Receivables {A}-{B}
FY2019	$32.629m	$28.527m	$12.174m
FY2018	$41.101m	$24.186m	$8.443m

Net receivables were down by $4.131m year on year. This was a key factor in allowing AWF to pay back $3m in debt.
Will 'Trade and Other receivables' be as well collected in FY2020: i.e. another incremental improvement? That will be something to look out for.

AWF really took a hammering on the market today, down over 10%, albeit on just 28,770 shares. The share price is back to early April 2020 levels now. I wonder if the market is looking at how AWF might continue to service their substantial debt? I took a guess that the Bennster would have the finances for AWF on the right track by EOFY2020. So how did it all work out?

Financial Year	2015	2016	2017	2018	2019	2020
EBITDA (Snoopy produced *) {B}	$12.729m	$11.945m	$12.751m	$11.751m	$7.679m	$11.593m
Finance Cost {C}	$2.109m	$1.333m	$1.193m	$1.297m	$1.380m	$1.584m
Interest Coverage {B}/{C} (target >3)	6.0	9.0	10.7	9.0	5.6	7.3
Net Bank Debt {D}	$18.608m	$21.870m	$32.383m	$29.731m	$26.643m	$29.822m
Leverage ratio {D}/{B} (target <3)	1.5	1.8	2.5	2.5	3.5	2.6

(* I calculated EBITDA from the income statement using the formula: EBITDA = NPBT + I (not including lease expenses) + DA. Likewise finance cost 'I' does not include lease expenses)

The banking covenants look to be back on track but only if you include 'Lease Depreciation' as part of the depreciation charges. However debt is still high, and won't go away until it can be paid off. So the big question is can EBITDA be maintained for FY2021, so that the bank maintains their confidence that debt load actually is manageable going into the future?

The dividend reinvestment plan raised a useful $1.703m in new capital over FY2020 (up from $1.569m in FY2019).

Trade and other receivables verses the equivalent payables were well controlled in FY2020, pointing to an improving pattern of debt collection.

	Trade & Other Receivables {A}	Trade Payables {B}	Net Receivables {A}-{B}
FY2018	$41.101m	$28.527m	$12.574m
FY2019	$32.629m	$24.186m	$8.443m
FY2020	$53.071m	$46.169m	$6.902m

It looks like FY2020 was another year of incremental debt collection improvement?

SNOOPY

[jg8512]

I haven't checked your figures closely Snoopy, but is the net debt of 23.643 right as at March 2020? I had $36m less cash = ~$30m .
ALso, are you excluding the contingent consideration (that's another >$3m of future cash to be paid out). and if revenues are weak, all those lease commitments still need to be paid.

[Snoopy]

I haven't checked your figures closely Snoopy, but is the net debt of 23.643 right as at March 2020? I had $36m less cash = ~$30m.

jg8512, you are quite right, but so am I. That $23.643m I had was under the column header '2020e', where 'e' stands for estimate. I was under the impression that AWF would be paying back debt over FY2020. But no, debt actually went up, largely because of the $10.520m shelled out for 'JacksonStone & Partners'. If you look further down in my post you will see that the actual net debt I calculated out at EOFY2020 was $29.822m - pretty much line ball with your figure.

ALso, are you excluding the contingent consideration (that's another >$3m of future cash to be paid out). and if revenues are weak, all those lease commitments still need to be paid.

Again good point. The EBITDA/I figure is a 'banking covenant' and I guess the reason for setting it at a maximum of 3 is to leave some headroom to pay other debts? I am happy to be corrected if my interpretation on this is wrong. The bank is pretty much at the top of any debt queue anyway, and as long as there is enough money left for them, then other creditors can 'get stuffed' (I think that is the correct banking industry parlance for such a situation).

Nevertheless I am a little unsure about 'I' the finance cost figure to use in my EBITDA/I ratio. On the balance sheet it is $2.084m. But if you turn to section A4 page 38, you will see that, for the first time, this figure incorporates and 'interest on contingent consideration' component and an 'interest on lease liabilities' component.

Interest on bank overdrafts and loans	$1.401m
Interest on contingent consideration	$0.101m
Interest on lease liabilities	$0.582m
Total Finance Costs	$2.084m

The previous year, only the equivalent of the $1,401m figure was used for the finance cost 'I'. So should I have added in those other two figures when calculating the finance cost or not? I remain a little confused on this point.

You are right about needing to find cash for the $1.463m contingent consideration for buying 'JacksonStone & Partners'. But there is still $3.259m worth of 'Customer Amortization' and $1.487m in 'Restraint of Trade' intangibles to be written off (p45 AR2020). That should more than balance out the contingent consideration payment due.

Personally I am not too concerned about the lease payment liability. The $9.098m lease liability is more than offset by the $11.107m of 'right to use' assets on the balance sheet. All the AWF employees have to work somewhere after all.

SNOOPY

[Snoopy]

It looks like the new NZ IFRS 9 standard on Financial Instruments will require the recognition of more losses by the likes of AWF on these labour contracts, whether these losses actually exist or not (from AR2019 p52):

"',,,requires lifetime expected losses for trade and other receivables to be recognised from initial recognition of the receivable."

Effectively AWF are being asked to use a 'probabilistic model', estimating their own write offs before they happen.

To some extent AWF is already doing this. AWF runs a 'provision for impairment' that is permanently on the books.

The extra provision from IFRS 9 for specifically 'trade receivables' (I take that to mean relating to the AWF sub brand?) is $0.371m (p71 AR2019). If we then move to the actual 'Provision for Impairment of Trade Receivables' (p53 AR2019), we can see that this extra $0.371m has gone straight in. In fact, following the write offs during the year, the 'Provision Impairment Balance' would have ceased to exist without this extra payment! Did the IFRS 9 standard change occur just in time to avoid a negative provisions coming onto AWF's books (if that is even possible)?

I am a little uncomfortable with the concept of holding a 'provision for impairment' at all. By having such a thing, bad deals can be 'smoothed out' by making a small annual addition to the impairment provision. Yet inside the provision there is a whopping write down that does not affect that year's profit and loss statement. Actual write downs for 'Trade & Receivables' over the last few years, compared to the consummate changes in provisions, have lined up as follows:

	2014	2015	2016	2017	2018	2019	Row Sum
Impairment Losses in P&L {A}	($0.358m)	($0.306m)	($0.510m)	($0.699m)	($0.655m)	($1.109m)
Impairment Write backs P&L {B}	$0.291m	$0.137m	$0.100m	$0.228m	$0.594m	$0.360m
Net Impairment in P&L {A}+{B}	($0.067m)	($0.169m)	($0.410m)	($0.471m)	($0.061m)	($0.749m)	($1.927m)
Actual Trade & Receivables Write down	($0.088m)	($0.204m)	($0.163m)	($0.163m)	($0.815m)	($1.034m)	($2.291m)
Provision for Impairment Balance	$0.377m	$0.342m	$0.589m	$0.897m	$0.143m	$0.229m

Over time, the last two rows should sum to about the same total. We can see that the declared result has incorporated a lower loss than the actual figures written off. But over the years a difference of $0.364m is not material. The one thing that strikes me about this table is that the 'write backs' have been quite high. One way to interpret that is that the initial write offs were overly conservative. I guess that is a good thing.

The end of year 'Provision for Impairment' balance is low enough that there can't be that much smoothing of write offs between years going forwards. I say that is good. What is not so good is that 'big impairment hits' will be felt against profits without delay - not great when you are struggling to reduce your debts! My gut feeling is that the 2019 provision balance, given it is now meant to reflect probabilistic losses that might occur, is now too low.

Time to update the bad debt performance of the group. This is a good thing to keep an eye on if a company has substantial debt on its balance sheet already.

The 'Impairment Losses Recognised' may be found in AR2020 Section C6 p52.
The 'Impairment Losses Reversed' may be found in AR2020 Section C6 p52.
The 'Write-offs to bad debts during the year' may be found in AR2020 Section C6 p52.
The 'Provision For Impairment Balance' may be found in AR2020 Section C6 p51.

	2014	2015	2016	2017	2018	2019	2020	Row Sum
Impairment Losses in P&L {A}	($0.358m)	($0.306m)	($0.510m)	($0.699m)	($0.655m)	($1.109m)	($0.301m)
Impairment Write backs P&L {B}	$0.291m	$0.137m	$0.100m	$0.228m	$0.594m	$0.360m	$0.046m
Net Impairment in P&L {A}+{B}	($0.067m)	($0.169m)	($0.410m)	($0.471m)	($0.061m)	($0.749m)	($0.255m)	($2.182m)
Actual Trade & Receivables Write down	($0.088m)	($0.204m)	($0.163m)	($0.163m)	($0.815m)	($1.034m)	($0.123m)	($2.414m)
Provision for Impairment Balance	$0.377m	$0.342m	$0.589m	$0.897m	$0.143m	$0.229m	$0.361m

Over time, the second to last two rows should sum to about the same total. We can see that for FY2020 the declared result (Net Impairment in Profit & Loss) has incorporated a loss almost double the actual figures written off. But looking at the cumulative effects of 'declared' verses 'actual' write offs over the years, the summed difference of $0.232m is not material. The one thing that strikes me about this table is that the 'write backs' have been quite high as a percentage of the declared losses. The possible exception to this is FY2020 where the write back percentage was 'only' 15%. One way to interpret that is that the initial write offs were overly conservative. I guess that is a good thing.

For FY2020, the end of year 'Provision for Impairment' balance has gone up from what look to be abnormally low levels from the previous two years. It is now a high percentage of the actual impairment losses. This offers the potential for declared impairment losses to be manipulated. The potential manipulation I allude to is for losses that should have been written off to be held in a bloated provision figure. However, because the absolute level of impairment losses has shrunk, any such manipulation, if it exists, should not have a material effect on the annual net profit result. With the FY2019 result I suggested that the 2019 provision balance, given it is now meant to reflect probabilistic losses that might occur, could have been too low. IMO raising the end of year provision for FY2020 has gone some way to fixing this.

SNOOPY

[Snoopy]

What anyone sees as 'one off' charges' are usually a matter of opinion. Nevertheless, with all the turmoil in the construction sector, it might be worth adding all the write downs back onto the 'AWF Division' profit, to see what profits might have done if the construction sector had not 'gone bad'.

	2014	2015	2016	2017	2018	2019
AWF EBIT {A}	$7.471m	$7.498m	$7.067m	$8.726m	$4,858m	$1.260m
Actual Trade & Receivables Writedowns {B}	($0.088m)	($0.204m)	($0.163m)	($0.163m)	($0.815m)	($1.034m)
No redeployment of Overseas Workers (saving) {C}						$1.500m
AWF EBIT - Writedowns {A}-{B}+{C}	$7.559m	$7.702m	$7.230m	$8.889m	$5.673m	$3.794m

This table indicates that just being more diligent in choosing the short term labour contacts that you wish to help with is not going to fix things. CEO Bennett realizes this and has already announced an unspecified amount of background cost cutting. How much of the EBIT decline is due to the adoption of the new probabilistic bad debt model under IFRS 9 in FY2019? The answer is none, because I have used 'actual write off figures' in the above table.

AWF experienced the collapse of three construction sector clients in FY2018 that had a significant effect on the 'blue collar' side of the AWF business.

http://nzx-prod-s7fsd7f98s.s3-websit...389/288702.pdf

CEO Simon Bennet, aka 'The Bennster' vowed to fix things. So what actually happened over FY2020?

AWF divisional earnings can be found on p31 of AR2020.
Actual Trade & Receivables Writedown can be found on p52 of AR2020.

	2014	2015	2016	2017	2018	2019	2020
AWF EBIT {A}	$7.471m	$7.498m	$7.067m	$8.726m	$4,858m	$1.260m	$1.692m
Actual Trade & Receivables Writedowns {B}	($0.088m)	($0.204m)	($0.163m)	($0.163m)	($0.815m)	($1.034m)	($0.123m)
Assuming no redeployment of Overseas Workers (creating a saving) {C}						$1.500m (1)
AWF EBIT - Writedowns {A}-{B}+{C}	$7.559m	$7.702m	$7.230m	$8.889m	$5.673m	$3.794m	$1.569m

(1) From 29th May 2019 market update

This table could be the indicator that AWF's construction sector woes are far from over. Remember that the latest figures are for the year ended 31st March 2020, and that includes only a few days of the lock down. The figures for 'AWF blue collar' are so bad I am wondering if I have made a mistake ??!?? Are the glory days of AWF Madison as a group over? It looks like 'The Bennster' has some explaining to do!

The June 8th announcement to the market contains some answers:

"The reduction in Group revenue was driven by AWF, where revenue fell by 16% to $97.4 million (Snoopy edit: I think the technical market term for this is 'Ouch!'). This reflected a number of factors. The Group took the decision in 2018 to start relinquishing low margin, high cost-to-serve business in favour of customers with higher engagement levels. In addition, AWF has recently seen growth in permanent recruitment, which furnishes higher margins than temporary. Taken together, these factors constitute encouraging momentum by AWF towards a greater contribution to Group profitability."

A further insight into AWF's woes was given in the interim report from p6.

"Of more concern is the industrial relations activity we are seeing, with aggressive action by a number of Unions. We currently have a case before the Employment Relations Authority, taken by a Union, challenging the status of a number of our workers. We believe this case has no merit and, in fact, they are seeking an outcome that would have applied if the Employment Relations (Triangular Employment) Amendment Bill had not been amended prior to it becoming legislation. This action has resulted in a shift away from contingent workers in a large client of ours who is seeking to take a more conservative balance of permanent and contingent workers."

"I believe the Union is effectively bullying a number of clients in this way. It is unfortunate for our workers as now a number have been left without work opportunities as a result. We are confident this will be resolved and may, in fact, result in stronger case law for legitimate providers of temporary employees. A balance of permanent and contingent workers optimises a workforce and provides good opportunities for workers, and we are confident common sense will prevail. The result has seen our white collar segment"

There were further problems redeploying workers from Christchurch (interim report p6).

"We had expected to resolve issues with the Labour Inspectorate relating to an investigation which commenced in May of 2018, following an exaggerated and sensationalised report by the media on 9 May 2018. This investigation stalled our renewal process for Accredited Employer status with Immigration New Zealand and resulted in us having little flexibility to redeploy many workers based in Christchurch, who were surplus to requirements, with large construction projects drawing to a close. We expect to announce a long overdue pathway forward in the coming weeks."

I take no 'momentum encouragement' myself from any of this! In a post Covid-19 world, are all the big corporates out there looking to hire new people? I can see a shift back to short term contracts, and more grindingly low margins for AWF blue collar temp workers.

SNOOPY

[Snoopy]

From the 29th May 2019 press release:

"Bennett said AWF had reduced its cost base, and was now geared to return 4% to 6% EBITDA on turnover approaching $120 million."

So lower and upper bound estimates of EBITDA for AWF are as follows:

$120m x 0.04 = $4.8m, $120m x 0'06 = $7.2m

Since these figures already include the cost savings I was trying to estimate in my post 791, I have removed those from my updated calculation.

	Lower Estimate FY2020	Higher Estimate FY2020
EBITDA (AWF Division)	$4.800m	$7.200m
EBITDA (Madison/AbsoluteIT)	$5.597m	$5.597m
less Head Office Admin and Expenses	($2.649m)	($2.649m)
less Net Interest Expense Paid	($1,232m)	($1.232m)
less Depreciation and Amortization	($3.445m)	($3.445m)
equals Net Profit Before Tax	$3.071m	$5.471m
less Income Tax @ 30%	($0.921m)	($1.641m)
equals Net Profit After Tax	$2.150m	$3.830m

Second Impression? It is looking like a slow crawl back from the construction industry supply profit bomb hole. At best profits will be back to FY2014 levels. But we have one more year of $2m odd in hidden amortizing customer relationship cashflows for FY2020. That means the amount of cash available to pay the dividend this year is $4.150m and $5.830m. Even with the growing number of shares on issue as a result of the DRP (let's say 36m in total by this time next year) that allows a 'dps' payment of 11.5c to 16.2c without 'borrowing from the bank to pay the dividend'. I have to admit that dividend payment might be looking shakey by FY2021. But Simon and his team will have had a couple of years 'reinventing the business for growth' by then. So all will be 'hunky dorey' by then, right?

I always find it a useful exercise to go over old predictions and see where you went wrong.

	Lower Estimate FY2020	Higher Estimate FY2020	Actual Profit FY2020
EBITDA (AWF Division)	$4.800m	$7.200m	$1.962m
EBITDA (Madison/AbsoluteIT)	$5.597m	$5.597m	$7.156m
less Head Office Admin and Expenses	($2.649m)	($2.649m)	($2.876m)
less Net Interest Expense Paid	($1,232m)	($1.232m)	($2.084m)
less Depreciation and Amortization	($3.445m)	($3.445m)	($6.194m)
equals Net Profit Before Tax	$3.071m	$5.471m	$3.897m
less Income Tax (Scenario Estimates @ 30%)	($0.921m)	($1.641m)	($1.220m)
equals Net Profit After Tax	$2.150m	$3.830m	$2.677m

The actual tax paid over FY2020 was at a rate of:

$1.220m / $3.897m= 31.3%

The actual result was somewhat in the middle of my higher and lower estimate. But digging deeper that AWF division contribution was a clangor. Far far worse than my worst fears. The fix for the 'Auckland Commercial Construction' problems was to avoid the construction industry (AR2020 p12)!

The white collar contribution was up. But that was mainly due to 10 months contribution from the newly acquired 'JacksonStone' 'C' level placement consultancy. Take out that and both Madison and AbsoluteIT were down (revenue down a combined 8%).

Head office expenses were up as "directors and management all took pay reductions" (AR2020 p12)! Wasn't FY2020 also the year that the company gained the full value of putting Madison and Absolute It into the same locations in Auckland and Wellington? It is hard to reconcile administration expenses going up in the circumstances I have outlined.

The interest bill went up because I didn't model the $10.520m purchase of 'JacksonStone'. 'JacksonStone' certainly contributed to increasing White Collar profit But the downside was a much higher interest bill and restraint of trade amortisations that mostly erased the acquired business's positive EBIT contribution. I can't help wondering if in the 'quest for growth' the eye was taken off the existing balls in hand. I am not saying the acquisition of 'JacksonStone' was necessarily wrong long term, albeit the acquisition timing perhaps could have been better. I am saying that increased debt generally doesn't go away. But increased profits can.

If we go back to the interim report p7, 'the Bennster' had high hopes for Net Profit Before tax:

"We expect to achieve NPBT (Net Profit Before Tax) for the full year of above $4.2 million – nearly 50% higher than prior year. An upper range is not provided due to some uncertainty in the level of improvement within AWF and the speed at which we can ramp up temp numbers within Madison."

This was all pre Covid-19 of course. But even so, to fall short of your lowest expectations by: $4.200m - $3.897m = $0.303m, is not a great result.

SNOOPY