Turkey wants more FSRU: Is it really worth it?

It has become usual news for Turkey cutting gas flows to power plants during winter months in recent years. The problem is peak residential consumption for heating. Gas supplies to power plants are interrupted for the sake of continuous supply to residential and industrial customers.

Long term import contracts of Turkey are far from providing sufficient offtake flexibility during winter. To solve peak winter demand problem, Turkey indeed had some good options for the last 10 years, such as building the Tuz Golu (Salt Lake) underground storage facility, capacity expansion of Silivri underground storage facility, and capacity expansion of Marmara Ereglisi LNG regas terminal. Nevertheless only the upgrade of the Silivri storage has been realised so far (there is still more upgrade to be done in Silivri though). Tuz Golu has just commissioned this year, after tens of years it was first planned, and it is finally expected to start to contribute to supply for the peak demand next winter (Dec 2017).

As the capacity expansion projects delayed so far, winter peak gas demands began to hurt more, with longer blackouts due to much more significant interruptions of power plants, as happened this winter.

Under these conditions Turkey preferred to employ its first FSRU at the end of Dec 2016. It has 20 million cubic meters (mcm) per day send-out capacity, where 14 mcm/d of it is practically in use. In January 2017 four conventional cargoes of LNG are regasified from the FSRU to the Tukish gas network: A good relief for the gas balance of the transmission grid indeed. Moreover, it is understood that 2 or 3 additional FSRUs are planned to be employed, 1 by BOTAS and other(s) by private players.

Now first question is ‘who consumed the FSRU-supplied volumes this winter?’ The answer is ‘power plants’, as other consumer segments, namely industrial and residential customers, have priority over power plants in terms of gas supply security and they would be supplied in any case with the current long term supply sources.

We don’t have public and transparent daily data for the real gas demand of the gas fired power plants in Turkey. However we can approximately find the volume that have been cut off to the power plants during peak winter demand periods.

Average daily gas demand of power plants during the months of December is around 40 mcm. According to the monthly gas market report of EMRA (Energy Market Regulatory Authority) for December 2016, monthly gas consumption of the power plants is dropped nearly by 600 mcm compared to the same month of 2015. Indeed this 600 mcm figure is fairly consistent with the interruptions which have been realised in the range of 50-90% of the daily gas demands, ordered by the gas TSO (BOTAS) during the last 20 days of December 2016.

Of course there are other factors of electricity generation such as the effect of the hydro, the price dynamics at the power market or the impact of cutting off the more efficient new power plants while continuing to supply the less efficient old ones, etc. Assuming these factors more or less the same along the previous years, we can interpret that the actual gas demand has been around 0.5-1 bcm above the actual consumption in the recent years in Turkey.

Following figure demonstrates the comparison of gas consumption and actual gas demand in the recent years in Turkey:

Source: EMRA natural gas market reports and author’s analyses

Now let’s estimate the gas volume that would be cut-off to the power plants in the coming years if the conditions in 2016 were assumed unchanged (i.e., no additional storage capacity, no FSRU). Weather conditions were severe in December 2016. Even so, if we assume – as a worst case scenario – that this gas shortage happens twice per year in the coming years, there should be an unmatched demand of around 1.2 bcm per year due to the peak winter.

Very important point worth mentioning at this point is that even during the bearish aggregate gas demand of last 3 years, residential demand keeps rising in Turkey as shown below:

Source: EMRA gas market reports

Residential demand grew by 25% during the last 3 years, where industrial demand did only 12% and demand for gas to power declined by 29%. Apparently residential consumption drives the aggregate gas demand in Turkey, meaning that the demand becomes more concentrated on peak winter months (December, January, February).

Interesting point is that Turkey has more than enough supply contracts which adds up to 52 bcm per annum. When the idle capacity at the LNG terminals included, this amount even increases to 58 bcma. Note that annual demand has barely been reached to the level of 49-50 bcm so far, it follows a bearish trend for the last 3 years, and it is not expected to increase above 58 bcma in the coming years. Hence, speaking of supply shortfalls, they happen to be only in the months of December, January and February. Indeed in many of the recent years, such supply problem has been experienced only in just one or two of these months, not all of them in the same year.

Thus in fact the supply shortfall is around 40-50 days with a roughly estimate of 1.2 bcm in total per year, which is a typical problem to be solved by additional storage capacity, as Turkey’s total annual import contract volumes is enough to meet this additional demand when there is available storage capcity. Lack of sufficient underground storage capacity led Turkey to buy additional spot LNG cargoes during winter for the last 10 years. And because of very slow pace of new investments, and therefore insufficient additional storage and LNG regas terminal capacities, now Turkey employs FSRUs for 365 days for the sake of meeting peak demand of 40-50 days per se.

Note that the regasification fee of the current FSRU is USD 0.7 per MMBtu more than the regas fee of the Egegaz and BOTAS LNG terminals.

Besides, as the real need for FSRU is at most for the 3 peak winter months, Turkey needs to import additional LNG cargoes during the remaining 9 months to keep the FSRU tanks cool enough throughout the year. This means not only additional fee for using the FSRU, but also it means additional and very unnecessarily undertaken risk exposure for the current long term take-or-pay contracts of Turkey, especially of BOTAS as it uses the FSRU at present. Any imported LNG molecule through FSRU during the off-peak 9 months will be off-taken instead of the current long term contracts and eventually will increase annual and summer take-or-pay risks of Turkey.

On the other hand Tuz Golu storage facility will commence withdrawals in Dec 2017 and it is expected to have a daily withdrawal capacity of 44 mcm. This is slightly more than the level of the daily gas volume which is cut off during peak winter demand days indeed. This facility’s contribution to the system is much greater in terms of volume, efficiency and effectiveness compared to the contribution of an FSRU. It has almost triple of the practical send-out capacity of the current FSRU. And constitutes no risk for the take-or-pay positions of the current long term contracts, even mitigates these risks indeed.

Moreover, expedition of the upgrade investments at BOTAS LNG terminal and Silivri UGS will add additional supplies to the system in winter peaks.

Can more than one FSRUs and upgraded storage and LNG regas terminals altogether concurrently be usefully operated in Turkey?

Assuming there won't be an unexpected upwards shift at the current gas demand dynamics, the answer can be yes only if Turkey expands bidirectional interconnector capacities at the borders: Greece, Bulgaria, Georgia and Iran. This will facilitate Turkey’s supply optimisation. At the moment, Turkey cannot make any optimisation of this kind through cross-border flows due to lack of excess capacity. Or another viable ‘yes’ option would be BOTAS having its own FSRU which would enable BOTAS to get commercially use of it in other gas markets in the region during the 9 months off-peak period.

Otherwise Turkey must think twice and make the necessary benefit/cost analysis whether it is the best strategy to employ more FSRUs besides other more efficient supply flexibility options it has now, and/or it is expected to have in the very near future.

Remember, bird in the hand is worth two in the bush…