‘Unbundling’ – what it is, what it isn’t, and what it could mean for our ferry services.

February 12th 2022

In the last couple of weeks there has been some coverage in newspapers about ‘Project Neptune’. Project Neptune is a study being undertaken by consulting firm Ernst & Young for Transport Scotland, into the future organisation of Hebridean ferry services. It has not yet been published, but its recommendations are expected to be opened for public consultation soon.

The full detail of the study’s remit isn’t known, but we understand it will look at the roles of Transport Scotland, CMAL and CalMac, and whether there might be better ways of delivering ferry services, including “…an analysis of the challenges and opportunities associated with options for decentralisation (unbundling of routes into smaller packages).”

It is widely anticipated that one main recommendation of the report will be ending the split between the vessel owning company (Caledonian Maritime Assets Ltd, CMAL) and the vessel operating company, David MacBrayne Ltd. CMAL was created back in 2006 as a way to comply with the perceived requirements of EU competition law. There was much debate at the time around whether such a split was indeed required by the EU, and whether other more conventional arrangements could have met the EU requirement. But that argument is largely moot now, since the UK is no longer in the EU.

It is likely therefore that CMAL will be dissolved, or at least be made responsible only for ports and harbours, rather than vessels. Certainly this committee would not mourn the loss of CMAL, given our experience of the organisation. There also seems to be little support for CMAL in any Scottish political party, in the light of the fiasco surrounding hulls 801 & 802.

What seems to have less clear political support is the concept of ‘unbundling’. News that the Ernst & Young report was to include consideration of “…unbundling of routes into smaller packages.” led to press reports of the impending ‘privatisation’ of the ferry network, and questions during FMQs of Feb 3rd:

In newspaper articles, public debate online, and among MSPs, there seems to be a lack of understanding of what unbundling is, and a conflation with privatisation (either intentional or otherwise). It seems that mention of unbundling raises the spectre of private ferry services being left to the whims of the profit motive, with fares rising, services reducing and timetables cut.

However, unbundling does not equal privatisation.

There are many arguments in favour of unbundling, as recognised by Western Isles Council in a press release from last week. When the current system is failing in so many different ways, we agree with Comhairle nan Eilean Siar that the topic needs to be explored – but first it needs to be explained. So that’s what this post is about – explaining what unbundling actually is, and looking at some of the benefits it could bring.

So how are services tendered now?

The Clyde & Hebrides Ferry Contract (CHFS) is put out to tender every 8 years, with the current contract expiring in 2024.

‘Caledonian MacBrayne’ no longer exists as a company.

Instead Caledonian MacBrayne (or CalMac) is just a brand name for a collection of publicly-supported services that are awarded under contract to the best bidder. In the first CHFS tender of 2008, competing bids were received from David MacBrayne Ltd and Serco, with David MacBrayne winning out. In the second CHFS tender of 2016, again David MacBrayne won the contract after the Serco bid failed to progress (Had Serco won, the branding would have remained as ‘Caledonian MacBrayne’).

Similarly, Northern Isles services to Orkney and Shetland are put out to tender regularly, with David MacBrayne and Serco being the main competitors.

So we already have competitive tendering for public ferry contracts in Scotland, open to bids from private companies. Services to the Northern Isles are operated by a private company (Serco), with fares and service levels set by government in return for a subsidy to make up the difference between fare revenue and operating costs. The same basic arrangement applies to Clyde and Hebrides services – but in that case, the winning bid came from David MacBrayne, who happen to be state owned.

So it’s clear that a public service ferry route is not ‘privatised’ simply as a result of being put out to tender, nor is it ‘privatised’ if it’s operated by a private company, since both of those things already apply to Scottish public service ferry routes. No-one is claiming that Northern Isles or CHFS ferry services are privatised, including the First Minister as recently as FMQs three days prior to writing this.

This is common sense if we draw a comparison with our public road network – we have private companies building our roads, and private companies maintaining our roads – but we still have a state managed, state funded public road system.

Just because a public service is delivered by a private company, it is not considered to be privatised.

If ownership of the operating company is not what distinguishes a public ferry service from private, what does?

In short, Public Service Obligation. Where the free market could result in high fares, an inadequate service or perhaps no service at all, but there is a social need for the service to operate, then the state can decide to apply a Public Service Obligation on the service. PSOs apply to our ferry services, but also to many bus services and even air services. West Coast Motors are a private company, but they operate bus routes with a PSO and receive a subsidy in return for maintaining minimum service levels. Loganair operate air services to and from the Hebrides and the Northern Isles, most of which would not exist without benefitting from a PSO and the government subsidy that goes along with it.

All modes of public transport in the Highlands and Islands have some degree of PSO. Private companies receive government funding and in return are obliged to deliver defined public transport services, be that in terms of route operated, timetable or fares. A modern Hebridean ferry service could not exist without PSOs and government subsidy, and therefore there is literally no-one who seriously advocates for privatisation of the Hebridean ferry service, including us. Privatisation that means the removal of PSOs and government subsidy would be mad. That is the scenario that is sometimes cited as the argument against unbundling – that it equals the removal of PSOs and a Wild-West free-for-all where only lucrative routes are operated and profit is the only motive. No-one wants that.

So if unbundling isn’t privatisation, what is it?

It simply means that rather than offering the CHFS tender in one large ‘bundle’ resulting in one company having a monopoly over all 25 routes, the contract is instead split up into a number of smaller bundles, perhaps down to individual routes. Rather than issuing one large tender for the entire network, there would instead be a series of smaller tenders for parts of the network.

So what’s best – one big bundle, or a bunch of smaller ones?

When the possibility of unbundling has been raised before, David MacBrayne, the Trades Unions and some island communities were opposed to it. The government subsequently opted to offer the CHFS routes as one large contract.

There are three main advantages to a single large whole-network contract put forward by proponents –

1. That it’s the thin end of a privatisation wedge, that will ultimately lead to reduced timetables, higher fares and poorer services.
2. That a single large contract results in economies of scale that keep running costs down.
3. That a single contract service is more resilient, enabling vessels to move around the network to wherever they are needed, either in the event of breakdown or to cover for annual service.

On the contrary, proponents of de-bundling say that smaller contracts result in better outcomes in all three areas. Here’s how –

1. Service levels and fares

In an unbundled system, every aspect of the service delivered could be still be specified by government, just as it is now. Government could specify the fares to be charged; the minimum timetable that should operate; whether a route should be catered or not; how the booking system would operate; the emissions from the propulsion system, and any other detail desired. They would have exactly the same control over every aspect of the service as they currently do, but it would be up to the bidding company to decide how best to deliver it. So the only way that timetables could be reduced or fares increased would be if the government directed it. The level of control that the government have over the service would be decided by government themselves. So there is no loss of control over service levels or fares, just because a route was being tendered on its own rather than together with 24 others.

Proponents of de-bundling say that it could actually lead to an improvement in service levels beyond that specified, because of the competition at tendering time.

A tendering company could achieve a competitive advantage over another by proposing service enhancements, like an improved timetable, faster ferries, or a longer operating day.

However, this greater competition can only be delivered if –

• Bidding companies make their own ferry design and procurement decisions.
• Bundles are made as small as practical, (perhaps single routes), resulting in a greater variety of companies and organisations able to bid for them.

A large network contract is very costly and time-consuming to bid for, and comes around only once every 6 – 8 years. It takes hundreds of thousands of pounds and hundreds of people-hours to prepare a bid for the Northern Isles or CHFS contracts. It is very high-risk, and only the largest companies can afford to consider it. That is one of the main reasons that only Serco and David MacBrayne have made serious bids for either of the two big whole-network bundles. (In-fact, it is highly probable that the Serco bid for the CHFS contract of 2016 that was deemed non-compliant was a deliberate act by Serco. Even they may have decided that the investment and risk required for the bid process was not worth it. If a service company the size of Serco is put off by the CHFS tender, perhaps it demonstrates that the contract is already too large and complex)

However if the tenders were smaller, they would be more open to smaller private companies and public organisations to bid for. This is evidenced by the Northern Isles services – several other smaller companies expressed a desire to tender for individual routes to Orkney and Shetland when first tendered in 2006, but they subsequently withdrew when the decision was taken to tender as one single network contract.

There are a multitude of small ferry and general-marine operators in Scotland who if given the opportunity, would be very well placed to bid for a 1 – 3 route contract. This is exactly what the Western Isles Council have proposed – they would like the opportunity to operate their own ferry services, just as the Orkney and Shetland Island councils do very successfully. They feel that not only could they operate the services more cost-effectively, but by bringing the management to the islands, they would have a more island-responsive service.

Other routes might attract bids from existing private ferry operators – Pentland Ferries and Western Ferries for example. Or small commercial boat operators already active in the area might seek to diversify into ferry operations. There is even the prospect of island communities themselves setting up community-owned and controlled ferry companies. This is a possibility that we have ourselves commissioned a piece of consultancy on, and will be the subject of a subsequent post.

But whatever form the bidding organisations take, smaller contract bundles will encourage more competition at tendering stage. Rather than receiving one bid from the incumbent monopoly operator (currently David MacBrayne Ltd), Transport Scotland might find they have three or four competing proposals for each route.

Those competing proposals would be judged on a combination of value-for-money and service enhancement. One of the main ways that a bidder can gain a competitive advantage is by specifying and designing their own vessels. Rather than being obliged to use ferries from the CMAL catalogue as at present, they would be free to propose replacement vessels of their own. So the bidder would be motivated to ensure that the vessels they buy would be efficient, good value-for-money, productive and reliable. Good ferry design and procurement would be one of the key means by which a bidder would gain a competitive advantage in the tender. By using vessels that burn less fuel, cost less to build or can be crewed more efficiently, they would be able to offer service enhancements, a lower subsidy requirement, or a combination of both.

To demonstrate how the service and vessels could improve as a result of unbundled tendering, we can draw on an example from Norway, where almost all routes are tendered individually.

The service between Bodo and the southern Lofoten Islands is one of Norway’s most exposed, and has a crossing time of around 3 hours. In exposure and service length, it is very similar to services to the outer isles and Islay, for which two new vessels are currently in the process of being procured by CMAL.

Although they have very similar requirements in terms of route length and capacity, CMAL / David MacBrayne have specified vessels that compare very poorly with the Landegode and her sister-ships:

		MF Landegode, Norway.	New Islay Ferries
Year delivered		2012	expected 2024
Size (Length / beam / draught)		96m / 17m / 4.2m	94.8m / 18.7m / 4.0m
Car Capacity		120	100 – 107
Passenger Capacity		390	350
On-duty Crew		12	27
Daily operational hours		20 hours	14 hours

Faced with very similar requirements, Torghatten have arrived at very different solutions to the CMAL/David MacBrayne designs. The Landegode can operate for nearly 50% longer every day, and requires less than half the number of crew. If CMAL/David MacBrayne had chosen designs like the Landegode, they would have been able to offer the people of Islay an additional 2 return sailings per day; and taxpayers an approximate saving of £1.2 million in crew costs each year.

Or they could also have chosen to use the crew cost savings to increase services elsewhere – ie more sailings on more routes. Making a vessel more productive (ie fewer crew) does not necessarily result in over-all employment reductions – the other outcome is that more services and vessels can be operated with the same total number of crew.

The difference between Torghatten and CMAL/David MacBrayne is motivation. Torghatten had a commercial motivation to propose a vessel with low running costs and long operating hours, so that their bid was competitive. They won the contract because they required a lower operating subsidy from government, and/or could offer a better timetable to users.

When specifying the vessels for Islay, CMAL/David MacBrayne had no such commercial motivation. Whatever the operating deficit of the new Islay vessels, the taxpayer will pick it up; and because under the current tendering arrangements any competing bidder is obliged to operate the same CMAL vessels as the incumbent, there is no opportunity for an alternative operator to propose any other vessel.

In Norway, smaller bundles enable greater competition between prospective operators; and that competition results in better outcomes for both users and taxpayers.

2. Running costs

As shown by the Landegode example above, those bidding for unbundled Norwegian contracts have a greater incentive to operate productive and efficient vessels, in order to be competitive at tendering time. The counter-argument to that is that a whole-network contract achieves cost savings as a result of economies of scale and the buying power of a larger company. There is just one head office to pay for, and greater buying power for things like dry-docking.

However, there is lots of evidence that the lack of competition at tendering time is a far greater negative influence than any gains from the economies of scale in a monopoly.

As shown above, competition to pitch the best service with the lowest subsidy requirement led to improved efficiencies in the ferries serving the Lofotens. The Landegode is just one example where Norwegian vessels have much more efficient crewing, but she is not the exception. The Loch Frisa coming to Mull in the summer of ’22 operated with a crew of four when still known as the Utne in Norway. She is expected to have a crew of seven when she takes up service on the Craignure-Oban route under David MacBrayne.

It is very easy to find lots of examples from both the UK and abroad that demonstrate the poor productivity (ie over-crewing) of CalMac vessels:

		CalMac Finlaggan	CalMac new Islay ferry	Pentland Ferries Alfred	Standard bi-directional ferry, Norway.
Car Capacity		85	100 – 107	98	80 – 100
On-duty crew		29	27	14	12

So it does appear that David MacBrayne / CMAL lack an incentive to bring their crewing levels in line with industry norms when designing new vessels.

Not only that, but there are many instances across the fleet where opportunities to reduce crewing exist, but those opportunities are not taken. The MV Isle of Lewis is the most egregious example of crewing being maintained at un-necessarily high levels –

The Isle of Lewis is the second-largest vessel in the CalMac fleet. After being retired from the Ullapool-Stornoway service following the arrival of the Loch Seaforth in 2015, she was assigned to the Barra-Oban route. Although the longest, Barra-Oban is one of the quietest major routes in the network. The Isle of Lewis can carry up to 680 passengers with a minimum crew (dictated by the MCA) of 24. She also has a lower passenger carrying ‘mode’, that reduces her limit to 132 people with a minimum of 18 crew. As can be seen from the graph of carryings below, she could operate on the lower ‘mode’ throughout the winter and still have ample capacity for the service. This would enable a significant reduction in the number of crew on duty.

Despite having this opportunity to make cost savings of approximately half a million pounds per year, David MacBrayne choose not to.

The example of the Isle of Lewis is repeated around the network – passenger capacities are far in excess of the number of passengers actually carried. A major driver of crewing levels is the ‘minimum safe manning’ required in the event of an emergency. A vessel with more passengers requires more crew to help load lifeboats in the event of an emergency; therefore running vessels with passenger certification in excess of the number of people actually on-board is a direct cause of wasteful over-crewing, yet David MacBrayne do not appear to have any incentive to use their crew more productively and cost-effectively.

In the current whole-network monopoly contract, where any potential competitor is obliged to use the CMAL vessel and crewing regime already in service, the incumbent has no incentive to find economies, and the prospective new operator has no opportunity to find economies.

The Caledonian MacBrayne network is regarded by international industry experts as grossly over-staffed and unproductive when compared with more efficiently run ferry services. The tendering process that tends to suppress competition and prescribe the vessels that must be used, is considered by those same experts to be the prime cause of the system’s inefficiency.

Proponents of unbundling draw on examples like the MV Isle of Lewis to demonstrate that contrary to a monopoly driving economies of scale, it instead engenders complacency and waste.

3. Service Resilience

The third major defence of a single large monopoly contract is that it enables the operator to move vessels around the network as necessary to cover annual servicing and un-planned unavailability of vessels (ie breakdowns).

At the time of writing (February 2022), the CalMac fleet is suffering from an unprecedented number of breakdowns and dry-dock over-stays. In order to cope with the vessel shortage, ships are being moved between routes, so that some level of service is maintained, albeit out of timetable.

This is exactly the kind of situation that supports the view that Hebridean ferry services are inter-dependent on one another. If the ferry that serves Coll and Tiree is broken down, then the ferry that serves Mull can be borrowed to lend a hand. If the ferry that serves Islay is stuck in dry-dock longer than planned, then a ferry can be brought up from Arran to stand-in.

Inter-dependence between routes in this manner is not a necessity however – it is a strategic choice that has been made by David MacBrayne, Transport Scotland and CMAL. The three organisations currently responsible for planning vessel replacements have repeatedly chosen to accommodate traffic growth by building larger vessels, rather than adding more. This results in each major route having 1 or perhaps 2 large vessels instead of 2 or 3 smaller ones.

Other operators choose a different strategy – take for example Western Ferries on the Clyde. They have just one route, joining Cowal to McInroy’s Point on the South shore of the Clyde. They operate four vessels, and adjust the number in operation to meet demands of the time and season. With four vessels, they have four times the redundancy of a route with only one. Only at peak times does the service rely on all four; and if one is unavailable they suffer only a 25% reduction in capacity. By contrast, a breakdown on a service with just one ferry results in 100% loss of capacity. (detractors may say ‘but that’s the Clyde, not the Minch’ – but the principle is exactly the same, although the style of vessel used may be different).

Western Ferries operate a self-contained, self-reliant and highly resilient ferry service, without having to borrow vessels from neighbouring routes when they have a breakdown.

We can contrast Western Ferries’ strategy with the David MacBrayne /CMAL/ Transport Scotland strategy, by looking at the impact of a breakdown that is ongoing at the time of writing

The MV Hebridean Isles operates services from Uig across the Minch to Tarbert and Lochmaddy. Both routes are worked by one vessel, which currently has a broken engine and sits idle in Lochmaddy. With just one vessel working two routes, there is no resilience to breakdown whatsoever, without bringing in a vessel from another route. At the moment however (Feb 12th 2022), there are so many vessels with unscheduled technical issues, that no vessels can be spared. So because of the fault with one engine, both the Uig-Lochmaddy and Uig-Tarbert routes have come to a standstill for several days.

In ‘normal’ circumstances, a vessel would be borrowed from another nearby service, such as the Mallaig-Lochboisdale route. That is also a single-vessel route however, so both services end up being disrupted.

The alternative option for services out of Uig is to have two smaller vessels, one working each route. In the event of breakdown, the second vessel would be able to cover, and rather than disrupting a neighbouring route or having 100% loss of service (as are the current options), the service would suffer just a 50% reduction in capacity.

Opting for a small number of large vessels rather than a large number of small vessels is a strategic choice. It is not an inherent and unavoidable consequence of operating ferries in the Hebrides.

The ‘one large vessel’ model is being repeated however, with the planned (but very much over-due) introduction of hull 802 to replace the MV Hebrides. Local communities and industry experts advised that services from Uig should be worked by two smaller vessels, because that would give double the resilience without disrupting other routes.

As demonstrated by current performance, the CalMac network is far from resilient. Breakdown on one route often leads to a cascade of borrowing from other routes; far from being an advantage, that is a weakness.

On an unbundled route, a strategy more like Western Ferries’ would give the service in-built resilience without the need to borrow from other routes. The Craignure-Oban service could operate with three vessels for example, making it far more resilient than the current 1.5 vessel set-up. It could reduce to two in-service vessels during winter, with each taking it in turns to dry-dock.

On the smallest routes that could not realistically justify two vessels (like for example Tobermory-Kilchoan), then there would be a need for a ‘roving spare’ that to cover for dry-docking and breakdown. But that is very similar to the method currently employed on the minor routes – a spare vessel stands in whilst the main vessel is serviced. In an unbundled system, that service could be provided by another operator on commercial terms, or neighbouring operators might come to mutual agreements to cover one another. Just like a bus operator can hire in a spare when his own bus is having a service.

What would an unbundled Hebridean ferry service look like?

As suggested by Western Isles Council, there is no reason why a multiplicity of ferry operating companies cannot all work under the same Caledonian MacBayne brand if that was desired. Common ticketing from a shared ticketing system could easily be retained. ‘Caledonian MacBrayne’ and ‘CalMac’ are trademarks owned by the government, and currently David MacBrayne (or more specifically their subsidiary CalMac Ferries Ltd) are simply granted the right to use the trademark for the duration of the CHFS contract. There is therefore no barrier to the government granting that same right to any other operator.

So unbundling means neither privatisation nor ‘breaking up‘ the CalMac network. The unified Caledonian MacBrayne ‘brand’ that is synonymous with the Hebrides could remain, as would a single website, unified ticketing and central marketing.

How would the current monopoly contract transition to an unbundled system?

The perceived difficulty in moving from one system to another is an argument sometimes used against unbundling. What would happen to all the CMAL vessels? Would there not suddenly be a fleet of expensive vessels that new operators might not want?

The logical and pragmatic answer to this is that routes would be unbundled on a phased basis. Competitive tendering will work best if one or two routes come up for tendering each year, enabling smaller operators to tender more easily and with less risk. Routes could be released for tendering progressively, and thereafter a rolling schedule of tender renewals would be established with a different route coming up for tendering each time. David MacBrayne would be free to compete on these tenders of course – but there would be increased likelihood that their bid would not be the only one received, and all bidders would have both the incentive and the opportunity to propose better and more cost-effective ways of delivering the service.

A key part of unbundling is that the vessel operator is also the vessel owner. Any tender submission would include a vessel replacement plan, laying out a program of vessel renewal over the course of the contract. Responsibility for vessel design and procurement would therefore move from the government to the private operator, which is where the professional competence lies. Private procurement is also much quicker than public sector processes.

It would therefore make sense to progressively release routes for unbundled tendering based on the age of their current vessels. As the vessels on a route reach their natural replacement age, that route can be opened to tendering. The bidders would therefore start a 10-15 year contract with vessel replacement occurring the beginning of that period.

So is this something we should advocate for on Mull and Iona?

Unbundling is coming to the fore at this moment not just because of the poor state of our ferry system and the problems at Fergusons, but because the CHFS contract will be coming up for renewal in just two years. If a new contract model is to be introduced, the next opportunity is 2024. The routes with the oldest vessels are the obvious first candidates for unbundling. That puts the Craignure-Oban service at the top of the list of unbundling candidates, because after the renewal of both of Islay’s vessels by CMAL between now and then, the MV Isle of Mull will be the oldest vessel in the fleet.

This article is obviously not neutral – MIFC believes there are many potential benefits from unbundling, and like Comhairle nan Eilean Siar, we believe the whole topic needs exploration and examination. But our job is to represent the community, not just propose our own ideas. So we want to hear from you on this topic, please tell us what you think. Please comment or email us if you have concerns, ideas or suggestions on aspects of this topic, or anyting that we have not covered here. We will follow up on this with broader consultation and survey work. Debate on this topic needs to be properly informed, so we hope this makes a good start at explaining what unbundling could mean.

“A consultation on a change in how ferry services are delivered allows a healthy debate and if this leads to more local accountability and service delivery that places island needs first that can only be a good conversation to have.” Councillor Uisdean Robertson, Chair of Transportation, Comhairle nan Eilean Siar

But there is more ….

As mentioned further up, one of the potential outcomes of unbundling is that island communities might bid to operate their own ferry services, via a community-controlled or community-owned ferry company. Read about that here.