Companies whose sales volumes or costs change with the weather come to these contracts when they try to reduce swings in their profits through trades with financial institutions and insurers.

Weather indexes and how degree days are counted

The payments on a weather derivative depend on an index built from observations of temperature or rainfall. The index must be an objective value that neither party can influence, and in Japan the observations published by the Japan Meteorological Agency (JMA) have been used1. It is a derivative whose underlying asset is a weather index rather than a stock price or an interest rate.

A common temperature index is the degree day. Consider a hypothetical example in which the base temperature is 18°C and the daily average temperatures over three days are 10°C, 14°C, and 19°C. Heating degree days (HDD) add up the differences on days when the average is below the base: 8 plus 4 plus 0 gives 12°C days. Cooling degree days (CDD) add up the differences on days when the average is above the base, so over these three days there is only 1°C day, from the 19°C day. A document from the Ministry of Land, Infrastructure, Transport and Tourism also uses the terms cooling degree days and heating degree days for indexes that estimate the heat needed for cooling and heating, and the unit is °C days2.

In an example in a report commissioned by the JMA, the daily average temperature is the day's maximum and minimum temperatures added together and divided by 2. The rules that the US exchange CME (Chicago Mercantile Exchange) filed in May 2024 for futures on US cities do the same. The base temperature is 18°C in the report's example and 65°F (about 18.3°C) for the CME futures on US cities. Because those futures count in Fahrenheit, their values are in °F days3. The period is chosen to match the season: the report's example uses CDD from July 1 to September 30, and CME's contracts include HDD from November to March4. The hotter the summer, the larger the CDD, so the index clearly separates a cool summer from a very hot one5.

Besides temperature, indexes can be built from rainfall over the period. Others count days: days with at least 1 millimeter of rain, days with at least 1 centimeter of snow on the ground, and days with a maximum wind speed of at least 15 meters per second. Some indexes combine the CDD of several cities, and others combine temperature and rainfall6.

To choose the index, non-life insurers, weather service companies, and others analyze how the index relates to the user's earnings, using past weather data and financial data7. The contract also fixes the weather station whose observations are used. An example in a 2007 paper uses the AMeDAS station at Iwamizawa (AMeDAS is the JMA's network of unmanned regional weather stations)8.

Option payments and profit and loss at expiry

In the option form, the buyer pays an option premium. The buyer receives the amount by which the index ends above the strike price (call type) or below it (put type), multiplied by a fixed amount per unit. If the index does not end above the strike (call type) or below it (put type), the buyer receives nothing. The main terms are the strike, the amount per unit, and the payment cap (the maximum payout)9.

Consider a hypothetical example in which Company G sells heating fuel, and its sales fall when the winter is warm. In a normal year, its HDD for the winter (December 1 to the end of February) is 1,000°C days, and each °C day below that reduces its profit by 100,000 yen. Company G buys a put option with a cap on the payout (a put spread) from a non-life insurer on the following terms.

Term Value
Index Winter HDD (base 18°C)
Strike 900°C days
Amount per °C day 100,000 yen
Payment cap 20 million yen (reached at an HDD of 700°C days or lower)
Premium 1.5 million yen

If the HDD falls below 900°C days, Company G receives 100,000 yen per °C day. At 800°C days, the difference from 900 is 100°C days, and multiplying by 100,000 yen gives 10 million yen. Below 700°C days, the payout stays at the cap of 20 million yen. This has the same shape as buying a put with a strike of 900°C days and selling a put with a strike of 700°C days. The report commissioned by the JMA describes this combination of a bought put and a sold put as the most common product in weather derivatives10.

−500+50+100+150+200700900Profit/loss (°C days)Winter HDD (°C days)ProfitLossBreakeven 885
Figure 1Company G's profit and loss on a bought HDD put spread
Figure description

This figure shows the profit and loss on Company G's weather derivative for each level of winter HDD. The vertical axis shows profit and loss including the premium, in °C days; converting at 100,000 yen per °C day gives the amount in yen. At an HDD of 900°C days or more there is no payout, and the loss equals the premium, −15°C days (−1.5 million yen). Below 900°C days, profit and loss rises by 1°C day for every °C day that HDD falls. It reaches 0 (the break-even point) at 885°C days and a profit of 185°C days (18.5 million yen) at 700°C days, and it stays at 185°C days below that. The warmer the winter, the larger the payout, but nothing is paid beyond the cap.

Looking only at the weather derivative, Company G loses the premium in a cold winter. Combined with the change in its core business profit, the results are as follows.

Winter HDD (°C days) Change in core business profit (million yen) Payout from the weather derivative (million yen) Premium (million yen) Total (million yen)
1,100 10 0 −1.5 8.5
1,000 0 0 −1.5 −1.5
950 −5 0 −1.5 −6.5
800 −20 10 −1.5 −11.5
600 −40 20 −1.5 −21.5

In a cold winter (1,100°C days), core business profit rises, and the total is positive even after paying the premium. In a warm winter (800°C days), core business profit falls by 20 million yen, but Company G receives 10 million yen, so the total falls by only 11.5 million yen. On the other hand, a winter that is warm but not warm enough to pass the strike, such as 950°C days, brings no payout, and beyond the cap, such as at 600°C days, the payout does not keep up with the loss. The option seller, for its part, receives the premium in exchange for paying up to the cap in a warm winter.

Swaps pay in both directions

In the swap form, the parties agree on a reference value for the index. If the index ends above it, one party pays; if it ends below, the other party pays, in each case the difference multiplied by a fixed amount per unit. Unlike the option form, no premium changes hands. The swap lets two businesses whose profits move in opposite directions with the weather exchange their risks11.

Consider a hypothetical example in which Company G enters into a swap with a business whose profit falls in a cold winter, with a reference value of 1,000°C days and 100,000 yen per °C day. If the HDD is 900°C days, Company G receives 10 million yen, which offsets the 10 million yen fall in its core business profit. If the HDD is 1,100°C days, Company G pays 10 million yen, which offsets the 10 million yen rise in its core business profit. Company G limits its loss in a warm winter but gives up the gain in a cold winter.

In a real example, Tokyo Electric Power Company (TEPCO, as it was then) and Tokyo Gas entered into a swap (a temperature risk exchange contract) on Tokyo's average summer temperature for August and September 2001. A hot summer raises revenue for the power company and lowers it for the gas company, and a cool summer does the opposite. Under the terms, TEPCO would pay if the average temperature was above 26.5°C, and Tokyo Gas would pay if it was below 25.5°C. The average temperature that summer was 24.8°C, and Tokyo Gas paid TEPCO about 320 million yen. The deal was directly between the two companies without a financial institution, and in the report TEPCO listed lower fees and other costs, compared with buying a similar product from a financial institution, among its benefits12.

Growth in Japan

Weather derivatives began in 1997 as trades between US energy companies. In Japan, financial institutions were able to handle them after a 1998 law, which the report calls the Financial System Reform Act, took effect, and the first trade was completed in June 1999. Regional banks also began offering them in 200113.

Use differed between the United States and Japan. In the United States, temperature trades between power and gas companies were the core of the market, individual deals were large, and many used standardized contract terms. In Japan, as of the March 2002 report commissioned by the JMA, small and medium-sized companies in many industries were making small trades using indexes such as rainfall and the number of snowy days as well as temperature, and small standardized products were also spreading14. A 2007 paper also describes domestic trades as mainly tailor-made, with terms set for each customer15.

The users are industries whose sales volumes or customer numbers change with the weather. A summary of an interview with an insurer, published by the Ministry of the Environment, lists ski resorts, golf courses, beverage makers, power and gas companies, apparel makers, and agriculture among the users16. What temperature changes is not the price per unit but the volume of sales or demand, so weather derivatives cover volume swings that interest rate and currency derivatives cannot17. Power futures on electricity, whose demand changes with temperature, are explained in Power.

The difference from insurance is whether the loss is assessed

Insurance assesses the loss that actually occurred and covers the amount up to the policy limit. A weather derivative pays a fixed amount when the observations meet the conditions, without assessing whether there was a loss or how large it was. With no assessment process, payment arrives quickly, but it does not always match the actual loss18. A weather derivative can cover not only damage from natural disasters but also lower sales or higher costs caused by heat or cold19.

Under the Financial Instruments and Exchange Act, figures from weather observations published by the JMA and others are "financial indicators". Transactions such as paying and receiving money based on the difference between an agreed value and the actual value of a financial indicator are OTC derivative transactions20. On the other hand, the same interview summary gives examples in which products with almost the same content as those sold in Japan as weather derivatives are treated as weather index insurance in Thailand, India, and France21.

Basis risk (the gap between the loss and the payout)

Because a weather derivative pays based only on observations, the amount received can differ from the actual loss. The risk of this gap is called basis risk22. A company may have a loss but receive nothing because the index does not meet the conditions, or it may receive a payout because the index meets the conditions even though it had no loss23.

The strike and the payment cap also create gaps (the 950°C day and 600°C day rows in the table for Company G).

Weather futures listed on a US exchange

According to a filing that CME submitted to the US Commodity Futures Trading Commission (CFTC) in May 2024, futures and options on weather indexes are to be listed on CME. The indexes are HDD, CDD, and the Cumulative Average Temperature (CAT), which adds up the daily average temperatures over the period, with contracts for individual cities in the United States, Europe, Tokyo, and elsewhere24. Since there is nothing to deliver for weather, the futures are cash-settled at expiry using the value of the index. The options, when exercised, become futures positions25.

The terms of the seasonal weather futures that CME filed for listing in the same filing are as follows26.

Item US cities (HDD, CDD) Tokyo (CAT)
Contract value Index × USD 20 Index × 2,500 yen
Daily average temperature Average of each day's maximum and minimum temperatures Average of 24 hourly observations from 1:00 a.m. to midnight the next day
Observations US National Weather Service JMA
Settlement Cash settlement Cash settlement

The main users are energy companies whose sales volumes change with temperature, and insurers and reinsurers also trade, as do hedge funds and pension funds looking for assets whose prices have little relationship to their other holdings27.

Notes

  1. Japan Meteorological Agency, "Ⅲ．天候リスクマネジメントによる企業価値向上" (III. Increasing corporate value through weather risk management), in "企業の天候リスクと中長期気象予報の活用に関する調査 報告書" (Report of the study on corporate weather risk and the use of medium- and long-range weather forecasts), March 2002, pages 62 and 64 ("天候デリバティブの仕組み" (How weather derivatives work)) (in Japanese). ↩

  2. Ministry of Land, Infrastructure, Transport and Tourism, "環境負荷に係る検討について（4.環境負荷に関する地域差の検討（その３））" (Study of environmental load: 4. Regional differences in environmental load, part 3), Table 4.4 and notes 5 and 6 ("冷房デグリーデー" (cooling degree days), "暖房デグリーデー" (heating degree days), "(℃日)" (°C days)) (in Japanese). ↩

  3. The 18°C example and the average of the maximum and minimum temperatures are from Japan Meteorological Agency, "Ⅲ．天候リスクマネジメントによる企業価値向上" (in Japanese), pages 64–66. The 65°F base, the average of the maximum and minimum temperatures, and counting in Fahrenheit are from Chicago Mercantile Exchange Inc., "CME Submission No. 24-141 (43 of 144)" (May 2024), Exhibit B (Chapter 405, Rule 40501). ↩

  4. Japan Meteorological Agency, "Ⅲ．天候リスクマネジメントによる企業価値向上" (in Japanese), pages 64–65, and Chicago Mercantile Exchange Inc., "CME Submission No. 24-141 (43 of 144)", Exhibit B (Chapter 405, Rule 40501, "Heating Degree Days Strips"). ↩

  5. Japan Meteorological Agency, "Ⅲ．天候リスクマネジメントによる企業価値向上" (in Japanese), pages 64–66 (example of CDD for Nagoya from July 1 to September 30). ↩

  6. Japan Meteorological Agency, "Ⅲ．天候リスクマネジメントによる企業価値向上" (in Japanese), pages 64 and 67 ("複合インデックス" (Composite indexes)). ↩

  7. Ministry of the Environment, "保険会社（株式会社損害保険ジャパン）へのヒアリング結果要旨" (Summary of an interview with an insurer (Sompo Japan Insurance Inc.)), page 1 ("インデックスと当該企業の収益等の相関を分析" (analyze the correlation between the index and the company's earnings)) (in Japanese), and Ayako Yokouchi, "農業経営における天候リスクマネジメント技術導入手法" (in Japanese), page 228 (Table 3, "過去の気象データと財務・経営データを用いた相関解析" (correlation analysis using past weather data and financial and management data)). ↩

  8. Ayako Yokouchi, "農業経営における天候リスクマネジメント技術導入手法" (Introduction of Weather Risk Management Technology in Farm Management), page 228 (Table 3) and page 231 ("観測点 アメダス岩見沢" (Station: AMeDAS Iwamizawa) in Table 5) (in Japanese). The description of AMeDAS is from page 63 of Japan Meteorological Agency, "Ⅲ．天候リスクマネジメントによる企業価値向上" (in Japanese). ↩

  9. Ayako Yokouchi, "農業経営における天候リスクマネジメント技術導入手法" (Introduction of Weather Risk Management Technology in Farm Management), Agricultural Information Research, Vol. 16, No. 4, 2007, page 231 ("天候デリバティブ取引条件の設定" (Setting the terms of a weather derivative trade): strike value, unit amount, and payment limit) (in Japanese). ↩

  10. Japan Meteorological Agency, "Ⅲ．天候リスクマネジメントによる企業価値向上" (in Japanese), page 66 (the example of a Put Spread trade on Nagoya CDD, and footnote 4, "天候デリバティブでは最も一般的" (the most common in weather derivatives)). ↩

  11. Japan Meteorological Agency, "Ⅲ．天候リスクマネジメントによる企業価値向上" (in Japanese), pages 77–78 ("冷夏・猛暑リスクの交換を図る対策の検討と評価" (Study and evaluation of measures to exchange the risks of cool and very hot summers)). ↩

  12. Japan Meteorological Agency, "Ⅲ．天候リスクマネジメントによる企業価値向上" (in Japanese), pages 77–80 ("東京ガス（株）との気温リスク交換契約とその結果" (The temperature risk exchange contract with Tokyo Gas and its results), the table "気温リスク交換契約の概要と結果" (Outline and results of the temperature risk exchange contract) on page 79, and "気温スワップの効果及び評価" (Effects and evaluation of the temperature swap) on page 80). This part of the report is a presentation by a study group member from Tokyo Electric Power Company. ↩

  13. Japan Meteorological Agency, "Ⅲ．天候リスクマネジメントによる企業価値向上" (in Japanese), pages 62–63. ↩

  14. Japan Meteorological Agency, "Ⅲ．天候リスクマネジメントによる企業価値向上" (in Japanese), pages 63–64 ("日本における天候デリバティブ市場" (The weather derivatives market in Japan)). ↩

  15. Ayako Yokouchi, "農業経営における天候リスクマネジメント技術導入手法" (in Japanese), page 228 ("天候リスクのヘッジスキームモデル" (Model of a weather risk hedging scheme)). ↩

  16. Ministry of the Environment, "保険会社（株式会社損害保険ジャパン）へのヒアリング結果要旨" (in Japanese), page 1 ("保険会社における天候デリバティブ等の取組実態" (Insurers' activities in weather derivatives and related products)). The users are those named in the interview with the insurer. ↩

  17. Japan Meteorological Agency, "Ⅲ．天候リスクマネジメントによる企業価値向上" (in Japanese), page 68 ("企業における活用例" (Examples of use by companies)). ↩

  18. Ayako Yokouchi, "農業経営における天候リスクマネジメント技術導入手法" (in Japanese), page 227 (the text and Table 2, "天候デリバティブのメリット・デメリット" (Advantages and disadvantages of weather derivatives)). ↩

  19. Ayako Yokouchi, "農業経営における天候リスクマネジメント技術導入手法" (in Japanese), page 227 ("補償範囲が広範" (broad coverage) in Table 2). ↩

  20. "金融商品取引法" (Financial Instruments and Exchange Act, in Japanese; English translation: Japanese Law Translation), Article 2, paragraph 25, item 2 (financial indicators), and paragraph 22, items 2 and 4 (over-the-counter derivatives transactions). ↩

  21. Ministry of the Environment, "保険会社（株式会社損害保険ジャパン）へのヒアリング結果要旨" (in Japanese), page 1 (treated as weather index insurance in Thailand, India, and France). ↩

  22. Ministry of the Environment, "保険会社（株式会社損害保険ジャパン）へのヒアリング結果要旨" (in Japanese), page 1 ("保険とデリバティブの主要な違い" (Main differences between insurance and derivatives)). ↩

  23. Ayako Yokouchi, "農業経営における天候リスクマネジメント技術導入手法" (in Japanese), pages 227–228 (Table 2, and "ヘッジ不完全" (incomplete hedge) and "オーバーヘッジ" (over-hedge)). ↩

  24. Chicago Mercantile Exchange Inc., "CME Submission No. 24-141 (43 of 144)", the list of contracts to be listed and Exhibit A ("Cash Market Overview"). ↩

  25. Chicago Mercantile Exchange Inc., "CME Submission No. 24-141 (43 of 144)", page 12 ("Settlement Type", "Futures Final Settlement") and Exhibit A ("no deliverable supply"). ↩ ↩2

  26. Chicago Mercantile Exchange Inc., "CME Submission No. 24-141 (43 of 144)", pages 11–12 ("Trading Unit"), Exhibit B (Chapter 405, Rule 40501), and Exhibit D (Chapter 412, Rule 41201). ↩ ↩2

  27. Chicago Mercantile Exchange Inc., "CME Submission No. 24-141 (43 of 144)", Exhibit A ("Cash Market Overview", "Weather Market Overview"). ↩

  28. Chicago Mercantile Exchange Inc., "CME Submission No. 24-141 (43 of 144)", Exhibit B (the list of observation sites in Chapter 405, Rule 40501) and Exhibit A. ↩