On the Link between Spot and Forward Power Prices: A Comparative Analysis of German and Hungarian Power Market Efficiency

• Název práce v češtině: O vztahu mezi spotovou a forwardovou cenou elektřiny: ; Komparativní analýza efektivnosti německého a maďarského trhu
• Název v anglickém jazyce: On the Link between Spot and Forward Power Prices: ; A Comparative Analysis of German and Hungarian Power Market Efficiency
• Klíčová slova: Riziková prémie, Spotový trh s elektřinou, Forwardový trh s elektřinou, Efektivita trhu, Nevychýlenost forwadové ceny
• Klíčová slova anglicky: Risk Premium, Spot Power Market, Forward Power Market, Market Efficiency, Forward Rate Unbiasedness
• Akademický rok vypsání: 2013/2014
• Typ práce: diplomová práce
• Jazyk práce: angličtina
• Ústav: Institut ekonomických studií (23-IES)
• Vedoucí / školitel: prof. PhDr. Ladislav Krištoufek, Ph.D.
• Řešitel: skrytý - zadáno vedoucím/školitelem
• Datum přihlášení: 30.05.2014
• Datum zadání: 30.05.2014
• Datum a čas obhajoby: 22.06.2015 00:00
• Místo konání obhajoby: IES
• Datum odevzdání elektronické podoby: 15.05.2015
• Datum proběhlé obhajoby: 22.06.2015
• Oponenti: Mgr. et Mgr. Pavel Doležel

Předběžná náplň práce v anglickém jazyce

This thesis strives to investigate several features of a developing power market in Hungary. Such market lacks sufficient liquidity, continuity of prices as well as transparency. On top of that, crucial information for estimation of power prices such as maintenances of significant generation units or changes in available cross-border transmission capacities is often not easily accessible, frequently changed or delivered very late. This is relevant also to most of Balkan countries since they are well interconnected with Hungarian power market. Given such low transparency and resulting poor ability of fundamental forward prices formation, market participants put substantial emphasis on recent outcomes of spot auction than in a more mature market. Latest settlements thus influence the whole forward curve.

This is the main hypothesis I intend to deal with. The reasoning is following: Since power is a non-storable good, the price in spot auction is an indicator of real power balance. As mentioned, the information set concerning the Hungarian market is very limited so traders tend to overreact on this auction, although the reason for a large difference between traded price and settlement price might be just a one-off issue. German market, as the most liquid, deep and transparent power market in Europe will serve as a benchmark. Data on forward prices will be collected from brokers and power exchanges aggregated at a Trayport platform, which covers practically the entire volume traded.

Hypotheses:
1) Forward prices react on result of spot auction immediately.
A first hypothesis is that a differential between day-ahead price and actual spot auction outcome has an impact on forward prices. The main explanatory variable will be the surprise of spot auction (difference between traded day-ahead price and price settled).This variable should have impact on the difference between the month-ahead price traded before the auction result and after the outcome of the exchange is published, which is at 11:30 am. Since there are usually no important new fundaments released after 11:30, there should not be an omitted variable bias. The model needs to control especially for change in German month-ahead price as German price is a benchmark and usually lower bound of Hungarian price. In general, the significance of German market increases with lower spread and reaches its maximum if there is no congestion at Hungarian-Austrian or Hungarian-Slovak border. Therefore, the presence of congestion needs to be taken into account as well.

2) Risk premia of weekly contracts are autocorrelated.
The second hypothesis arises from similar reasoning as the last one, which is the overreaction on recent settlement prices. In other words, a certain stickiness of forward prices is assumed. Since I use the price of front week traded on Friday, the calculation of projected settlement of the current week needs to be done. This is going to be made up of settled prices of working days and forward price of weekend. The difference between the projected settlement and closing forward price of next week is thus expected to have an impact on the risk premium of week to follow. I will control for any ex post change in fundaments, which include the flows from neighbour countries, Hungarian consumption as well as generation and change in German prices.

3) Risk premium is decreasing over time both in Hungary and Germany. However, German risk premium is substantially lower. Efficiency of the market is improving and errors in spot price predictions are being reduced.

In this section, we simply compare the ex post risk premia in both markets. Subsequently, the unbiased forward rate hypothesis (UFH) is to be tested. The methodology will be similar as suggested by (Haugom & Ullrich, 2013). The estimation with rolling window will be used to reveal the evolution of parameters in the UFH equation in time. The intuition is that overall volatility of forward prices and risk premia slowly decrease in time as the market is developing. Not only proprietary traders can predict the forward prices better, but also generators dispatch power in a more efficient way and flow from other countries is expected to be more efficient as well due to market coupling and more sophisticated market actors active at cross-border trading.

Core bibliography:

Bessembinder, H., & Lemmon, L. (2002). Equilibrium pricing and optimal hedging in electricity forward markets. Journal of Finance 57 , pp. 1347-1382.

Böhm, B., Haas, R., Huber, C., & Redl, C. (2008). Price formation in elektricity forward markets and the relevance of systematic forecast errors. Energy Economics (31) , pp. 356-364.

Cartea, Á., & Villaplana, P. (2008). Spot price modeling and the valuation of elektricity forward contracts: The role of demand and capacity. Journal of Banking & Finance (32) , pp. 2502-2519.

Fama, E., & K., F. (1987). Commodity Futures Prices:Some Evidence on Forecast Power, Premiums and the Theory of Storage. The Journal of Business Vol.60, No.1, The University of Chicago Press , stránky 55-73.

Furió, D., & Chuliá, H. (2012). Price and volatility dynamics between elektricity and fuel costs: Some evidence for Spain. Energy Economics (34) , pp. 2058-2065.

Haugom, E., & Ullrich, C. (2012). Forecasting spot price volatility using the short-term forward curve. Energy Economics (34) , pp. 1826-1833.

Haugom, E., & Ullrich, C. (2013). Market efficiency and risk premia in short-term forward prices. Energy Economics (34) , pp. 1931-1941.

Joëts, M., & Mignon, V. (2011). On the link between forward energy prices: A nonlinear panel cointegration approach. Energy Economics (33) , pp. 1170-1175.

Kellard, N., Newbold, P., Raynert, T., & C., E. (1999). The relative efficiency of commodity futures markets. Journal of Futures Markets, Vol.19, Issue 4 , pp. 413-432.

Nomikos, N., & Andriosopoulos, K. (2012). Modelling energy spot prices: Empirical evidence from NYMEX. Energy Economics (34) , pp. 1153-1169.