Monday, 23 February 2015

Tungsten (TUNG) ... it's a small world part II

Monday 23rd February 2015

A principal reason why I have a bearish stance on Tungsten is down to the the forward curve for Bank of England (or for that matter that the FED or ECB) policy rate expectations.

Bank of England - policy rate expectations
Source: Bloomberg
I will expand upon this further below.

But first ... 

As highlighted in my prior post (it's a small world), Tungsten was floated on AIM in October 2013. It raised £160 million (£149 million net of costs) and acquired the e-Invoicing platform, OB10, also in October 2013. 

This was a large float for AIM. Indeed the largest since 2008, excluding investment companies.

OB10

At the time of acquisition, OB10 had been loss making for the three years prior. In fact, it had been loss making throughout its 14 year history, but the AIM Admission document only provides three years of financials of granular information. 
  • Revenue had been growing at a relatively pedestrian pace of 7-9% per annum, to £17.8 million in 2013. This was hardly indicative of OB10 gaining market share.
  • Pre-tax losses had increased from £3 million in 2011 to £3.5 million in 2013.
  • Approximately £50 million had already been invested in OB10 since it was formed in 2000 through to 2013.
  • Its share premium balance stood at £44.7 million as at April 2013.
  • Its accumulated losses totalled £52.3 million by that point also.
  • Net liabilities had risen from £2.3 million in 2011, to £7.5 million in 2013. 

Surprisingly for a company centred on the invoicing market, its own record of receivables management had deteriorated sharply. Trade receivables past due or impaired as a percentage of net trade receivables had risen from 33% in 2010, to 62% by 2013. 

It would appear to me that OB10 was showing few signs of becoming profitable any time soon. Rather, OB10's auditor seemed to believe that without the acquisition proceeds from Tungsten, that OB10 would not have been a going concern:
(b) Going concern 
This historical financial information relating to the Group has been prepared on the going concern basis, which assumes that the Group will continue to be able to meet its liabilities as they fall due for the foreseeable future. The use of the going concern basis relies on the receipt of the net proceeds from the offer of shares of Tungsten Corporation plc, the owner of the Group following the admission of Tungsten Corporation plc’s ordinary shares to AIM.
OB10 was acquired by Tungsten for £73 million in cash and £28 million in equity. 

Disruptive Capital

While OB10 had been plodding along but racking up losses for over a decade, Tungsten's development began a little more recently. Tungsten was founded in February 2012 by Edmund Truell and Danny Truell. Up until the date of IPO in October 2013, the well regarded founders had invested £9.6 million in the company. 

According to Tungsten's Admission Document, the company was also advised by Disruptive Capital. Disruptive Capital was exclusively engaged by the Tungsten Board to identify and recommend investment opportunities to Tungsten. 

As well as being the CEO of Tungsten, Edmund Truell is also the Chairman of Disruptive Capital. 

It's a small world. 

In the 14 or so months since Tungsten was founded in early February 2012 to its reporting period 30 April 2013, Tungsten's administrative expenses racked up as below: 
5. ADMINISTRATIVE EXPENSES  
Directors’ salaries and social security ____________  £591,866 
Transaction costs  __________________________   £3,340,065 
Advisory fees ________________________________ £333,358 
Office accommodation ________________________  £209,333 
Irrecoverable VAT ___________________________  £133,952
Post and stationery ____________________________  £67,414
Advertising __________________________________  £69,057
Auditors remuneration _________________________  £25,000
Other expenses ______________________________  £135,010 
Administrative expenses _____________________ £4,905,055
The transaction costs relate to the aborted IPO in 2012. Other Administrative costs are £5,040,000 in relation to the fair value benefit of the LTIP Incentives.
Although the founders had put an impressive £9.6 million into Tungsten, £591,866 had been taken out in salary related costs, and a further £333,358 had been paid in advisory fees. Given that Tungsten was being exclusively advised by Disruptive Capital, it's probably fair to assume that the bulk of the advisory fees winged their way to Edmund Truell's advisory firm.

And then there's that £3,340,065 cost related to an aborted IPO. It's not altogether clear why a first stab at IPO was aborted in 2012; although the markets did have a wobble mid way through the year so perhaps the answer lies there? Nonetheless, the aborted IPO carried a hefty cost.

Ex-ante to the aborted IPO, Disruptive Capital had seemingly expected to charge the company fees following flotation. However, as the IPO was cancelled, and ex-post there was no immediate prospect of Disruptive Capital getting paid on this part of the arrangement, a new mechanism was formed in order for this to be remedied.
17.2 Arrangements with Disruptive Capital 
A number of agreements between Company and the Subsidiary on the one hand and Disruptive Capital on the other which were entered into in May 2012 in the expectation of an admission to the main market of the London Stock Exchange. These included the ability to charge fees from the point of admission to the main market. As the admission did not take place, the Directors of the Company agreed with Disruptive Capital to engage them to provide origination, recommendation, negotiation and execution of potential investments on the basis of a charge of cost to Disruptive Capital. These arrangements were reviewed regularly by the Board and the costs approved before invoices settled. 
The Company will not continue these arrangements after the Admission and will only engage with Disruptive Capital if there are services the Board consider are appropriate to obtain from them. 
An aggregate of £1,293,666 has been paid to Disruptive Capital by the Company in the 12 month period to 9 October 2013, the latest practicable date prior to the publication of this document. An additional fee of £2 million (exclusive of VAT) is payable to Disruptive Capital on Admission. 
Fortunately for Disruptive Capital, it would seem that while Tungsten paid its advisors £333,358 in the period 2 February 2012 to 30 April 2013, and absorbed £3,340,065 in (aborted) transaction fees, it paid Disruptive Capital a total £1,293,666 in the period 10 October 2012 to 9 October 2013. Then Tungsten paid out an additional £2 million to Disruptive Capital after the second (and this time successful) go at IPO.

To be fair to the management, they put their hands in their pockets for the £12 million placing in September last. Rather oddly however, while most of the board members which took part in the placing received allocations of around 97% of their prior day indications of interest, poor old Michael Spencer got cut back to 50%. Perhaps he'd had a bad night? Or perhaps he'd already dipped his quill to begin his letter of resignation? He gave notice in late December last.   

The bank

The AIM Admission Document also outlined Tungsten's ambition to acquire a bank. And it bought the UK branch of First International Bank of Israel (FIBI Bank) on 10 June 2014.

Tungsten paid £29.5 million for FIBI Bank.

As highlighted below, FIBI's banking licence was valued at £3.3 million. It had £21.4 million in current assets (mainly composed of debt instruments and other investments), and had £1.9 million in liabilities (largely customer deposits).

Goodwill of £6.8 million was attributable to the expertise of FIBI's staff, customer service capability and future opportunities. Whether FIBI's staff were - prior to acquisition - fully versed in invoice discounting, I do not know. The structure of the asset base and details in the AIM Admission Document suggests that they weren't doing much of it, if any, and that the asset base comprised mainly short term UK gilts and certificates of deposit. Further, valuing future opportunities is a fairly subjective exercise. But there you have it.

Final fair values at the acquisition date of FIBI Bank
Source: Tungsten Plc
A subsequent £7.5 million was dolled out following the acquisition of FIBI Bank, as "investment in operations."

Tungsten's original aim for the bank was seemingly to deploy its own capital through invoice discounting. This capital was targeted to stem from corporate and institutional bank deposits, this allowing Tungsten a relatively low cost of capital.

While Tungsten viewed a banking licence as bringing a range of benefits, it would appear that it also acknowledged (in its AIM Admission Document), that a full UK banking licence is not necessary for invoice discounting and that a non-bank finance company structure could be used in its place.

I would reckon that there is no serious prospect of meaningful deposits until the company can demonstrate it can monetise its asset base in order to fund its liabilities, be they operational or deposit related. But I may be over cynical in this regard.

In the event, this appears to be ringing true, as Tungsten has been largely unsuccessful in raising either debt or bank deposits and has resorted to a third party provider of capital in the form of its arrangement with Insight.

And so it looks to me, that as well as seemingly overpaying for a loss making, cash absorbing, insolvent (according to its auditor) business, that Tungsten has also likely squandered - in the short term at least - a good £37 million in purchasing and brushing up a banking licence. Although it did get c. £20 million in short term gilts and other stuff.

But all this is very much in the rear view mirror. Bulls are focused on the big bucks ready to roll in.

And so with the bank in place and all set to go on invoice discounting, this brings us back to the beginning of this post and policy rate expectations ...

Rates

According to Tungsten, their cost of funding from Insight is similar to the cost of funding its own bank is presented with. That being somewhere in the order of 3%.

A 3% cost of capital is worth bearing in mind when observing the Royal Bank of Scotland's invoice finance pricing outline:
Royal Bank of Scotland - Invoice Financing Pricing Schedule
Source: Royal Bank of Scotland
A new client is probably looking at an average cost for invoice discounting of:

  • + 1% arrangement (one off) 
  • + 1% service charge
  • + 0.9% (12 month LIBOR) +2.75% (mid-point)
This comes to 5.65%. 

Further ongoing invoice discount requests may see this drop to 4.65% as due diligence on the underlying client has already been performed/charged.

Tungsten's cost of capital is indicated to be in the order of 3%. I suspect it's a little higher than that and has been rounded down, but let's assume it's 3%.

I reckon Tungsten's average net interest margin is therefore likely to be something like:

5.65%
less 3%
= 2.65% on new clients.

And that Tungsten's average net interest margin is therefore likely to be:

4.65%
less 3%
= 1.65% on recurring clients.

Let's assume an average of 2.15%. I suspect this is generous but let's assume it.

That would suggest in order to obtain a cool £10 million in net invoicing fees to Tungsten, it needs to provide invoice discount finance to £465 million of invoices.  

Is that likely?

Bibby Financial Services claims to be the UK's leading independent invoice finance specialist. It reckons it has advanced approximately £388 million in invoice finance (both discounting and factoring) last year.
Bibby Financial Services is the UK's leading independent invoice finance specialist and a trusted provider of cashflow funding solutions to 7,000 businesses, handling annual client turnover of £4.9 billion and advancing in the region of £388 million. 

Or another way

In the AIM Admission Document, Tungsten highlights:
Euro Banking Association and Bank of England figures suggest that invoice discounting penetration rates in the UK are between 9 per cent. and 10 per cent. Suppliers to buyers on the OB10 network would typically experience an invoicing period of 45 to 60 days. On this basis, the Company has assumed that the book of discounted invoices would turn over five to six times a year. Assuming discount rates at the lower end of current market levels, this would give an average annualised discount rate of the order of between 8 and 10 per cent. per annum. As an illustrative example, assuming the UK average invoice discounting penetration rate of between 9 per cent. and 10 per cent. applied to OB10’s £19 billion UK invoice value for the financial year ended April 2013, this would equate to £1.8 billion of invoice financing being advanced over a year, with approximately £325 million advanced at any one time.
I reckon this is somewhat misleading.

Firstly, although invoice discounting penetration rates may well be 9-10% in the UK (Bibby's data above suggests 8%), it is unlikely that those 9-10% of the suppliers are seeking invoice discounting five to six times a year.

Rather, they may be doing it just once or maybe twice a year, when cash is tight. Otherwise, if utilising invoice discounting every 45 to 50 days, then as highlighted by the RBS fee structure above, this is costing them around 6% of revenue. A lot of suppliers are lucky to make 6% margins!

It might also lead some to think that as penetration is 9-10%, that Tungsten will automatically scoop this rate up on invoicing that passes across its network. Yet there are many, many invoice discount providers to choose from. In fact there are loads. The likely choice will be the invoice discount provider the supplier is already tied to/signed up/banking with (saving on any sign up cost, such as that 1% for due diligence).

I reckon it's going to take Tungsten a long time to even come close to convincing as much as 25% of those OB10 UK based suppliers that use invoice discounting to switch to Tungsten. Especially the likes of FTSE 350 suppliers which will draw the big bucks, but may use it relatively infrequently anyway and when they do prefer to use their trusted, lower cost bank.  

And applying this to Tungsten's UK example means 2.5% penetration of OB10's £19 billion UK invoice value, or £475 million of invoice financing being advanced over a year. I.e. probably £10 million in net fees.

Of course, Tungsten has already advanced £10 million in financing for the two months ended 31 December 2014, so there's £215,000 towards the haul.

Modest rate rises

I reckon Tungsten's original strategy centred on low fixed capital costs drawn from institutional and corporate deposits and/or debt. And that they would make considerable returns against the backdrop of a rising interest rate environment, seeing the capital cost rise but their invoice discount rates rise rapidly higher.

And so this picture of modest rate rises over the next few years and protracted low rates thereafter probably presents a dilemma to what looks to me like pie in the sky market forecasts.

Bank of England - policy rate expectations
Source: Bloomberg
I remain short Tungsten.

Disclaimer: The information, discussions or topics referred to on this blog should in no way be considered “advice” to buy or sell anything. The information which may be referred to is freely available in the public domain and where required the source of information is referenced to for verification. While every effort has been made to ensure the veracity of any information contained within this blog, the author accepts no responsibility for the accuracy of any information contained within this blog or for the sources of information which may be referred to. Readers are responsible for their own actions and interpretation of the information contained within this blog. 

3 comments:

  1. Interesting commentary.
    One observation as an FD and experienced user of invoice discounting facilities.
    The 6% cost you quote is an annual cost not a transaction cost. Typically an invoice discount facility is pretty permanent in that the invoices acting as collateral are revolving - as old invoices are paid new ones are being generated. So for a client with customers paying on average every 2 months the cost is 1% of revenue. When SMEs are struggling to get either risk capital or unsecured bank loans this is relatively cheap and flexible finance. I am long on Tungsten because I believe the product is needed but your other concerns may be right and must say I'm not totally comfortable right now. The company needs to make an announcement in the light of todays sharp share price fall. Martin C

    ReplyDelete
  2. A fuller response

    https://longtailresearch.wordpress.com/2015/02/25/lord-i-will-hold-my-tung/

    ReplyDelete
    Replies
    1. Sah36uk,

      I have probably covered most of your comments in a follow up post. However, with regards to your points 3 (company links to Disruptive Capital) and 4 (buying a bank when they didn't need to), I would say:

      I have not suggested that the to-ing and fro-ing between Edmund Truell, Tungsten and Disruptive Capital was secretive in any way. It is clearly marked in the documentation. It’s up to you if you do not care about that. I find it noteworthy. And other aspects I have not touched upon.

      The buying of the First International Bank of Israel (UK), was pretty silly in my view. It suggests to me that the strategy was flawed. Firstly, I reckon it should be obvious that if you are a cash burning company, which could present balance sheet concerns, and that there is no certainty on monetizing your asset base, or your existing operation (OB10), then you are going to have a hard time drawing bank deposits or raising debt from corporates and institutions.
      Surely this was obvious from the start? Or why not secure some form of guarantee of deposits or debt on the basis of acquiring a bank before buying the bank in the first place?
      I’m pretty sure c. £20 million of gilts and other investments are worth c. £20 million. Yes. I’m pretty sure of that. Why do you suppose they may be more valuable?
      Is the cost of capital 3.5%? I had assumed it was 3%.

      Best
      Matt

      Delete