• No results found

Notes on Brattle paper 1. Transit routes analysed – general observation on German market areas

N/A
N/A
Protected

Academic year: 2021

Share "Notes on Brattle paper 1. Transit routes analysed – general observation on German market areas"

Copied!
12
0
0

Bezig met laden.... (Bekijk nu de volledige tekst)

Hele tekst

(1)

GTS/110414 Notes.doc 1

Notes on Brattle paper

1. Transit routes analysed – general observation on German market areas

On p35 Brattle states that: “One additional development since our 2007 report has been the consolidation of the German market areas […] the development of the gas trading hubs in Germany does not represent a new transit route.” Regardless of whether the traded hubs themselves provide any transit options, the market areas and the cooperation within them have – as the following extract from the Wingas website describes well:

““As far as the gas transport in and out of the GASPOOL market area is concerned, the system works very much like the EU’s Schengen Agreement,” Hümbs explains. “In order to get into the EU, citizens must show their ID cards at the border, after that they can move freely from country to country and do not have to show their ID again until they leave.”

“GASPOOL works according to the same principle: in order to feed the natural gas into the pipeline network of the market area, the transport customer closes one entry contract with one network operator in the GASPOOL market area.

“The natural gas can then be delivered to any exit point in the market area, provided the transport customer has closed an exit contract with a downstream network operator. GASPOOL also links up six EU countries: Poland, the Czech Republic, Denmark, the Netherlands and Belgium are all connected to the German market area thanks to their international network interconnection points. This means transport customers now have one single balancing zone between these countries. “The two-contract model also applies here – transport customers can transport gas from Ellund in Denmark through the market area to the Czech Republic, for example, with just one entry and one exit contract,” Hümbs explains. Traders in the market area do have to be aware, however, that no additional transport capacities were created with GASPOOL.”

Source: http://www.wingas.com/1409.html?&L=1

So market areas in Germany give shippers additional options with reduced costs, as the pancaking of TSOs’ individual tariffs has disappeared (e.g. enter Emden on GUD, exit Eynatten on Wingas, while Gaspool cooperation between the two TSOs prevents pancaking charges saving the shipper an additional entry and exit charge).

2. Brattle Table 5 – Emden to Belgium

Notes

2.1. The Table shown on p37 as Table 5 is actually Table 7 – the results in it do not relate to routes to Belgium. These are not shown in the report.

2.2. Brattle decides on p35 that the most expensive route between Emden and Belgium via the Netherlands will be decided by the highest GTS exit tariff, which is at Zandvliet. This is because Brattle states on p36 (last paragraph) that “Fluxys charges a uniform entry tariff at all entry points”. This is not true: Fluxys has two entry charges, one higher charge for points with compression attached, the other for points without compression. S’Gravenvoeren and Zelzate are both points with the higher charge, and of these two points Zelzate has the higher exit charge from the GTS system. Therefore the highest cost route via the Netherlands (including entry into Belgium) is actually via Zelzate, not Zandvliet. We have called this the ‘ADL GTS route’ in our calculations. 2.3. From the percentage results given in the paper and the results shown in the other Tables, we can

(2)

GTS/110414 Notes.doc 2 assume shipment of 1 TWh of gas and a Zeebrugge day ahead price of 24.347 €/MWh (Heren EGM, 23/03/11) to calculate a commodity charge of 0.15 €/kWh/h/y. We also assume high load factor (8000 hours).

2.4. Wingas routes are calculated from “Emden (Bunde)”. The Bunde entry point to the Wingas system shows GTS as the connected network, not Gassco. Therefore in strict terms there is an entry/exit tariff combination for GTS to add to the Wingas costs (GTS entry at Emden and exit at Bunde). In this case, however, using Bunde as an additional entry point provides another plausible tariff comparison and therefore it seems reasonable to use it in this way.

2.5. Brattle states that Wingas does not show any exit capacity at Eynatten. This is true for the

remainder of 2011. There is however data on the Wingas website showing capacity available from 01/01/12 through into 2015. Other capacity information given by Brattle for OGE, Wingas, GUD and Thyssengas appears to be correct.

2.6. October 2012 is given by Brattle as the latest date for which capacity information is available. This appears to be true for OGE. However, the Wingas website provides data until at least 2015. The Fluxys website gives data to at least the end of 2012 and Thyssengas to 2014.

2.7. The description of the possible routes does not take into account the possibility of using Gaspool (entry GUD, exit Wingas – whether or not GUD can offer competition to GTS); we have added this route. Capacity is showing as available across the entire route 01/01/12 to 09/09/12.

2.8. Using Net Connect Germany market area is also possible entering Thyssengas at Emden and exiting OGE at Eynatten; we agree there is no capacity available on Thyssengas at Emden, but as Brattle ignores a lack of capacity for the other companies this is no reason to discount this route, so we have added it also.

2.9. Aside from that and the issue with the route via the Netherlands described, Brattle’s route list appears complete and correct. Eni Gas Transport Deutschland would be a further alternative via Net Connect Germany entry, but it appears it should be discounted as capacity is marketed at Eynatten only on a restricted basis.

(3)

GTS/110414 Notes.doc 3

3. Table 5 Results calculated by ADL

Table 1: Routes Emden to Belgium, showing Brattle’s route via GTS and ADL’s alternative route via Zelzate (results highlighted in orange)

Table 1 above is similar to what we believe Brattle intended to publish as Table 5, with the addition of routes via Gaspool, NCG and an alternative GTS route via Zelzate.

Note the percentage results in our table are slightly different to Brattle’s (p36 states the OGE route is 118% of the price of the GTS route, whereas we have 117%), which is due to Brattle not including a commodity element in the Fluxys entry charge.

The route we have added via Gaspool is the cheapest alternative at 107% if the price of the GTS route via Zandvliet; it is 98% of the GTS route via Zelzate, which incorporates the higher Fluxys entry tariff described in point 2 above.

(4)

GTS/110414 Notes.doc 4 Table 2: Routes Emden to Belgium, showing Brattle’s route via GTS and ADL’s alternative route via Zelzate

(results highlighted in orange) – GTS FEES INCREASED BY 10% FOR COMPARISON

In Table 2 we show the results incorporating a 10% increase in the GTS tariffs as requested; all other tariffs remain the same.

4. What does this mean for Brattle’s Table 1, Emden to Belgium conclusions (p5)

4.1. We agree neither OGE nor Wingas is currently offering capacity at Eynatten; Wingas does have capacity available from 01/01/12.

4.2. Wingas entry capacity is actually at Bunde, not Emden.

4.3. Yes, the cost of both of these routes is around 20% higher than the GTS route via Zandvliet. 4.4. Brattle has ignored the chance to use Gaspool and NCG. The Gaspool route is only 7% higher

than the Zandvliet GTS route. The NCG route is around 20% higher.

(5)

GTS/110414 Notes.doc 5

5. Brattle Table 6 – North to South Germany

5.1. Brattle asserts at the top of p38 that “Boscholtz [sic] is the only H-gas exit from the GTS network into the Open Grid Europe network”. This is not true, as the Oude Statenzijl H-gas connection to OGE also exists and even has capacity available. We have included this route as an alternative GTS route.

5.2. H-gas connections to other TSOs whose networks or market areas reach southern Germany also exist at Bocholtz (Thyssengas and ENI DT) and at Oude Statenzijl (GUD and Wingas). We have calculated these routes as alternative GTS routes.

5.3. The route chosen by Brattle for GTS currently has no available capacity at the OGE entry point at Bocholtz. This is true until April 2013 at least, which is the limit of the published information. (It seems curious that no reference is made to this in this section of text, whereas available capacity on the alternative routes is described)

5.4. The routes entering Germany at Emden have not included Thyssengas or GUD as alternatives. There is no capacity available from Thyssengas until 2013 at least and limited capacity from October 2011 to September 2012 from GUD. We have calculated tariffs for these routes. 5.5. The Wingas route enters Germany at Bunde; in this case it forms our interconnection to GTS at

OSZ, for the route GTS Emden – GTS OSZ – Wingas Bunde/OSZ. This highlights the problem with assuming Emden and Bunde are interchangeable for calculating these routes.

5.6. Brattle states that exit fees in Germany are not included as the shipper would have to pay them whether travelling via GTS or not; this is true, but the exit fee will differ depending on the network or market area and the delivery point. This could in theory make it preferable for a shipper to choose a route which involved entering Germany on a particular network or in a particular market area.

6. Table 6 results calculated by ADL

In Table 3 we show our calculations for Brattle’s routes and our additional routes. We have also listed the availability of capacity at non-GTS entry points.

We agree that the GTS route chosen by Brattle has a price 2.54 €/kWh/h/y higher than the OGE route and 2.34 €/kWh/h/y higher than the Wingas route (note that the Wingas entry point at Bunde also features later in our table as part of route 6, this time connected to the GTS network).

The difference is not so high if an alternative GTS route is used: we have highlighted in orange results for the route that has capacity available, which is via OSZ onto OGE’s network. The GTS route is then 2.11

€/kWh/h/y higher than the OGE route and 1.91 €/kWh/h/y higher than the Wingas route. The lowest priced GTS route via GUD at OSZ has no capacity available, but it would be 1.98 €/kWh/h/y higher than the OGE direct route and 1.78 €/kWh/h/y higher than the Wingas route (and 1.71 €/kWh/h/y higher than the

(6)

GTS/110414 Notes.doc 6 Table 3: Routes Emden to South Germany, showing Brattle’s route via GTS, Brattle’s German routes and

(7)

GTS/110414 Notes.doc 7 Table 4: Routes Emden to South Germany, showing Brattle’s route via GTS, Brattle’s German routes and

ADL’s alternative routes via Germany and via GTS (results highlighted in orange for case where capacity exists) – GTS FEES INCREASED BY 10% FOR COMPARISON

(8)

GTS/110414 Notes.doc 8

7. What does this mean for Brattle’s Table 1, North to South Germany conclusions (p6)

7.1. Agreed, the GTS route is more expensive than the German alternatives, including the additional alternatives we found, and the capacity information is correct.

7.2. There is no capacity to enter OGE at the GTS connection point at Bocholtz until 2013 at least, which limits GTS’s ability to exercise market power when the German routes from Emden are congested.

7.3. Other TSOs can be used to enter Germany at Emden, although Thyssengas has no capacity and GUD has capacity only from 01/10/11.

7.4. There are alternative GTS routes which are cheaper than that chosen by Brattle, but they still have higher prices than the direct German routes from Emden.

8. Brattle Table 7 – Emden to France

8.1. On p39 Brattle states its choice of GTS route as via s’Gravenvoeren, Obbicht or Zelzate. The calculations in Table 7 use Zandvliet instead of any of these points.

8.2. Gas is said to enter the GRTgaz network at Quevy, Blaregnies or Taisnieres. Blaregnies and Taisnieres are effectively the same as Blaregnies is the name of the point on the Belgian side of the border and Taisnieres is the name on the French side.

8.3. Brattle asserts that only OGE has an exit point to France, but GRTgaz Deutschland also markets capacity at Medelsheim (at the slightly lower price of 2.19 €/kWh/h/y). There is no capacity available from GRTgaz Deutschland, just as there is none available from OGE.

8.4. There are further possible routes for this transit than those specified by Brattle. It would be possible to enter at Emden at GTS and flow via Germany to Medelsheim. Or enter Emden on OGE then flow via Belgium to Taisnieres. We have calculated results for these routes.

8.5. At the top of p40 Brattle states that the flow via Belgium, presumably from Germany, would be the same analysis as for Transit Route 1 (Table 5). This is not true, as Brattle does not take into account the charges for crossing Belgium (which depend on the route taken) nor the exit fee at Blaregnies.

8.6. It is true that there is no capacity available at Medelsheim or Wallbach.

9. Table 7 results calculated by ADL

In Table 5 we show the results of our calculations, attempting to replicate Brattle’s Table 7 and adding some further routes.

The results for the first route for OGE we can verify as correct.

(9)

GTS/110414 Notes.doc 9 lowest tariff is actually that for East Switzerland, so seems an odd choice. The Oltingue/Rodersdorf border point is located in the Middle Zone, which carries a higher tariff (33.4 CHF/m3/h/y compared to 19.8

CHF/m3/h/y for the East zone). A more plausible assumption is to use an actual transit tariff. These are not published, however, the Swissgas auction procedure for summer 2011 sets a published reserve price of 4.65 CHF/kWh/h (3.54 €/kWh/h) which we have used in our calculations. Our result for this route is 10.29

€/kWh/h/y compared to Brattle’s 8.98 €/kWh/h/y.

The GTS route calculated by Brattle contains a number of errors:

− The Fluxys Zandvliet entry charge should include a commodity fee

− No transit charges have been included – Fluxys requires payment for internal zone crossings for transit flows

− The exit charge at Blaregnies should include a commodity fee (in the table Brattle lists the exit point as Zandvliet instead of Blaregnies, which is clearly an error)

− Entry to France is listed as at Obergailbach when it should be at Taisnieres; the fee is also incorrect and should be 2.25 €/kWh/h/y

Our result for this route is 8.78 €/kWh/h/y compared to Brattle’s 5.91 €/kWh/h/y.

We therefore disagree with Brattle’s calculations of the differences between GTS routes and the alternatives: our calculations show the GTS route is 2.11 €/kWh/h/y higher than the OGE route and 1.51 €/kWh/h/y lower than the Swissgas route. This clearly has implications for Brattle’s argumentation following Table 7, which is based on the GTS route being the lowest price route.

A cheaper alternative route via GTS would be to travel via s’Gravenvoeren, shown in our table below as route 4. The result is 8.54 €/kWh/h/y, still 1.86 €/kWh/h/y higher than the OGE route and 1.75 €/kWh/h/y lower than the Swissgas route.

Our route 5, using GTS entry at Emden, connecting to OGE via OSZ and then to Medelsheim to enter France, has a similar total price as the GTS route via Belgium to France, at 8.79 €/kWh/h/y.

(10)

GTS/110414 Notes.doc 10 Table 5: Routes Emden to France, showing Brattle’s route via GTS, Brattle’s German routes and ADL’s

(11)

GTS/110414 Notes.doc 11 Table 6: Routes Emden to France, showing Brattle’s route via GTS, Brattle’s German routes and ADL’s

alternative routes via Germany and via GTS - GTS FEES INCREASED BY 10% FOR COMPARISON

(12)

GTS/110414 Notes.doc 12

10. What does this mean for Brattle’s Table 1, Emden to France conclusions (p6)

Referenties

GERELATEERDE DOCUMENTEN

At the time of writing in 2019, arrivals from Turkey to Greece have been increasing (although clearly not to the same scale as in 2015), and, as this research clearly

In the literature review performed during this research it appeared that the member states have very different types of drug policies.. This means that the decision on the

• What is the current production process, and what is the strategy for taking decisions on the salt stream in the Librox department.. • In what occasions is the salt sent towards

Management thinks that by using route planning software the problem of time windows will be easier to deal with as well, because customers with time windows are placed into a route

For the second hypothesis, the link between happiness-oriented and problem-oriented working style with regard to the ratings of trust in the counselor, the

First, the matrices A, B, C are reduced to a lower-dimensional triplet A, B, C, with B and C nonsingular, using orthogonal transformations such as the QR-factorization with

In a laser, coherent Jaynes-Cummings Rabi oscillations become irreversible emission events because of the stochas- ticity derived from all those incoherent processes that lead to

Nevertheless, the place is located at the trajectory to mount Kailsa in the Himlayan range, and mentioned in the grand pilgrimage route in the Mah  bh  rata,