DIF III - Global Co-Investment Fund L.P v DIF Capital Partners Limited

Coinmach and the Coinmach Transaction

1. At about the same time as the PPM was issued, BBLP was approached by Deutsche Bank, one of Coinmach's financial advisors, concerning the possibility of acquiring Coinmach. To consider this possible acquisition, the Coinmach Deal Team (the Deal Team) comprising Ms Berry Talintyre, Mr Jake Haines, Ms Sridhara Ramachandran and Mr Justin Levi was formed. Ms Ramachandran was an employee of BBLP, whilst the other three members of the Deal Team were seconded to BBLP from another BBLP subsidiary. The Deal Team was supervised by Mr Topfer. As outlined at [27] above, the potential acquisition was given the code name "Project Spin".
2. As has been noted at [4] above, Coinmach was the leading provider of outsourced laundry equipment services for multi-family or multi-dwelling housing properties in the United States.
3. There were a number of dimensions to Coinmach's business. By far the most important of these was its core "Route" business which involved leasing laundry rooms from building owners and property management companies, installing and servicing laundry equipment and collecting revenues generated from the use of laundry machines by building occupants. As was subsequently stated in an information memorandum prepared by the Deal Team (the Information Memorandum), the Route business was "characteri[s]ed by long-term customer contracts, low net customer attrition and consistent inflation-related price increases that generate stable and predictable cash flows that are largely insulated from economic cycles".
4. The Route business was identified in the Information Memorandum as having provided approximately 88.7% of Coinmach's 2007 financial year revenue. It was noted that the Route business was resilient to economic cycles, as was demonstrated by "stable revenues throughout recent periods of heightened vacancy in multi-family housing and a proven ability to successfully increase prices through the vacancy cycle".
5. The revenue derived from the Route business was a function of a number of variables, including the number of machines in the market, the proximity of competition to the multi-family or multi-dwelling buildings from which the Route business operated, the level of occupancy or vacancy rates in each such building, their amenity including the extent of the provision of shared laundry facilities, their geographic locality, the number of laundry loads per machine and the average price, described as the "vend price", per load of laundry. Some of these variables and, in particular, occupancy or vacancy rates of multi-family or multi-dwelling buildings, were in turn the function of broader macro-economic factors such as interest rates, the sub-prime mortgage market and tolerance for mortgage delinquencies.
6. As will emerge, the interaction of these integers or variables is of central relevance to the outcome of this appeal. For present purposes, it suffices to note that the number of laundry loads processed in Coinmach's Route business was not simply a function of the price of those loads. It appears to have been far more complex than that.
7. In respect of loads per machine, the Information Memorandum noted that: "Loads per Machine: vacancy rates (as measured by a machine-weighted regional average rate) reached historic highs in 2004 as interest rates were low, the sub-prime and 'creative mortgage' market expanded, and new construction was booming - Estimated annual decline in loads per machine of 3% from 2001 to 2007; - Larger average machine size also contributed to the decline in loads; and, - Loads per machine should improve as vacancy rates are expected to decline from current levels (which may be further supported in low 'B' and 'C' grade buildings by sub-prime defaults and tightening credit)."
8. In relation to vend prices, the Information Memorandum noted that: "Vend Prices: average vend price per load has increased from $0.93 in 2001 to $1.15 in 2007 (representing a 3.6% [Compound Annual Growth Rate]): - Coinmach typically has the ability to increase prices without landlord consent and does so on a regular basis (subject to monitoring the impact on loads per machine and machine profitability); - Ability to raise prices and maintain machine volumes is dependent on the pricing of substitutes such as laundromats and drop-off services which industry sources suggest typically increase prices in line with inflation; and, - In addition, the increasing use of card-payment machines (vs. physical coin payment machines) allows for additional flexibility to increase prices at 'non-quarter' increments."
9. Under the "Key Risks/Issues and Mitigants" section of the Information Memorandum, the following entries are of significance: "5. Vacancy/occupancy rates Increases in vacancy rates do result in reduced loads per machines and hence are a risk to the Coinmach business. However, the ability to increase vend prices has been used effectively in the past by Coinmach to address the significantly higher than average vacancy rates experienced over the last few years. Vacancy rates have reduced and the current sub-prime market upheaval and general tightening of credit is expected to only improve occupancy rates. In addition, the successful implementation of several of Coinmach's current business initiatives will serve to diversify the route business base and lessen the impact of any increase in vacancy rates. 6. Vend prices The ability to increase vend prices is typically only limited by the potential competitive threat of accessing a laundromat/drop-off service instead as landlords generally are supportive of any increase via the inclusion of regular price rises in the underlying lease contract and/or the joint commission/sharing arrangement whereby the landlord also financially benefits from such increases. Coinmach is well aware of the competitive threat posed by laundromats and sets its prices accordingly. The company has a strong history of achieving good vend price increases (3.6% CAGR over last 6 years) including in times of higher vacancy rates such as the last couple of years."
10. A preliminary non-binding indication of interest in acquiring Coinmach was submitted by BBLP to Deutsche Bank on 21 March 2007.
11. On 2 April 2007, the Deal Team sought an upcoming capital approval to incur up to US$1,000,000 in due diligence expenses following BBLP's shortlisting to provide a binding bid to acquire Coinmach. In its memorandum in support of this expenditure request, which was addressed to Mr Topfer, the Deal Team identified some six reasons why it believed that Coinmach was an attractive acquisition. These were: "(i) Leading provider in 'essential service' market (by a factor of 3 times) with barriers to entry including economies of scale (leads to cheaper sourcing of equipment and more efficient 'routes'), existing logistics system, upfront capex, and long term contracts and relationships with customers including alignment of incentives. Opportunities to grow both organically and via acquisition including utilizing existing core logistics system and service ability to satisfy other customer needs eg ATMs; (ii) Operations which are relatively resistant to changes in economic cycles, geographically diverse and with no customer concentration; (iii) Experienced management team with a proven track record of managing the core business, integrating acquisitions and developing new business. The CEO has indicated initial willingness to stay for a further 4-5 years; (iv) Ability to earn an attractive forecast IRR for 'infrastructure style' cash flows. Upside through major synergistic acquisitions, higher leverage and exit into lower return 'boring business' fund; (v) Complementary to the developing US Real Estate multi-family housing platform. The BNP REIT recently purchased by the US Real Estate team, for example, is a user of the Company's services and has provided initial positive feedback as to the Company and its business model; and, (vi) The opportunity to earn sizeable upfront and on-going fees for B&B (and B&B funds where relevant) and raise the profile of B&B and the new North American Corporate Finance team within the North American market. Successful completion of the Transaction could also provide the proposed DIF III fund with a seed asset."
12. In this memorandum, DIF III was identified as a potential source of US$50,000,000 in respect of the transactions.
13. During April 2007, the Deal Team prepared the Information Memorandum, to which reference has already been made, for delivery to a limited number of institutional and other sophisticated investors on a confidential basis, solely for their use in considering whether to purchase the debt instruments described in the memorandum.
14. The Information Memorandum provided a set of projected financials, described as the "base case", for a five year period. The Deal Team had constructed a financial model by reference to which the investment in Coinmach could be assessed and projections computed based on a range of assumptions.
15. Early versions of the financial model constructed by the Deal Team in April 2007 were not in evidence before the primary judge, but other documents, such as the Information Memorandum together with a report in relation to the transaction by KPMG dated 27 April 2007 (the KPMG report) indicated that, at that time at least, the financial model assumed a 2% increase in vend pricing per annum.
16. A detailed email of 14 April 2007 from Mr Haines to Ms Ramachandran (with her response integrated into the email) sheds some light on the process by which the model was developed, and the assumptions it contained were questioned and debated by the Deal Team. One aspect of this email which is of note is the following observation by Mr Haines: "One missing piece in the IM and our industry analysis is some real info on laundromats, dry cleaners and other substitutes (i.e drop off services). Can you please try to get some industry info on laundromats such as historic growth, recent trends, average pricing etc (all the obvious stuff in figuring out whether there is a structural shift to laundromats and whether our historic price increases are consistent with that of our key substitutes - something I think needs to be true to get comfortable with 2%+ growth). There may be a public laundromat company or the ratings or broker reports for DRA or TUC may have some info."
17. This observation picks up on those features of the Information Memorandum which have been highlighted above.
18. On 15 April 2007, Ms Talintyre reported to Mr Topfer as follows: "Hi Rob - made progress on 'base case' (still subject to final dd etc but making good progress): (i) Forecast IRR of 21.9% post fees to B&B (upfront fee of 2% of EV (excl costs) & 1% management fee on total equity amount) - assumes $14 price which we think need be at to win (ii) Upside to forecast IRR through better margin on debt (hope to get but leaving bit in in case of flex), lower make whole amount for 11% notes than allowed (again expected but being conservative), lower upfront to banks (think should get 25bp cheaper), exit at higher multiple/lower return than pay (assumed constant multiple at this stage), several specific business plan initiatives which they are currently working on but we have excluded from base case in interest of conservatism, acquisition of second largest competitor, JV with third largest competitor, re-fi later on etc-> should be able make a case for BBGP & hopefully avoid any side deal Current financing plan requires USD348m of equity, USD40m of junior notes (priced at straight 10% coupon & interest only - should get senior second lien notes at 9.5% (10% between ARG and parking)- plus will offer upfront fee of 1.75% (havent got in past but EBB now want)), USD600m of senior notes (aim margin 2.25 - 1 % amort) and USD460m of senior second lien notes (no amort- 9.5% coupon)-> total funds required= US1.448 bn Equity proposal - [100]m B&B, [100]m EBB and BBGP/BBGP cos combined, B&B bridge DIF [ 40m ] (DIF 1 not have $ due to Coogee Resources - in reality as long as DIF III get up and not delayed & agree to deal (initial like but Harry can't focus on until mid May), B&B $ probab wont be required or if are only for a short period of time - also B&B will get $28m fee on close) - so need [108]m … PS: we had a great discussion with BNP REIT CEO on Friday (Multi-family housing REIT purchase recently finalised by US RE guys) - they use Coinmach as a supplier and recently awarded them another contract over Mac-Gray (2nd largest competitor) - were very positive re Coinmach as a company (good service, good people etc), lowering of vacancy rates, potential opportunities moving forward as a group if secure etc - going to get him to speak to Coinmach CEO/Chairman as well and put in good word re B&B".
19. A number of points should be made about this email.
20. First, as at 15 April 2007, the forecast IRR was 21.9%.
21. Second, the base case by reference to which this IRR was calculated excluded a number of specific business plan initiatives that the current management of Coinmach was looking at which, if brought to fruition, carried a potential upside.
22. Third, the reference to EBB is a reference to Everest Babcock & Brown which was a separate fund which the Deal Team was targeting to participate in the transaction.
23. Fourth, the statement "as long as DIF III get up and not delayed & agree to deal" is a reference to the fact that DIF III had not yet, at that stage, been established although, as has been noted, the PPM had been issued.
24. Fifth, the reference to "Harry" in this email is plainly a reference to Mr Harry Nicholson, one of the Manager's two Investment Officers, as outlined at [41] above.
25. In the KPMG report to which reference has been made at [59] above, KPMG stated that management was "confident that it will be able to continue to raise vendor prices in the near future." The report also noted that management had indicated that "its pricing policy varies by geographic area and that price increases largely depend, among other micro-economic factors, on the pricing developments of nearby laundromats." The report also noted that it "is not uncommon that price increases are rolled back if the price increase has a negative impact on the revenue obtained per machine".
26. Implicit in this statement is that that would not invariably be the case and that revenue could be affected by other micro-economic factors, such as proximate competition. To this factor, one could further add vacancy rates in multi-dwelling units which may in turn be a function of larger macro-economic considerations such as mortgage rates. The KPMG report also recorded that there was a significant variance in price levels charged by Coinmach across different regions.
27. On 2 May 2007, the Deal Team prepared an internal Capital Approval Request (CAR) seeking approval for US$85,000,000 of equity funding from BBLP, subject to BBLP being selected as the preferred bidder to acquire Coinmach and ultimately securing a positive vote from shareholders so as to facilitate a 100% acquisition, satisfactory documentation, satisfactory confirmatory due diligence and the implementation of satisfactory management equity/incentive arrangements. The CAR replicated much of the information that was contained in the Information Memorandum to which reference has already been made. Of particular note, the CAR identified the likely fees and returns on the equity investment as being approximately 22% post company level tax and pre-management fee. The pre-management fee was identified as a 2% fee.
28. This forecast base case return on equity investment was said to be based upon an exit at a multiple equal to the entry multiple in approximately four years' time, and an achievement of the underlying base case financial projections.
29. On 3 May 2007, Mr Haines received an email from Mr Will Peterson, Portfolio Manager of the Everest Babcock & Brown Income Fund, which had been identified in Ms Talintyre's 15 April 2007 email, extracted at [62] above, as a potential equity investor. Mr Peterson's email was as follows: "…I have one additional question in relation to the base case assumptions for load rates on the core business. I am not sure if I am looking at this correctly and your clarification is much appreciated. The historical growth in loads per machine has been -3.2% in 2002, -4.5% in 2003, -3.5% in 2004, -1.8% in 2005, -2.8% in 2006 and -2.7% in 2007E. This is as per the financial model which you have provided. The forecasts assume that there is ZERO growth in loads per machine for each year in model. Whilst I understand that loads per machine are correlated to vacancy rates and that vacancy rates have declined in recent years and the trend is expected to follow I do not understand how do you get comfortable with that the trend of declines is corrected immediate next year and there afterwards. Is there some evidence of this coming through over the last 6 to 12 months when vacancies have started to decline? I ran a downside scenario whereby I assumed that growth in loads were -1.3% in 2008 and -1% in 2009 and ZERO thereafter. It appears that this had the impact of reducing the Base Case IRR to 18%. I am not sure whether this is a more realistic assumption as this might impact our investment decision. I would appreciate your view on this as the model appears to be highly sensitive to load growth. I would also appreciate as to how you went your capital approvals for the submission of your bid next Monday".
30. Mr Haines responded on the same day as follows: "…Loads per machine are driven by two primary factors: - vacancy rates (# of people in the buildings) and vend prices (in-building prices relative to substitutes such as the local laundromat). The trends you are observing reflect a combination of both (to different degrees) over time. The initial period declines are due to increases in vacancy rates while the decline in later periods (particularly 2006 and 2007) are driven by price increases (5.4% price increase in each of the last two years). We did a lot of analysis and examined the correlation of price increases to vacancy-adjusted loads per machine (i.e. nomalising for the impact of any vacancy moves) and discovered that at price increases about 2.5 - 3.0% one could expect a fairly linear reduction in load volumes. That said, at price increase below 2.5%, the impact on load volumes was fairly minimal. The forecasts in the model take into consideration the interplay between price and load volumes. That is, we have assumed 2% price growth per annum which we feel very comfortable reconciles to stable load volumes. If we were to assume higher price growth (say, 4%) we would have to incorporate an assumption about load declines (which in my view would be ~1% in that instance). Vacancy rates are still well above the long-term average and are expected to decline over the next 5 years. We also believe vacancy rates will decline further within the low 'B' and 'C' grade building segments as a result of subprime defaults, tightening credit etc which will benefit Coinmach. We have NOT built in any growth as a result of improvements to vacancy rates in our base case. I hope that helps. Let me know if any other questions arise or you would like to discuss this further". (emphasis added).
31. Following this exchange, Mr Peterson prepared a report for the EBB Income Fund investment committee. Its relevance for present purposes lies in the following observations which were no doubt informed by Mr Haines' email referred to immediately above: "Increase in Vacancy rates and its impact on Load rates per machine Loads per machine are driven by two primary factors: vacancy rates (# of people in the buildings) and vend prices (in-building prices relative to substitutes such as the local laundromat). However the ability to increase vend prices has been used effectively in the pass [sic] to address the significantly higher than average vacancy rates experienced over the last few years. Vacancy rates did peak in 2004 and have started to decline marginally over the last 2 years however the number of loads per machine is still declining each year but at a slower rate. The financial model assumes nil growth in the number of loads per machine compared to an average 3% p.a decline over the last 5 years. BNB has explained this trend due to a combination of both a change in vacancy rates and a change in vend prices (to different degrees) over time. The initial period declines are due to increases in vacancy rates while the decline in later periods (particularly 2006 and 2007) are driven by price increases (5.4% price increase in each of the last two years). BNB has stated there [sic] analysis indicated that at price increases above 2.5 - 3.0% one could expect a fairly linear reduction in load volumes. That said, at price increase below 2.5%, the impact on load volumes was fairly minimal. The forecasts in the model take into consideration the interplay between price and load volumes. That is, it assumes 2% price growth per annum which may result in stable load volumes however there is no evidence of this yet. Vacancy rates are still well above the long-term average and are expected to decline over the next 5 years. We also believe vacancy rates will decline further within the low 'B' and 'C' grade building segments as a result of subprime defaults, tightening credit etc which will benefit Coinmach. BNB has NOT built in any growth as a result of improvements to vacancy rates in the base case. We have run a scenario whereby loads per machine are declining at 1% p.a. over the remaining life of the transaction. This reduces the IRR (post BNB Fee pre-tax) to 16.0% p.a compared with the Base Case IRR of 22%. Vending Prices The ability to increase Vend prices is typically only limited by the potential competitive threat of accessing a Laundromat/drop-off service instead as landlords generally are supportive of any increase via the inclusion of regular price rises in the underlying lease contract and/or the joint commission/sharing arrangement whereby the landlord also financially benefits from such increases. The company has a strong history of achieving good vend price increases (3.6% CARG over the last 6 years) including times of higher vacancy rates such as the last couple of years."
32. On 3 May 2007, a document headed CAR Supplement was prepared by the Deal Team to be read in conjunction with the CAR of 2 May 2007, referred to at [71] above. This document contained a series of slides addressing the communal laundry market profile, historical performance and outlook. The summary slide was in the following terms (with emphasis added): "A Market Profile • The addressable market in the US for communal laundry facilities comprises ~16m apartments which can be segmented by 'grade' based on rent level and amenities and provides insight into resident laundry usage • The majority of laundry facilities are located within lower grade buildings (70% of the rental apartment market) which are considered low risk conversion to in-unit machines given structural and economic factors • Over 90% of Coinmach's machines are installed in low-middle income areas where many residents rely on communal laundry facilities • New apartments increasingly include in-unit machines (particularly 'A' and 'B' grade buildings) but new construction only represents 1.1% of current supply and the effect on Coinmach's addressable market is negligible • Assuming no increase in market share (a downside case given Coinmach's financial strength relative to smaller operators) new construction is forecast to impact the machine base by less than 0.3% per annum B Historical Performance • The historic performance of the business has been stable across a challenging period of elevated vacancy rates • Net attrition in the route machine base has remained flat after adjusting for non-recurring/abnormal losses • Loads per machine have declined 7.6% since 2004 which was largely driven by increasing vacancy and price increases • Average vend prices have increased by 12.6% since 2004 (5.4% in 2006 and 5.4% in 2007) • Discussions with management and further analysis suggests price increases can continue at levels below 3% per annum without impacting load volumes C Outlook • Vacancy rates are above long-term average levels and are forecast to decline over the next 5 years • Coinmach's leading market position and developed service platform support profitable above-market growth through further industry consolidation and increased outsourcing of both communal laundry facilities and in-unit rentals • Coinmach's lease run-off profile is relatively uniform which provides stability".
33. In relation to "historical performance", each of the slides contained a headline proposition accompanied by illustrative charts and graphs purporting to support the respective headline propositions. The headline propositions were as follows:

• the performance of the business has been stable across a challenging period of elevated vacancy rates, supported by consistent price increases;
• the machine base has remained relatively flat after adjusting for abnormal and non-recurring items;
• while vacancy rates have declined over the last couple of years, the effect on load values has been offset by aggressive vend price increases;
• however, vend price increases have been successfully effected with a less than commensurate decline in loads per machine;
• historical data suggests load volumes are insulated from price increases of less than 3% but are negatively impacted by price increases above those levels. (emphasis in original).

1. The last of these slides included the following graph: text version of graph (5085241, rtf)
2. The CAR and Supplement were circulated/distributed on or about 4 May 2007.
3. Final bids were lodged on 7 May 2007 and Babcock & Brown was subsequently appointed preferred bidder. A merger agreement was executed on 15 June 2007 with closing, following completion of due diligence, then estimated to take approximately 4 months.
4. By June 2007, the Deal Team had progressed with its due diligence and produced a set of slides entitled "Project Spin Summary Materials" (the Summary Materials). These were sent to Mr Neilson and Mr Nicholson by the Coinmach Deal Team on 6 July 2007.
5. The Executive Summary included the following two bullet points: "• Commercial diligence undertaken through a series of formal management meetings and site visits, access to electronic data room materials, a detailed Q&A process, and extensive analysis; • Legal, accounting and tax due diligence performed by external advisors".
6. Under the heading "Investment Thesis", the Summary Materials noted that Coinmach had been "successful in consistently raising prices during a period of heightened vacancy rates". Under the section "Communal Laundry Market Profile", Coinmach's historical performance and projections were described as follows: "• Coinmach has successfully managed through a very challenging period of high vacancy rates and has achieved 3% EBITDA growth since 2004 • The projected performance of the existing business is assumed to be consistent with historic performance and observed trends • Projections assume 0% growth in the machine base which is consistent with normalized historical attrition (albeit management predict growth in the machine base this year) • Load volume and price growth assumptions are conservative and do not include the impact of declining vacancy rates • Above-market growth is driven by further consolidation and focus on leveraging the existing route platform and systems • Operations are increasingly looking to monetize as vacancy rates improve and succession issues need to be managed • Coinmach is well positioned as the logical 'consolidator' given its scale, geographic footprint, advanced systems, ability to realize addition synergies versus is competitors, and proven record of integration." (emphasis added).
7. Under the heading "Valuation" in the Summary Slide, the Deal Team stated that: "• Base case projections yield an IRR of approximately 21.3% which is underpinned by the stability and defensive nature of the core business • Upside returns in excess of 50% are possible with the execution of further industry consolidation, more aggressive sourcing and pricing, and multiple expansion such as that which could be achieved through sell-down into a proposed 'Essential Services' fund".
8. The summary conclusions set out in the preceding paragraphs were in turn supported by detailed slides which both graphed past performance and described various key assumptions in the model. Of particular significance was a slide entitled "Load volume and price increase assumptions also reconcile with history". This slide contained the following text: "A. The primary drivers of load volumes are vacancy and price. There does not appear to be a structural shift to substitutes which is having a material impact on load volumes. B. Price is a primary lever to increase profitability and price increases are managed at a local level based on the local machine profile, particular lease arrangements and average pricing relative to substitutes: • Currently, it is understood that the company's target is $1.25 (i.e. all machines under $1.25 vend price will be selectively increased over the course of the next 12-18 months) • Machine revenues are monitored after each price increase to examine the effect on loads and cash flow per machine C. Since 2002/03 the company's focus has been on managing cash flow per machine and price growth has been a key element of this strategy D. The commercial due diligence process has focused on the relationship between price and load volume declines: • Correlation between price and load volumes on a city-by-city basis broadly support a relationship whereby price increases of 3%+ negatively impact load volumes (see slide 19) • Price increases below 3% do not appear to have a material impact on load volumes (i.e. do not force substitution) • This is supported by the economics of laundromats which are typically single-facility operations that are exposed to cost inflation of 3%+ per annum (utilities, retail rent, higher capital costs due to less purchasing power etc.) E. The base case assumes price growth of 2% per annum and 0% growth in load per machine volumes as a means of addressing the observed relationship between price and loads: • It is expected that there is further upside to the volume forecast as a result of improvements in vacancy rates generally and, more specifically, with respect to low 'B' and 'C' grade buildings which will be positively exposed to the impact of increasing sub-prime mortgage defaults (see slide 20) • No impact from declining vacancy rates is assumed in the base case model (i.e. represents potential upside)" (emphasis in original).

1. In his oral submissions, Mr Jackman, having pointed out the negative impact on loads per machine in the context of price rises of 4.3%, 3.1% and 2.7% for 2002, 2003 and 2004 respectively, said in relation to the data for 2005: "Then the next year, 2005, there's a growth in price per load of 1.3%, but that has a -1.8% impact on loads per machine. Now that of course would have been sufficient when one goes back to that sensitivity chart to push the IRR well under 23%. That is, if you only assumed a growth in price per load of 1.3%. And then to round it out the increase in 2006 of 5.4% led to a reduction in loads per machine of 2.8%. So any increase in price one takes from that causes a drop in loads per machine, and looking at those five years of data, there's no justification for thinking that a two and a half per cent increase in price would not cause well over a 1% reduction in loads per machine. And on the sensitivity chart that destroys the prospect of this investment getting over a hurdle rate of 23%."
2. By way of contrast with this historical data for 2002-2006, in terms of projections for 10 years from 2008, the model as sent to Mr Nicholson on 26 October 2007 assumed a price increase of 3.6% for 2008, and then regular price rises of 2.5% for each year to 2017, but with no growth or diminution in the % impact on loads per machine.
3. Mr Jackman attacked the 0% figure attributed to growth in loads per machine as an unjustified assumption which he submitted, almost wholly by reference to the historical data for the previous five years referred to above, a competent and independent due diligence would have discerned.
4. In this context, he also pointed to a sensitivity table contained in the model which showed that, consistent with Mr Haines' email of 26 October 2007 (see [95] above), an assumed annual price rise of 2% rather than 2.5% with a 0% assumed impact load growth would have dropped the IRR from 24.6% to 22%. The same sensitivity table illustrated that an assumed annual price rise of 2.5% with a 1% drop in load volumes would have yielded IRR of 21.9%.
5. An extract from the sensitivity table is set out below: table text version (82418, rtf)
6. Given the way in which Mr Jackman put his argument, it is relevant to observe that the sensitivity table also showed that if one assumed a 2.5% annual price increase but a negative 0.5% load growth, the IRR would be 23.3%, which was above DIF III's targeted IRR.
7. On 1 November 2007, Mr Nicholson distributed a proposal to the Manager's Investment Committee to "co-invest with B&B in Coinmach", noting that the Manager's Compliance Committee approval had been obtained earlier in the week. In this document, Mr Nicholson noted "[t]he equity returns, on base case projections, target an IRR of 24.6%" and incorporated the sensitivity tables from the model which I have referred to above. The proposal also noted that "[t]he various value accretion projects, such as industry consolidation, could increase IRR to 50%+".
8. Mr Nicholson stated in the proposal that: "The base case is naturally most sensitive to changing the assumptions about the route business as it is the dominant part of the company. These sensitivities show, if we accept price per load growth has been set conservatively with respect to historical growth, and that load volume is protected on the downside due to decrease in vacancy rates following the sub-prime meltdown, that preventing machine attrition due to poor management or enhanced competitor activity is the most critical feature of the business."
9. He also noted that: "As noted above, the projections in the bid model do not include any net growth in the route machine base (excluding growth as a result of recently won contracts). Management believes this is a conservative assumption and targets 1% annual growth in the machine base. This would result in additional EBITDA of ~US$15m."
10. Mr Nicholson further noted that: "Increases in vacancy rates do result in reduced loads per machines and hence are a risk to the Coinmach business. However, the ability to increase vend prices has been used effectively in the past by Coinmach to address the significantly higher than average vacancy rates experienced over the last few years. Vacancy rates have reduced and the current sub-prime market upheaval and general tightening of credit is expected to only improve occupancy rates."
11. The following further extracts from Mr Nicholson's proposal are also of relevance to this appeal. First, he noted that: "Finally, the bid model assumes Coinmach acquires ~US$25m of acquisitions per annum. This level of activity is below the aggregate value of opportunities currently being investigated by the company, and is in line with 'business as usual'. In particular an aggressive acquisition strategy of industry consolidation could rapidly eclipse this target."
12. He later said, in language that plainly derived from the Key Risks section of the Information Memorandum, as follows: "5. Vacancy/occupancy rates Increases in vacancy rates do result in reduced loads per machines and hence are a risk to the Coinmach business. However, the ability to increase vend prices has been used effectively in the past by Coinmach to address the significantly higher than average vacancy rates experienced over the last few years. Vacancy rates have reduced and the current sub-prime market upheaval and general tightening of credit is expected to only improve occupancy rates. In addition, the successful implementation of several of Coinmach's current business initiatives will serve to diversify the route business base and lessen the impact of any increase in vacancy rates. 6. Vend prices The ability to increase vend prices is typically only limited by the potential competitive threat of accessing a laundromat/drop-off service instead as landlords generally are supportive of any increase via the inclusion of regular price rises in the underlying lease contract and/or the joint commission/sharing arrangement whereby the landlord also financially benefits from such increases. Coinmach is well aware of the competitive threat posed by laundromats and sets its prices accordingly. The company has a strong history of achieving good vend price increases (3.6% CAGR over last 6 years) including in times of higher vacancy rates such as the last couple of years".
13. By 6 November 2007, all members of the Investment Committee had approved the transaction.
14. On 7 November 2007, Coinmach's half yearly accounts for the six months ending 30 September 2007 were published. These were summarised by the primary judge (at [232]) as follows: "The results for the second quarter became public on 7 November 2007. Revenue decreased by approximately $0.8 million, or approximately one per cent, for the three-month period ended 30 September 2007, as compared to the prior year's corresponding period. Revenue decreased by approximately $3.0 million, or approximately one per cent, for the six-month period ended 30 September 2007 compared to the prior year's corresponding period. The main explanation for the decrease was an increase in vacancy rates in certain geographic areas."
15. On 8 November 2007, the Board of the General Partner was notified that the Investment Committee of the Manager had approved the transaction and recommended it to the Board.
16. On 9 November 2007, the Board of the General Partner (two of whose members, Messrs Topfer and Neilson, were also on the Investment Committee of the Manager) resolved that entry into the Coinmach transaction would be in DIF III's "long term commercial benefit and [in its] commercial interests".

Consideration

1. Under the heading "The focus of the appeal" (see [30]-[38] above), the gist of Mr Jackman's argument on appeal was set out. In summary, it involved the following steps: 1. DIF III targeted investments with an IRR of 23% or more (see [3] and [30]-[31] and [40] above); 2. as originally presented to Mr Nicholson by the Deal Team, the base case in the financial model only indicated a rate of return of 22% (see [64], [71] and [97] above); 3. Underpinning the projected 22% IRR was a key assumption that vend prices would increase by 2% for each year of the model's operation from 2009 (with an increase of 3.6% for 2008), with no negative impact on the average number of washing machine loads on account of this assumed price increase (see [59], [74] and [85] above); 4. in order to achieve an IRR exceeding 23%, the assumed price rises in the model were increased at Mr Nicholson's request from 2.0% to 2.5% per annum (see [95]-[96] above); 5. the increase in the annual price rise by 0.5% from that in the base case was not accompanied by any decrease in the load volumes built into the model, that is, the model continued to assume a nil impact on washing machine loads notwithstanding the (increased) assumed price rise; 6. historical data for Coinmach suggested that there should have been a corresponding decrease in load volumes, as this had historically been the case when vend prices were increased (see [98] above); 7. this historical data was available to Mr Nicholson, as it was contained in the updated financial model he was sent on 26 October 2007 (see [95] above); 8. had proper due diligence been undertaken, the assumption that a price increase of 2.0% or 2.5% would have no impact on load volumes would not have been accepted, and it should have been appreciated that there would be a negative impact on load volume which would, in turn, have driven the IRR down to less than 23%; 9. by reference to the primary judge's unchallenged finding set out in [265] of the judgment (see [29] above), the transaction would not, in the circumstances, have been one that DIF III would have entered into.
2. In oral address, Mr Jackman succinctly summarised this submission as follows: "So we take from the model itself, without going to anything else, if all one did was to analyse the model as a freestanding document, one would conclude that looking at the historical experience one could not justify the assumption of zero change in loads per machine if one was to increase prices 2.5% each year. One would be very concerned at the fact that there are substantial losses for the first three years and cash flow is going to be absolutely line ball on an annual basis…".
3. I did not understand Mr Jackman to rest his case on the two considerations set out in the final sentence of his submission extracted above. Rather, I understood his reference to, and arguments in respect of, those two matters to reinforce how "tight" the model and its assumptions were. The considerations of losses for the first three years and the tightness of cashflows were not suggested, either individually or collectively, to have demonstrated a want of competence on the part of the Manager.
4. In any event, on the question of losses and tightness of cash flow, the primary judge held (at [236]) as follows: "It is apparent that the strategy behind the investment was to buy Coinmach, to increase its size, largely through further acquisitions which were to be funded in part by two facilities of USD50 million each (said to be 'for capex/acquisitions') and the reinvestment of profits; and then to sell the expanded business after three and a half years or so, at a profit. The return was to come from the profit on sale, not from income earned from the business during the time that it was held by Babcock & Brown. It is not surprising that the model assumed in that case that the business would not generate any significant profits. The critical matter was whether Coinmach would generate the predicted revenue. If it did not, that would have an effect on the ability of Coinmach to make further acquisitions, and consequently an effect on the ability of Coinmach to achieve the growth which underpinned the expected IRR. It would not necessarily affect Coinmach's viability. The Investment Committee Memo recognised that there was a risk that Coinmach would not be able to generate the anticipated growth. It also pointed to the potential for up-side. The question for the Investment Committee was whether the benefits of the transaction justified the investment notwithstanding the risks. The profit was irrelevant to that analysis."
5. The principal focus of DIF III's argument was on the suggested negative correlation, not reflected in the financial model, between price increases and load volumes.
6. Mr Jackman's argument had a superficial force and attraction, but that attraction overlooked the fact that the relationship between price rises and load volumes was, on the materials before the Court, highly complex, non-linear and a function of, or influenced by, many factors including geography, the size of any price increase, the rate of inflation, the general state of the economy and, in particular, its effect on the vacancy rates in multi-dwelling housing in which Coinmach's Route business operated (see [39]-[53] and [70] above).
7. That the relationship between price rises and vends loads was non-linear may be illustrated by the table setting out historical data for the years 2002-2006, an extract of which is reproduced at [98] above, and upon which heavy reliance was placed in argument.
8. Whilst the price increase for each of the years 2002-2006 shown in the table produced negative growth in loads per machine (which was essentially Mr Jackman's argument), the table reveals that micro-economic factors other than just price must have influenced the figures for load growth. This is illustrated by the fact that, for example, there was a greater negative impact on growth in 2004 following a 2.7% rise in price, compared to 2006, when the price rise was double that amount.
9. Further and not insignificantly for Mr Jackman's argument, all the price rises shown in the table extracted at [98] above, other than that for 2005, exceeded 2.5%. In other words, given the assumption made in the model as to a 2.0% or 2.5% price increase, the only historical data evidencing a price increase of less than either of those two figures was that for a single year, being 2005. The commentary in the materials produced by or emanating from the Deal Team acknowledged that there would be, or there would likely be, drops in load volumes for price increases of greater than 2.5% (see, for example, at [78], [85] and [88] above). The forensic significance of the figures in the table reproduced at [98] above was therefore much diminished.
10. Next, I turn to the graph reproduced at [78] above which was headed "Historical data suggests load volumes are insulated from price increases of less than 3% but are negatively impacted by price increases above those levels", and which lent apparent support to the critical assumption in the financial model which was the subject of Mr Jackman's attack. It was submitted that this graph showed no data at all for a price increase less than 3%, with the consequence that "it doesn't give support for the first part of the headline that the data suggests load volumes are insulated from price increases of less than 3%. There's just no data at all on that." It may be that the headline would be substantiated if the linear regression line on the graph were extended to 3%, but there was no evidence presented to the Court that would allow that conclusion to be reached one way or the other.
11. As the primary judge observed at [241], "the conclusion stated in the heading [to the graph] was based on an analysis which went beyond the figures for 2005", that is, those figures from the table upon which Mr Jackman relied as to historical experience (see [98] and [123] above).
12. Furthermore, this graph at [78] also illustrated, amongst other matters, that there was no linear relationship between a price increase and a diminution of load volumes, and evidenced the significance of geography for the purposes of analysing the impact between price increases and load volumes. Thus, in at least the cases of New York and Louisville, the graph suggested an increase in load volume notwithstanding respective price increases of over 4.0% and 5.0% in those locations. The graph plotted average 2006 and 2007 "vend price changes" excluding Data Outliers, which were identified as Atlanta and Columbus.
13. It is also noteworthy that the Y axis on the graph was entitled "Vacancy Adjusted Loads Per Machine Growth". The reference to "Vacancy Adjusted" highlights that one very important integer or factor affecting the relationship between price increases and load volumes was the vacancy rates in multi-dwelling buildings. The higher the vacancy rate, the less occupants in a multi-dwelling building and therefore a correspondingly lower number of people to use and, inferentially, a lower use of washing machines in such buildings. The position would be converse with lower vacancy rates.
14. In this context, whether or not the assumed neutral impact of price rises on vend loads for the 10 years covered in the financial model's projections was valid or reasonable would be influenced, in part at least, by an assessment of the multi-dwelling housing market for that period, and the level of vacancy rates. There were a number of statements made in the due diligence materials predicting a fall in vacancy rates, because of the predicted collapse of the sub-prime lending market which was expected to drive people back to multi-dwelling buildings in which Coinmach's core business was centred (see, for example, [51], [53], [74], [76], [85], [106], [108] and [110] above).
15. Subject to one qualification, neither the primary judge nor this Court had any material to question the validity of these statements, or to model or assess the impact of the macro-economic judgement that there would be a fall in multi-dwelling vacancy rates on projected vend loads, notwithstanding a posited 2.0% or 2.5% price rise. The one qualification was the following statement in the September quarterly report for Coinmach issued on 7 November 2007, after the Investment Committee of the Manager had made its recommendation and very shortly before the General Partner made its decision to proceed. That statement was: "Route revenue for the three months ended September 30, 2007 decreased by approximately $1.1 million, or approximately 1%, over the prior year's corresponding period. Route revenue for the six months ended September 30, 2007 decreased by approximately $3.1 million, or approximately 1%, over the prior year's corresponding period. Such decrease was primarily due to a decline in same store sales attributable to increased vacancy rates in certain geographical areas." (emphasis added). A similar statement had appeared in the report for the previous quarter.
16. A number of points may be made.
17. First, the report makes clear that this reference was confined to "certain geographical areas". It was not across the board.
18. Second, the extent of the increase was not identified, other than that the statement was made in the context of a very small relative drop in revenue.
19. Third, the statements in the due diligence materials referred to above in relation to decreases in vacancy rates were mid to longer term projections. As Mr Peterson of EBB said in his report for the EBB Income Fund Investment Committee, referred to at [75] above: "Vacancy rates are still well above the long-term average and are expected to decline over the next 5 years. We also believe vacancy rates will decline further within the low 'B' and 'C' grade building segments as a result of subprime defaults, tightening credit etc which will benefit Coinmach."
20. Fourth, vacancy rates were linked to the sub-prime market. In this context, Mr Jones, appearing for the PI Insurers, submitted that the period to which the reports related (being the quarters ending 30 June 2007 and 30 September 2007) were prior to the effects of the Global Financial Crisis and the collapse of credit markets impacting the market. In this respect, he noted that the primary judge recorded at [88] that the collapse of credit markets followed the collapse of Lehman Brothers in mid-September 2007. As such, it was suggested that there would or may have been some time lag in those economic events working themselves through the housing market. This was supported by the evidence of Mr Topfer.
21. It will be recalled that the primary judge was not satisfied that it had been established that the relevant assumptions in the financial model were invalid or unreasonable or otherwise shown to be wrong. In essence, his Honour held that DIF III had failed to discharge its burden of proving that competent due diligence would have demonstrated the invalidity or unreasonableness of such assumptions.
22. His Honour was evidently not satisfied that one could reason from the fact of the historical experience of price increases which, on one view, was relevantly confined to one year, namely 2005 (because in other years for which data was available, price increases exceeded 2.5%) to a conclusion that the projections based on price rises of 2.0% to 2.5% would have a downward impact on load volumes with a corresponding downward impact on IRR. His Honour observed that he was left "without full details of the analysis and expert evidence in relation to it" (at [230]). In this context, he was also cognisant of the fact that the Coinmach Deal Team had had discussions with Coinmach's management, and that "the model and the assumptions it contained were prepared on the basis of the information the team had obtained": at [229].
23. It is in this context that Mr Haines' email of 31 July 2007 to Messrs Flyer and Pratan (see [87] above), and his earlier statement to Mr Peterson around the analysis that had been done in relation to the impact of price rises on load volumes (see [74] above) assumed particular significance. It will be recalled that it was in his 31 July 2007 email, at [87] above, that Mr Haines referred to an excel workbook containing a "supplemental vend price and loads analysis" which he said was used "to assess the relationship between vend price increases and load declines and to get comfortable that increases of 2-3% were achievable with no material impact on load volumes."
24. Mr Jones placed considerable significance on this 31 July 2007 email and the workbook to which it referred and to which it attached. He submitted that the reference in this email to the Excel workbook was significant as, on the face of the email, the analysis contained in it was the source of the critical conclusion stated in the last sentence of the email, namely that: "We used this analysis to assess the relationship between vend price increases and load declines and to get comfortable that increases of 2-3% were achievable with no material impact on load volumes." The workbook attached to the email is some 3.15 megabytes in size. Mr Jones noted that "no submissions were made to the trial judge as to how [this] data … invalidated [Mr Haines'] conclusions". Mr Jones submitted that it "would require some expert assistance… at least to review the data." His point was that, prima facie at least, there was material and analytical data which was inconsistent with the case DIF III was putting, and that the burden of persuasion lay on it to displace the prima facie support for the validity of the key assumption in the financial model.
25. It was telling, in my opinion, that DIF III undertook no analysis, or at the very least, both at first instance and on appeal, presented no analysis of the data in the workbook attached to Mr Haines' email of 31 July 2007. If the conclusion in Mr Haines' email was justified by the workbook analysis, it would not, in my opinion, have been open to conclude that the assumption contained in the financial model as to the neutral impact on load volumes of price increases of 2.5% or less, by reference to which the Manager made its investment recommendation to the General Partner, was flawed or invalid. This is plainly the view which the primary judge took (see [136] above).
26. Whilst expert evidence may not have been required to establish causation (cf Fernandez v Tubemakers of Australia Ltd [1975] 2 NSWLR 190 at 197), the lack of analysis or engagement by DIF III with this material was even more conspicuous because, although DIF III did lead expert evidence in the form of an extensive report by a chartered accountant, Mr Angus Ross, that report did not interrogate the workbook or test Mr Haines' conclusion by reference to it. More fundamentally, it did not address the validity of the assumption that a price increase of 2.5% would not have a negative or downward impact on load volumes.
27. In Commercial Union Assurance Company of Australia v Ferrcom Pty Ltd (1991) 22 NSWLR 389 at 418 (Ferrcom), Handley JA observed that, consistent with Jones v Dunkel (1959) 101 CLR 298; [1959] HCA 8 (Jones v Dunkel), there is: "no reason why those principles should not apply when a party by failing to examine a witness in chief on some topic, indicates 'as the most natural inference that the party fears to do so'. This fear is then 'some evidence' that such examination in chief 'would have exposed facts unfavourable to the party'."
28. Although not argued in the present case and although not necessary to presently decide, an interesting question arises as to whether a Ferrcom inference may be drawn when an expert witness does not address a particular topic otherwise within their expertise. Whilst there are cases, including a decision of Handley JA, where such an inference has been drawn in relation to an expert (Ta Ho Ma Pty Ltd v Allen (1999) 47 NSWLR 1 at 4; [1999] NSWCA 202; see also Gordon Martin Pty Limited v State Rail Authority of New South Wales [2008] NSWSC 343 at [322]; Aristocrat Technologies Australia Pty Ltd v Global Gaming Supplies Pty Ltd [2009] FCA 1495 at [417]; cf Sigma Pharmaceuticals (Australia) Pty Ltd v Wyeth [2010] FCA 1211 at [462]) and cases where it was assumed that such an inference could be drawn but that it was not appropriate to do so in the circumstances (see, for example, Australian Securities and Investments Commission (ASIC) v Rich (2009) 236 FLR 1; [2009] NSWSC 1229 at [478]-[480]; Harris v Bellemore [2010] NSWSC 176 at [136]), the question has not, so far as I am aware, been considered at the level of principle.
29. It is sufficient to note that, in the present case, there was relevantly an absence of expert evidence on the subject of the reasonableness or validity of the assumption that was at the fulcrum of DIF III's case. Whether or not Mr Ross would have been appropriately qualified to provide that particular evidence is another question.
30. As with the primary judge, on a matter which was undoubtedly of real complexity, being affected as it was by the interplay of a range of factors and dependent upon an understanding of a particular industry and market in a foreign jurisdiction, this Court was essentially left to speculate as to: (a) whether or not the key assumption in the financial model that Mr Jackman attacked was invalid, unjustified or unreasonable; and (b) whether the exercise of due diligence by the Manager would have ascertained this, leading to a different analysis and consequently yielding a different recommendation (or no recommendation at all).
31. As Dixon CJ observed in Jones v Dunkel at 304, in an action for negligence, the action must fail unless evidence is offered "to the reasonable satisfaction of a judicial mind", supporting some positive inference implying negligence. In this context, competing inferences of equal degrees of probability are inadequate, and the choice between them must not be "a mere matter of conjecture": at 304-305. The court is not authorised "to choose between guesses … on the ground that one guess seems more likely than another or the others": at 305. The Chief Justice continued (at 305): "[t]he facts proved must form a reasonable basis for a definite conclusion affirmatively drawn of the truth of which the tribunal of fact may reasonably be satisfied." See also Kitto J at 305ff.
32. The inadequacy of conjecture in the context of satisfactory proof of causation was discussed by Spigelman CJ in Seltsam Pty Ltd v McGuiness; James Hardie & Coy Pty Ltd v McGuiness (2000) 49 NSWLR 262; [2000] NSWCA 29 at [84]-[86] (Seltsam) as follows: "It is often difficult to distinguish between permissible inference and conjecture. Characterisation of a reasoning process as one or the other occurs on a continuum in which there is no bright line division. Nevertheless, the distinction exists. Lord Macmillan in Jones v Great Western Railway Co (1930) 47 TLR 39, in the context of stating that a possibility that a negligent act caused injury was not enough, said (at 45): 'The dividing line between conjecture and inference is often a very difficult one to draw. A conjecture may be plausible but is of no legal value, for its essence is that it is a mere guess. An inference in the legal sense, on the other hand, is a deduction from the evidence, and if it is a reasonable deduction it may have validity as legal proof. The attribution of an occurrence to a cause is, I take it, always a matter of inference.' After referring to this passage, Sir Frederick Jordan in Carr v Baker (1936) 36 SR(NSW) 301 said (at 306): 'The existence of a fact may be inferred from other facts when those facts make it reasonably probable that it exists; if they go no further than to show that it is possible that it may exist, then its existence does not go beyond mere conjecture. Conjecture may range from the barely possible to the quite possible'."
33. The fact that, in the present case, it was possible that the assumption made in the financial model - that there would be a neutral impact on load volumes notwithstanding an annual price rise of 2.0% or 2.5% - was flawed, is insufficient to establish causation on the balance of probabilities: Seltsam at [80]-[81], citing St George Club Ltd v Hines (1961) 35 ALJR 106 at 107; Bonnington Castings Ltd v Wardlaw [1956] UKHL 1; [1956] AC 613; Tubemakers of Australia Ltd v Fernandez (1976) 50 ALJR 720.
34. For all of the above reasons, DIF III has failed to demonstrate that the primary judge erred in his conclusion that there had been a failure to establish that the investment in Coinmach would not have proceeded had the Manager exercised due diligence. The "prior question" that his Honour had identified in [314] of his judgment, extracted at [21] above, namely: "whether having conducted additional investigations, the Manager would have reached different conclusions or made different recommendations from those that it did", was not addressed by any evidence, as the primary judge pointed out, and this lacuna was not filled by pointing to historical data contained in the financial model itself. The subject matter was far more complex than that simple forensic technique, albeit skilfully deployed, on appeal, allowed.
35. DIF III's failure to demonstrate that the primary judge erred in finding that it had failed to establish loss or damage means that its appeal to this Court must be dismissed with costs.
36. MEAGHER JA: This judgment adopts the abbreviations and terms used in the judgment of Bell P. I have had the benefit of reading his Honour's judgment and agree with his reasons for concluding that the first issue identified at [15] should be resolved in favour of the Manager, a subsidiary of Babcock & Brown Ltd, and the PI Insurers. That issue arose under grounds 14 and 15 of DIF III's amended notice of appeal.
37. The PI Insurers undertook the defence of DIF III's claim against the Manager, and were also sued directly by DIF III pursuant to Civil Liability (Third Party Claims against Insurers) Act 2017 (NSW), s 4(1). The remaining two issues in the appeal concern whether the Manager would have been entitled to an indemnity for its alleged liability to DIF III.
38. The first of those issues, raised by ground 18 of the amended notice of appeal, is whether the primary judge erred in rejecting DIF III's contention that its claim made in the proceedings against the Manager was a third party Claim covered by the PI Policy.
39. The remaining issue, raised by ground 1 of the PI Insurers' amended notice of contention, is whether the "conflicts" exclusion in cl 7 of the policy would have been engaged. The application of that exclusion depends on whether any insured "liability" arises out of or is based upon or attributable to "directly or indirectly, any conflicts of interest arising out of, based upon, relating to or in connection with investment banking activities or any research report".
40. Although these remaining issues are not dispositive, I propose to deal with the first, but not the second which depends on a close analysis of the scenarios in which DIF III's claim might have succeeded. Those scenarios were not the subject of findings by the primary judge (J [362]).

Was DIF III's claim to be considered as first made during the policy period?

1. The relevant provisions of the underlying policy are extracted by the primary judge at J [319]-[327]. They are not repeated in these reasons, which instead summarise the effect of those provisions.
2. The alleged liability of the Manager is for breach of cl 3.1(g) of the Management Agreement, involving a failure to "exercise all due diligence and vigilance in carrying out its functions, powers and duties". In particular, it is alleged the Manager breached its duty in not conducting appropriate financial due diligence by undertaking its own modelling that challenged the growth assumptions made by the Coinmach Deal Team. Those assumptions resulted in the modelling of price per load growth in the base model being increased from 2% to 2.5% without any adjustment to the load volumes, raising the modelled IRR by 2.6% to 24.6%.
3. The policy indemnified against third party liability arising out of claims "first made during the Policy Period". That period was from 1 September 2008 to 1 September 2009. Any written or oral demand alleging an actionable breach of duty is a Claim under the policy. Clause 7 provided that a third party Claim "is considered to be made" when the insured's management, including directors, the managing director, and members of executive committees established by the board, become "aware of any fact, circumstance or event which could reasonably be anticipated to give rise to a Claim at any future time".
4. That clause also provided that it was a condition precedent to coverage in respect of any Claim that written notice be given to the insurers "at the earliest practical moment, but in any event within 30 days after the expiration" of the policy. Such a notice, whether of an actual or potential third party Claim, was required to include "full particulars of such actual or potential Claim, including the identity of the actual or potential claimants, the location where such actual or potential Claim has been or is likely to be made... the specific allegations made or anticipated to be made, and the facts, circumstances and events giving rise to such actual or potential Claim."
5. If such a written notice was given, any "subsequent legal proceedings for damages brought against the Assured... as a direct result of any matter or matters for which written notice has been given... whether such proceedings are brought during or after the expiration of the Policy Period, is considered to be a third party Claim first made" at the time the written notice of such matter or matters was first given.
6. DIF III commenced the underlying proceedings against the Manager in 2018. No notice, either of that claim or of any "fact, circumstance or event" which could reasonably be anticipated to give rise to a Claim under the policy, was notified to the PI Insurers by Babcock & Brown or the Manager before those proceedings were commenced.
7. It was contended that the pleaded claim was a third party Claim first made during the policy period by the following process of reasoning. During that period, the Manager, here relevantly Mr Nicholson, became aware of circumstances "which could reasonably be anticipated" to give rise to a claim at some future time. The Manager omitted to give written notice of that matter to the insurers within 30 days of the expiry of the policy. The proceedings, insofar as they include the claim for breach of cl 3.1(g), were brought in 2018 "as a direct result" of the matters of which notice would have been given, but for that omission. Had that notice been given the pleaded claim would have been "considered to be" a third party Claim first made at the time that written notice would have been given. Finally, the effect of Insurance Contracts Act 1984 (Cth), s 54 was that the PI Insurers could not "refuse to pay" the Manager's claim to an indemnity by reason only of its omission to give written notice of its having become aware of the relevant matters during the Policy Period. That possible outcome of the application of s 54 is not contested between the parties, on the basis that it accords with the decision in FAI General Insurance Co Ltd v Australian Hospital Care Pty Ltd (2001) 204 CLR 641; [2001] HCA 38.
8. As formulated by the primary judge, the relevant questions included whether the Manager first became aware of such circumstances during the Policy Period and whether the Claim as made was a "direct result" of those circumstances as they would have been notified (J [339]). The primary judge dealt with the first of these questions at J [343]-[348]. He concluded that the evidence did not establish that any director or member of the senior management of the Manager became aware of such circumstances during the policy period (J [344]-[347]). His Honour noted that neither Mr Topfer nor Mr Green gave evidence of any facts from which it might be concluded that they became aware of circumstances that could reasonably be anticipated to give rise to a claim; and that Mr Nicholson and Mr Neilson were not called to give any such evidence. The primary judge also found that none of the nine communications relied on (J [328]-[336]) suggested that DIF III might make any claim against the Manager or members of the Investment Committee (J [348]).
9. In this Court, DIF III relied on only three of those communications, being those which came to the attention of Mr Nicholson, but not on their face to the attention of any other member of the Investment Committee. That correspondence is dated 13 November 2008, 17 November 2008 and 5 February 2009, and summarised by the primary judge at J [329], [331] and [335] respectively. It is to be recalled that the transaction was concluded on 20 November 2007 in the context of the global financial crisis which proceeded between mid to late 2007 and early 2009.
10. The first of those communications, dated 13 November 2008, recorded views of Mr Stahel of BBGP concerning the "distressed" position of the Coinmach investment. BBGP had separately invested approximately USD 70 million through a wholly owned subsidiary. Mr Stahel's view was that the prospect of holding "the investment long term to realise value through EBITDA growth" was at that time "impossible". The email contains no suggestion that the funding or capital deficiencies which the company faced were not a consequence of the global financial crisis or that they, or the reason for their arising, should have been discovered or appreciated at some earlier time by the Manager or its Investment Committee. Nor does it suggest that the company's "distressed" position was in any way related to or a consequence of over-ambitious growth assumptions in the final version of the Coinmach revenue model. Finally, the email contains no suggestion that the Manager might be held responsible for third party losses if the investment ultimately failed.
11. The second communication is an email of Mr Stahel to Mr Nicholson dated 27 November 2008. That email refers to an earlier email exchange between Ms Talintyre and Mr Cumming of RBS in November 2007 in which, as Mr Cumming later recalled, Ms Talintyre "stated that the Coinmach deal should go ahead - although agreeing it being un-commercial - purely for the reason of collecting a substantial fee from BBGP and other satellite fund investors". From Mr Nicholson's perspective at that time, as disclosed in his response to Mr Stahel dated 1 December 2008, before taking the matter any further he was "interested in seeing the email trail".
12. The relationship between Babcock & Brown and RBS with respect to the Coinmach transaction is dealt with at some length by the primary judge. In June 2007, RBS agreed to take USD 136 million of equity in Coinmach with the intention of selling down that interest with the assistance of Babcock & Brown (J [58], [71]). Following the Lehman Brothers collapse in mid-September 2007, RBS sought to withdraw from that underwriting commitment (J [88], [94]). By early November 2007, RBS considered that the likely equity loss it would suffer in the event that it complied with its funding commitment was more than the break fee it would incur in not meeting that obligation (J [129]). In that context there were email exchanges between Ms Talintyre and Mr Cumming (J [132]-[134]). In particular, Ms Talintyre's email of 8 November 2007 pointed out that should RBS default on its funding commitment, and B&B not find a replacement equity provider, there would be significant expenses incurred, including a termination fee to Coinmach, B&B's loss of its own fees, and significant legal, accounting and related expenses, all of which might be claimed from RBS.
13. The evidence does not indicate whether Mr Nicholson received copies of the email exchanges referred to above, or whether there were any further email exchanges which might be those recalled by Mr Cumming in his conversation with Mr Stahel. In that uncertain state of affairs, Mr Stahel's report did not of itself provide a sufficient or sound basis for thinking that a claim of some kind might be brought against the Manager because of conduct of Ms Talintyre, who was the leader of the Coinmach Deal Team.
14. The final email is dated 5 February 2009 and attaches a memo from Mr Nicholson to the DIF Compliance Committee attaching a draft valuation of Coinmach as at 1 October 2008 undertaken by KPMG LLP. That valuation concludes that Coinmach's interest-bearing debt exceeded its enterprise value, with the consequence that the value of Coinmach's equity was nil. In the memo, Mr Nicholson stated: The decline... in the equity value of Coinmach is primarily due to the drop in the EBITDA multiples of the selected comparable companies used in the Market Method, and the impact of the general weakening of the US and global economy on Coinmach's projected cashflow and applied discount rates within the discount cash flow method.
15. It was submitted on behalf of DIF III that these three communications were to be considered together, and in the context of the Manager's subsequent admission in the legal proceedings that it had breached its obligation to undertake due diligence by not conducting its own modelling that challenged the growth assumptions made by the Coinmach Deal Team. The communications were said to indicate that in late 2008, the investment was "distressed" and that the original investment strategy to hold the equity interest to realise value through EBITDA growth was no longer feasible; that in November 2007, Ms Talintyre had regarded the Coinmach deal as "un-commercial"; and that by October 2008, as assessed in early February 2009, the value of Coinmach's equity, and accordingly DIF III's investment, was nil.
16. Reference was made to the Court of Appeal decision in Euro Pools Plc v Royal & Sun Alliance Insurance Plc [2019] EWCA Civ 808; Lloyd's Rep IR 595 at [39], where Dame Elizabeth Gloster, with whom Hamblen and Males LJJ agreed, summarises the approaches taken to the construction and application of clauses such as cl 7; that is, provisions which deem claims subsequently made, that arise out of circumstances of which the assured first became aware during the policy period, to have been made during that period.
17. Those principles include: first, that the provision should be construed with a view to its commercial purpose, being to provide an extension of cover "for all claims in the future which flow from the notified circumstances"; secondly, that consistently with that purpose, a provision which refers to circumstances that "may" give rise to claims sets a "deliberately undemanding test"; thirdly, that a notification need not be limited to particular events and may be to a "problem" described in general terms if that problem of itself may give rise to a claim, and notwithstanding that the quantum and character of such claims, or the identity of claimants, may not be known at the date of notification; and fourthly, that whilst the insured necessarily has to be aware of circumstances which might reasonably be expected to produce a claim (or in this case, could reasonably be anticipated to give rise to a claim), that does not "predicate that the insured needs to know or appreciate the cause, or all the causes, of the problems which have arisen, or the consequences, or the details of the consequences, which may flow from them."
18. Finally, and accepting that these principles must necessarily give way to the language of the particular clause in question, the position remains, as Toulson LJ said in HLB Kidsons (a firm) v Lloyd's Underwriters [2008] EWCA Civ 1206; [2009] Bus LR 759 at [142], that the notifiable circumstance must be such that it "may reasonably be regarded in itself as a matter which may give rise to a claim" (or in the language of the clause here, it must be a circumstance that "could reasonably be anticipated to give rise to a claim").
19. Turning to the matters relied on by DIF III, it is not contended that the fact of the Manager's admission that it did not undertake due diligence to test the growth assumptions in the model was of itself a matter that could reasonably be anticipated to give rise to a claim. That would depend on whether there were other circumstances which, taken with that circumstance, indicated or suggested that there was a "problem" which might of itself result in claims.
20. For the reasons given above, the second of the communications relied on did not provide any basis for thinking a claim of some sort might be brought against the Manager. The remaining two communications concern Coinmach's financial position in October and November 2008. Whilst the fact of its distressed financial position made it likely that there were persons or entities, including DIF III, who would suffer losses as a result of its failure, that was not of itself a matter which could reasonably be anticipated to give rise to a claim against the Manager. And as the primary judge concluded at J [348], there was nothing in any of those communications which suggested that DIF III or any other party had a basis for making a claim arising out of conduct of the Manager in relation to the DIF III investment.
21. It follows that ground 18 of the amended notice of appeal is not made out. It was not established that during the policy period the Manager became aware of any circumstances that could give rise to a claim, let alone that the 2018 claim was brought as a direct result of such circumstances.

Amendments 29 June 2020 - Insurance Contracts Act 1984 (Cth) s 54 added to coversheet. Link to text version of graph added at [78] Link to text version of table added at [103] 15 June 2021 - Revision to [172] - paragraph reference changed in HLB Kidsons (a firm) v Lloyd's Underwriters [2008] EWCA Civ 1206; [2009] Bus LR 759, from [141] to [142]. DISCLAIMER - Every effort has been made to comply with suppression orders or statutory provisions prohibiting publication that may apply to this judgment or decision. The onus remains on any person using material in the judgment or decision to ensure that the intended use of that material does not breach any such order or provision. Further enquiries may be directed to the Registry of the Court or Tribunal in which it was generated. Decision last updated: 15 June 2021