World Wood Price Methodology
RISI publishes two separate price estimates for the world pulp market. One, which traditionally has been published in World Pulp Monthly, reports a single point estimate for the prices of major grades of wood pulp in North America, Europe, and Asia. The other, traditionally published in PPI This Week (Europe), Pulp & Paper Week (North America) and PPI Asia News (Asia), is a range estimate (high and low, which are occasionally the same) within which the large majority of transactions are believed to have occurred. This document describes the methodology used for these price estimates, herein referred to as the Point Estimate (PE) and the Range Estimate (RE).
The PE and RE surveys were developed independently when the respective publications were published by different companies. Going forward, news editors and economists will share and compare information regarding survey results, thus increasing the amount of information used to determine both the PE and the RE. This sharing of information will be limited, however, in cases in which a confidential relationship exists between the surveyor and a respondent. The RE survey is conducted by news editors associated with PPI This Week, Pulp & Paper Week, and PPI Asia News The PE is set by economist associated with World Pulp Monthly, based on all of the information collected as part of the RE survey, plus additional survey information gathered independently by the economist.
Determination of Point Estimates and Range Estimates
All price discovery methods, no matter how statistically rigorous they claim to be, run the risk of being "gamed" or undermined by survey responders, through means such as selective reporting of transactions or misreporting of prices. Even a process that is based on a weighted average of actual invoices can be manipulated if respondents report transactions selectively (that is, non-randomly).
RISI's price reporting methodology addresses this problem in two ways. First, prices are collected from both buyers and sellers in an attempt to balance out potential bias from respondents. Second, our methodology allows the news editor or economist who is doing the survey (hereafter, "the surveyor") a small degree of latitude to supplement the survey results with other information in order to arrive at the price that best represents market conditions. The surveyor, for example, can use his or her best judgment in assigning more weight to responses from those survey participants who have been the most reliable sources for pricing in the past. That the surveyor is allowed judgmental latitude in estimating prices is an especially important point to note for the PE survey---point estimates are not arrived at using a fixed calculation or weighting scheme. (Larger transactions are given more weight than are smaller ones, however, which is consistent with the concept of a weighted average.) Application of expert judgment based on years of reporting on and studying these markets has been one of the keys to achieving the quality and acceptance of RISI's price estimates.
Where survey information is limited due to small size of the market, or when there are substantial conflicts in the information provided by survey respondents, the surveyor may factor non-survey information into the estimate. Such information can include current supply and demand fundamentals, historical relationships between product prices over time, and information as reported by companies to the media or in financial reports.
Revisions to the prices are made infrequently, in the event of new information becoming available or when there is an ex-post revision or adjustment to prices by the parties. Retroactive rebates, or the invocation of a "baisse" clause, for example, may trigger a price revision. For the surveys that are done more than once a month, if a price change for that month is questionable, it will not be shown until the end of the month or the beginning of the following month, once it has been accepted by both sides of the market. If the price change is realized partially across the market, the RE surveys will reflect this in the ranges chosen. (See "Other Transaction Specifications" below.)
Survey Procedure
Upon joining the survey pool, a respondent is asked to disclose the specification of the products that they actively buy or sell, the average volumes involved, and the nature of the transactions that they engage in (spot or contract, indexed or open market). This information is used by the surveyor to put the respondent's responses into perspective, and is updated periodically.
Most contacts with survey respondents are made by phone. Some are made by email. Participants in the survey include both buyers and sellers of paper. Agents and brokers are included, in most cases as sellers.
The amount and type of information that a respondent provides each period varies across individuals and companies. In the ideal case, the respondent would provide a number of prices for various transactions that occurred during the week or month (depending on the survey) along with the volumes involved in each and any special terms. Most respondents, however, report a single average number or a price range for the time period for each grade of paper that they buy or sell. A small minority will report only changes in price level from the past period's value. All of these responses will be taken into consideration by the surveyor.
The mix of buyers and sellers varies by product, but an effort is made to adequately cover both sides of the market. The number of respondents varies by product but is always sufficient to determine a representative price for the period.
RE transaction prices are collected weekly in Europe, biweekly in Asia, and monthly in North America. Spot prices are collected biweekly for all regions. PE prices are set on a monthly basis, by the last business day of the month.
Key Transaction Specifications
For Europe and North America, price estimates reflect transactions in which the seller is a pulp mill or pulp distributor and the buyer a paper mill. For Asia, and China in particular, the buyer may be a paper mill or an onshore trader/distributor
North American prices include free delivery to the paper mill. European and Asian prices include free delivery to a major port (and are thus CIF or C&F prices) with delivery costs from the port to the paper mill being extra. This is in keeping with the conventions of the market.
Prices for two types of transactions are reported on each continent: "contract/relationship" transactions, and "spot" transactions
Contract or relationship transactions are deals between buyers and sellers who have the expectation of doing business with each other on an ongoing basis in the future. A "contract" transaction typically involves a transaction volume that is agreed to ahead of time, with the price to be negotiated on an ongoing basis, usually month to month. A "relationship" transaction is less formal than a contract transaction, in that there is no fixed volume agreed to ahead of time, and again prices are negotiated on an ongoing basis, usually month to month. Contract and relationship deals are similar in that both involve the expectation of doing business together on an ongoing basis in the future.
Spot transactions involve transaction volumes that are more volatile over time that are contract/relationship transactions. Spot transactions are more opportunistic and do not involve the expectation of continuing business being done between buyer and seller in the future. Spot transactions are generally negotiated on a "net price" basis, in which discounts or rebates are not applied to the quoted price. Spot prices tend to be lower than contract/relationship prices during weak markets, and higher during strong markets.
Three important pricing concepts apply to contract/relationship transactions. The list price is the pre-discount price that is announced (perhaps publicly) by the pulp producer. The transaction price or effective list price is the price that a buyer and seller actually agree upon, before customer-specific discounts are applied. (Customary customer-specific discounts are given for reasons such as fidelity or size of purchase.) Finally, the net price is actual price paid by the buyer, net of all discounts and rebates (except for small end-of-year rebates in some cases).
Unless otherwise noted, RISI pulp price estimates for Europe and North America and parts of Asia reflect transaction prices (or effective list prices) of pulp. In a "sellers' market", transaction prices will generally be the same as list prices. In a "buyers' market", however, transaction prices will generally be lower than list prices. The difference between the announced list and the transaction price may be called a "temporary competitive allowance" that is generally available to most buyers in the market regardless of the size or purchasing pattern. Temporary market allowances are not the same thing as customer-specific discounts, which may be as low as 2-3% in Asia and may range from 5-15% or sometimes higher in North America and Europe.
Other Transaction Specifications
Arms-length transactions: Only transactions between non-affiliated parties are included in the survey. To be included in the survey, a transaction must be conducted at "arms length." Prices for internal transfers between units of a company are excluded from the survey.
First-quality tonnage: Prices are for all first quality tonnage sold.
Distressed transactions: Distressed or unusual transaction are not included in the price.
Timing: For Europe and North America, price estimates for a given period of time generally reflect shipments made in that period, rather than orders placed in that period that will be shipped in future periods. For Asia, RE prices for a given period of time generally reflect orders placed in that period. Asian PE prices generally reflect shipments, not orders.
Locked-in prices: In some cases, when prices are changing, some buyers may be locked in to the previous period's price, either because of contractual reasons or because they are able to negotiate special treatment with suppliers. In the RE surveys, it is common to have the price range reflect both the new price and the "locked in" price, as long as the "locked in" price represents a substantial amount of tonnage transacted in the market. If the "locked in" tonnage is not substantial, the old price will not be shown. In the PE survey, an attempt would be made to approximate the "weighted average" of the new and old price, based on available information, if the tonnage transacted at the old price is substantial.
Terms of payment: Price estimates assume normal and customary payment terms.
Delivery: Prices are delivered prices within a standard delivery zone. Any surcharges that a customer pays for additional freight because of his location or drop-off requirements are not reflected in the estimated price.
Specification changes: These descriptions describe the methodology as of 2006. The prices series involved are up to 20 years or more in length. In the earlier years of these surveys, somewhat different specifications may have applied.
Circularity
RISI price estimates are used within the industry for many purposes, including setting prices for contract transactions in which both parties choose to tie the transaction to an index. RISI is aware that a fraction of transactions in the market may be tied directly to its the point price estimates. In theory, this raises concerns about circularity of the survey process.
For this reason, RISI's policy is to survey only for transactions that are conducted in the open market; i.e., that are not tied to the price index itself. Some respondents may be involved in both open market and "indexed" transactions. In these cases, every attempt is made to use only information relating to open market transactions.
Auditing
Because they involve confidential sources and require the application of the judgment of the surveyor, the surveys are not audited by an outside party. RISI has instituted internal management controls to check on survey methodology and accuracy.
Non-Disclosure of Information
All RISI price surveys are confidential. Names of individuals and companies participating in the survey are never disclosed to the public or to other survey participants. RISI's policy is to neither confirm nor deny either directly or indirectly the participation of any individual or company at any time. Confidential or proprietary information that is collected as part of the survey process, including prices reported by individual survey participants, is not shared with the public or with other survey participants.
Absence of Financial Interest
RISI recognizes that in the normal conduct of their jobs, its employees may have access to confidential or proprietary information regarding companies or markets. RISI prohibits its employees who conduct price surveys from having a financial interest in companies whose values are directly and substantially affected by prices they report. Excluded from this prohibition are holdings in mutual funds or other diversified portfolios that might include financial assets related to such companies. For example, if an employee is involved in a price survey for newsprint, he or she may not directly hold financial instruments issued by newsprint producers or newspapers, because the price of newsprint substantially affects the values of both the producing and the consuming companies. The employee could however invest in mutual funds or other diversified financial instruments that include some of the above-described companies. Also prohibited is a financial interest in any financial instrument that derives its value substantially from the reported prices of products included in the survey.
Commercial Usage
RISI price estimates are copyrighted information. Any commercial use, including hedging or settling of contracts or trades, requires a subscription. For more information, please contact info@risiinfo.com
Disclaimer of Warranty
Although RISI collects information from sources who are believed to be reliable, RISI does not warrant or guarantee the accuracy or completeness of its price estimates. Actual transaction prices will often vary from the estimated level.