Printer Friendly

Economic benefits of a restored oyster fishery in Chesapeake Bay.

ABSTRACT A simple reduced form inverse demand model is used to determine the price flexibility of Chesapeake Bay oyster production when other producing regions of the United States have their oyster production held at average levels for 2002 to 2006. With the current reduced state of oyster demand compared with earlier periods, a restored Chesapeake oyster fishery that could sustain annual harvests of 4.9 million bushels would result in a predicted price that is significantly lower than the minimum observed from 1950 to 2006 and presumed to be unprofitable. A survey of oyster industry participants suggests that an equilibrium price is more likely to be around $20.06 per bushel. Using the inverse demand model, the predicted level of production is about 3.2 million bushels. Industry members believe that a restored industry, based on Crassostrea ariakensis, would have a greater percentage of oysters going to the shucked market compared with Crassostrea virginica.

KEY WORDS: economics, demand, restoration, oysters, Suminoe oyster

BACKGROUND

As part of the deliberations over introducing the Suminoe oyster (Crassostrea ariakensis [Fujita 1913]) to the Chesapeake Bay, the United States Congress mandated the preparation of programmatic Environmental Impact Statement (EIS). The notice of intent for the EIS states that the goal is to evaluate oyster restoration alternatives to restore the Chesapeake Bay oyster population to a level that could sustain harvests comparable to those in the 1920 to 1970 time period. (Federal Register: January 5, 2004 (Volume 69, Number 2)] [Notices] [Page 330 332] From the Federal Register Online via GPO Access [wais.access.gpo.gov].) Historical harvest figures indicate that the annual average harvest, computed on a decadal basis, of Chesapeake Bay oysters from 1920-1969 ranged from 3.6 million bushels to 5.8 million bushels, and averaged 4.9 million bushels. Presumably, this is the harvest target range of the proposed restoration. Although these harvest figures are mostly from the period prior to the onset of outbreaks of the protozoan parasites MSX (Haplosporidium nesloni) and then Dermo (Perkinsus marinus) in Chesapeake Bay, and the period when the Maryland oyster repletion program began, there is little reason to believe that these harvest levels were biologically self-sustaining. Nevertheless, we use this figure of a 4.9 million bushel annual harvest in the following analysis of the commercial oyster industry, recognizing that the population necessary to sustain such a harvest may be significantly greater than the one that existed during the 1920 to 1969 period.

To achieve the stated goal, the EIS proposes eight alternatives, including a no action alternative as well as a combination of alternatives. The remaining six proposed actions include:

1. Expansion of the existing native oyster restoration program.

2. A harvest moratorium with compensation for fishermen.

3. Aquaculture of native oysters.

4. Aquaculture of triploid nonnative oysters.

5. Introduction of an alternative strain of Crassostrea ariakensis or an alternative nonnative species.

6. Introduction of diploid Crassostrea ariakensis and termination of native oyster restoration programs.

In this paper we analyze the potential direct economic benefits of oyster restoration. Direct benefits are those related to the oyster fishery. Ecological and other benefits and costs will also be considered in the EIS, but are not dealt with here.

DIRECT BENEFITS--THE CHESAPEAKE BAY OYSTER FISHERY

The United States Oyster Market

The Committee on Normative Oysters in Chesapeake Bay report for the National Research Council (2004) does an excellent job of reviewing the current and historical role of the Chesapeake Bay oyster fishery in the US oyster market. We briefly review and update the supply and demand situation here, but refer the reader to that source as well as Lipton & Kirkley (1994) and Muth et al. (2000) for greater detail.

Oyster Production

As can be seen in Figure 1, the US supply of oysters (including both Crassostrea virginica Gmelin 1791 and Crassostrea gigas Thunberg 1793) has fallen by almost 60% from around 80 million pounds of meats in the early 1950s to less than 33 million pounds in 2006. (A few other oyster species are harvested in the United States, but the level of landings is inconsequential compared to C. virginica and C. gigas.) Figure 2 uses the same landings data to show the market share in terms of production provided by each region of the country. The small growth in Gulf of Mexico and Pacific landings with a declining total harvest translates into large increases in market share for those two regions. In 2006, these two regions supplied 93% of the market.

Oyster Demand

There is a limited number of studies of oyster demand (see Lin et al. 1991, Berry 1992, Keithly & Diop 2001) and few of recent vintage to provide meaningful information that can be used to predict the consumer response to a revitalized Chesapeake Bay oyster fishery. As can be seen in Figure 3, one gets a different picture of price responses in the oyster fishery depending on whether they are looking at nominal or real prices. Nominal prices show a steady increase over the 1950 to 2006 period. However, when adjusting for inflation, we see a stable real price of around $3.00 per pound of meats, except for a brief period of high prices from 1987 to 1992. (Prices are adjusted by the consumer price index for all urban consumers with the 1982-1984 = 100 series and then adjusted to 2006 dollars (source: http://www.bls.gov). A consumer index rather than a producer index is used because in Chesapeake Bay, oyster revenues represent direct income to fishermen.)

[FIGURE 1 OMITTED]

The increase in aggregate oyster prices from 1987 to 1992 shown in Figure 3 corresponds with the period of increased production and market share expansion for New England oysters. New England oysters are mainly targeted for the higher value half shell market. In 1986, New England oyster prices rose 40% over the previous year's level. This higher price was followed by an increase in New England production. By 1993, production in New England hit an all time high for the study period of 8.4 million pounds of meats and a 22% share of the market by volume. At this peak level of New England production, New England oyster prices had dropped to 42% of the 1986 high. Subsequently, New England production fell from its peak of over 8 million pounds of meats to less than half a million.

[FIGURE 2 OMITTED]

[FIGURE 3 OMITTED]

The large decline in Chesapeake Bay oyster production over the last 56 y and the lack of a response in real price to this decline provides strong evidence that the demand for oysters has also declined during the period. If demand had not declined, prices would have risen because of the increased scarcity. This notion is supported by the analysis of consumer survey data presented in Lipton & Kirkley (1994) and detailed in Berry (1992) that compared oyster consumption in three different periods using three different food consumption surveys for 1977, 1980, and 1987. (These are the USDA National Food Consumption Surveys of 1980-1981 and 1987-1988 and the 1981 NMFS seafood consumption survey.) These studies do not test for factors that led to the decline in demand, but speculate that this is a reflection of changes in tastes and preferences of US consumers.

A Model of Inverse Demand for Chesapeake Bay Oysters

Estimating the inverse demand for oysters (i.e., price as a function of quantity) can help in making projections about future oyster prices under different scenarios by providing price flexibility estimates. [See the recent work on inverse demand systems for fish by Park et al. (2004).] The inverse demand relationship can also be used to estimate changes in consumer surplus, an approximation of the welfare measure used in benefit cost analysis.

An inverse demand model is specified as:

[P.sub.CB] = [alpha] + [delta]D + [[beta].sub.CB] [H.sub.CB] + [[beta].sub.MA] [H.sub.MA] + [[beta].sub.NE] [H.sub.NE] + [[beta].sub.SA] [H.sub.SA] + ([[beta].sub.G] + [[beta].sub.VV] [D.sub.VV])[H.sub.G] + [[beta].sub.PC] [H.sub.PC] Eq. (1)

where P is the ex-vessel price per pound of meat and H is reported landings subscripted by the region: CB = Chesapeake, MA = Mid-Atlantic exclusive of Chesapeake Bay; NE = New England; SA = South Atlantic; G = Gulf of Mexico and PC = Pacific coast. D corresponds to a dummy variable that takes the value 0 for the 1950 to 1978 period and 1 for the 1979 to 2006 period. The dummy variable is used to capture a structural shift in demand as suggested by the Berry (1992) study. There is also a binary dummy variable subscripted Vv, corresponding to the 1991 to 2006 period when California required warnings regarding consumption of Gulf of Mexico oysters because of the risk of the bacterial pathogen Vibrio vulnificus. Keithly and Diop (2001) found that the relationship of Gulf of Mexico oyster prices and Chesapeake prices dramatically shifted as a result of this concern. The subscripted [alpha], [delta], and [beta] correspond to model parameters to be estimated using ordinary least squares regression. The model is estimated for the 1950 to 2006 landings and value data available from the National Marine Fisheries Service. [Data available at (http://www.st.nmfs.noaa.gov/stl/commercial/index.html.)]

Results from the inverse demand model estimation are given in Table 1. The model has strong predictive capability with an [R.sup.2] of 0.80. All model parameter estimates are significantly different from zero at the 95% confidence level except for New England landings, which is not significant and the Vibrio vulnificus dummy variable, which is significant at the 90% level. Chesapeake, Gulf, and Mid-Atlantic landings negatively effected Chesapeake price, whereas, South Atlantic and Pacific prices had a positive and significant effect. The dummy variable parameter was negative and significant indicating a downward structural shift in demand when comparing the pre-1979 period to the 1979 to 2006 period.

Price flexibility, the inverse of elasticity, provides an estimate of the ratio of the effect of a percentage change in quantity on the percentage change in price. Calculated at the mean of the data used in the estimation, the price flexibility for Chesapeake Bay oyster production is -0.37. Thus, a 1% increase in Chesapeake Bay production will result in a 0.37% reduction in Chesapeake Bay price. The current situation is nowhere near the mean of the data. Thus, using this estimate of flexibility would not be reflective of expected changes from current conditions. The mean annual Chesapeake Bay catch for the 1950 to 2006 period is around 17 million pounds, whereas the 2006 harvest was only about 289 thousand pounds. Mean Chesapeake Bay price per pound is $3.91 and the 2006 price was $4.66.

To simulate the effect on Chesapeake Bay price of a large increase in local production from current levels we calculated a predicted price based on the last five years of data (2002 2006) for each of the areas of harvest. The predicted Chesapeake Bay price per pound of meats, using Chesapeake Bay harvests equal to the five year average is $4.67 per pound (in 2006 dollars).

To predict the ex-vessel price for Chesapeake Bay oysters under a restored fishery scenario, we convert the presumed target level of 4.9 million bushels to pounds of meat using an assumption of 7 pounds of meats per bushel. Thus, Chesapeake production would be 34.3 million pounds of meats. Keeping all other area's production constant at the 2002 to 2006 level, the predicted Chesapeake Bay price resulting from this increase is $1.93 per pound of meats, which translates back to an ex-vessel price of $13.49 per bushel.

Because cost data are not collected for the oyster fishery, it is difficult to comment on the supply (harvest) response to an increasing stock of oysters in the Chesapeake Bay. However, an ex-vessel price of $13.49 per bushel in 2006 dollars is significantly below any historical low price for oysters and it is reasonable to assume that harvesting oysters at this price would not be profitable. The historical low price in Chesapeake Bay was in 1974 when the price per bushel in 2006 dollars was about $20. The historical decline in demand estimated by the model and the increase in quantity supplied from other regions result in a harvest quantity from the Chesapeake region that was once an observed market outcome to no longer be feasible. For example, if we use 1950 demand conditions and the 1950 to 1955 average quantity supplied from other regions, the predicted Chesapeake price would be $22.89 per bushel, presumably a price level that would be profitable to fishermen. Thus, although the restoration of a population of oysters sufficient to sustain harvests at the 4.9 million bushel level may be technically feasible; it is not economically feasible under projected market conditions and we would not expect a viable fishery at that level. It should be noted that this simple price response model does not capture the full suite of market interactions between regions. A more sophisticated multimarket model would need to be developed to reflect how increased supply from the Chesapeake region will affect prices and thus, production from other regions. Here, we have kept other region production constant. Nevertheless, ultimately the market place and consumer demand will determine the total harvest and price of oysters from Chesapeake Bay. The market will tend toward an equilibrium price and harvest that equates prices across regions, although accounting for quality differences (e.g., perceived safety) and transportation costs.

An Industry-Informed Scenario of a Restored Oyster Fishery

We surveyed the oyster industry to gain insight into industry members' beliefs about the market that will result from a restored oyster fishery in Chesapeake Bay. Surveys were mailed or hand delivered to 16 oyster dealers in Virginia and 12 dealers in Maryland. Responses were received from 10 Virginia oyster dealers and 5 Maryland dealers. The information gained from this survey can be used to understand the beliefs held by key industry leaders regarding the economic consequences of particular actions taken in regard to the oyster fishery. This information can be used to:

(1) Provide an informed but biased prediction of potential industry status;

(2) Understand the motivation for the level of industry support for a particular proposed action;

(3) Compare with the results from the statistical modeling of the oyster market.

The first survey question attempted to determine what oyster dealers felt they would have to pay to oyster growers or fishermen for either C. virginica or for C. ariakensis if Chesapeake Bay production of either oyster was equal to the stated goal (4.9 million bushels). With the limited knowledge they have about the market for C. ariakensis', Virginia dealers felt on average that it would sell at a premium relative to C. virginica; whereas, Maryland dealers felt that the native oyster would sell at a premium. The Virginia results are skewed by two responses (small sample size because of the small size of the extant industry is a problem for this entire part of the analysis) where the dealers guessed that C. ariakensis would have a huge premium over C. virginica. Seven out of 10 Virginia dealers felt that they would pay the same or a lower price for C. ariakensis than they would for C. virginica, and all the Maryland dealers felt the C. ariakensis price would be the same or lower than C. virginica.

Responses to the first question also provided an indication of industry beliefs regarding price responsiveness to the increased Chesapeake Bay production at the harvester level. For dealers in both states, they felt that oyster prices to harvesters would decline about 22% from current levels of $25.61 per bushel to $20.06 a bushel for a restored C. virginica fishery. (The survey was conducted in 2004, prices have been inflated to 2006 dollars using the consumer price index.) The median response for most likely price for both species was $21.34 a bushel.

The percentage of product that goes to the higher value halfshell market will have a significant effect on industry revenues. We asked the dealers participating in the survey what they thought was the current size of the halfshell market for oysters harvested from Chesapeake Bay or for shellstock imported into the Bay region for distribution. There was a wide variation by dealer in response to this question, but the median response was that 35% of the current Chesapeake Bay oyster market is for the halfshell market.

We asked the industry representative what they thought a restored oyster fishery market would look like in terms of halfshell versus' shucked product. The median response was that a C. virginica fishery restored to historical harvest levels would consist of a 30% halfshell market. On the other hand, they thought a C. ariakensis based fishery, would be about 22% half shell.

We polled the industry on how their output prices would change as a result of a restored fishery based on either C. virginica or C. ariakensis. We got responses for prices for shucked and halfshell oysters and whether they were "wild" or "cultured." A summary of the median responses is given in Table 2. Industry expectations are that there will be no difference in price between the species for the shucked product, but C. virginica will have approximately a $2.13 to $2.67 premium per hundred count over the C. ariakensis price regardless of whether the product is harvested from the wild or cultured. Compared with Fulton Fish Market wholesale prices, the $40.55 price per gallon of shucked oysters represents a 21% decline in price. Halfshell market prices are dependent on where the product is from. No recent prices for Chesapeake halfshell oysters were available, but the $18.67 to $21.34 price range is below the price of $42.68 per hundred count for Connecticut cultured oysters and above the $15.47 price for Gulf of Mexico wild oysters.

We also asked the industry to provide their estimates on how their profitability might vary depending on the species and product form. Whereas not equal to economic rent and the appropriate welfare measure that we seek in a benefit cost analysis of the proposed actions, the responses do provide a relative indicator of the industry's perception of what the potential benefits may be. Their responses indicated expected differences in profits between species for a shucked product, but not for a halfshell product (Table 3).

From the earlier mentioned survey, we can aggregate the data to develop a scenario of what industry experts believe a restored Chesapeake Bay fishery might look like. We are assuming that the industry will reach a state where it is harvesting 4.9 million bushels a year from the Chesapeake Bay. The industry believes that regardless of the species used, this will translate into an ex-vessel value or gross fishermen/grower income of $94.1 million. This is 41% lower than we would estimate using 2006 prices, so there is some acknowledgment in the industry responses that an increase in supply of oysters will lead to some decrease in price.

According to the survey participants, the difference between an industry, based on C. virginica versus C. ariakensis, starts to manifest itself in the difference regarding halfshell versus shucked product, with 30% and 22% going to the halfshell market for each species, respectively. The implication is that there will be a different value and profit at the wholesale level. To calculate the revenues and expected profits, we assume production of 1 gallon of oysters from a bushel and a 225 oyster count to a bushel. Thus C. virignica, under this scenario yields at the wholesale level $138.7 million worth of shucked product and $70.4 million worth of oysters for halfshell. The figures are slightly different for C. ariakensis with $154.7 million for the shucked market and $45.9 million for the halfshell market. Whereas revenues are slightly higher for C. virginica, profits are higher for C. ariakensis with profits of $42.7 million compared with $39.5 million.

In summary, given limited direct knowledge regarding C. ariakensis and the large amount of uncertainty in projecting changes caused by large scale shifts in the oyster market, industry experts see little difference in an industry based around C. ariakensis compared with C. virginica. This industry analysis does not address the time frame or the cost necessary to create this new market level. Fishermen or grower incomes and processing industry profits reported here are not welfare benefits that can be used in a benefit/cost analysis of the proposed actions.

Combining the Industry Scenario With the Inverse Demand Analysis

The earlier mentioned industry scenario appears reasonable in foreseeing a 22% decline in ex-vessel oyster prices with a restored oyster fishery, but the results are more optimistic than the inverse demand analysis, which predicted a 59% price decline. One possibility is that the industry has realistic expectations about what are sustainable ex-vessel prices relative to the upstream market they face and their costs of processing and handling oysters, but not a good understanding of the aggregate size of the market that will lead to those prices.

A reasonable approach would be to take the industry price predictions and then determine what size fishery would result from those prices. To do this, we used our inverse demand equation and solved for Chesapeake Bay landings as a function of predicted price and the harvest in all other regions at the average for the 2002 to 2006 period. Thus, the $20.06 price per bushel for either species would translate into an ex-vessel price of $2.87 per pound of meats. (This is one area where there may be a significant difference in impacts between the two species if C. ariakensis has a higher meat yield per bushel than C. virginica.) Solving for the quantity that results in this price, yields an estimate of Chesapeake Bay landings of about 27 million pounds of meats or 3.2 million bushels. At this level, Chesapeake Bay fishermen and producers gross income would be $65.1 million, significantly less than the industry prediction of $94.1 million, but significantly more than the 2006 income of under $1.3 million. Table 4 summarizes the scenario, carrying forward the analysis to the next market level, based on the industry scenario of wholesale prices.

CONCLUSION

In addition to all the caveats mentioned above, there are a number of other issues that need to be considered related to the benefits of a restored commercial oyster fishery in Chesapeake Bay. For example, the earlier mentioned analysis does not reveal a large distinction in benefits between fisheries based on C. virginica versus C. ariakensis. The only major distinction taken into account earlier is because of a slight difference in industry beliefs about the size of the halfshell market for each species. If there is a major difference in demand for the two products, particularly if there are true taste and quality differences, this could have an impact on the results. Other factors that might impact results include differentials in the costs of harvesting and product yields.

Another important factor to considering when weighing the potential benefits of a restored fishery is the time frame over which the benefits might accrue. It will take many years to restore oyster resources in Chesapeake Bay to a level that would support the level of a sustainable fishery we anticipate. Because much of the expense of the proposed alternatives will occur early in the process, this timing will have an impact on calculation of benefits. Lipton et al. (1992) discuss the role of the timing of benefits and costs related to the potential introduction of C. gigas into Chesapeake Bay. Because the data suggests that there might not be a significant difference in commercial fishery benefits, once the industry has been fully restored, regardless of the alternative chosen, the greatest difference among alternatives in this regard may be the timing of restoration. Thus, everything else being equal, from the viewpoint of the commercial fishing industry and oyster consumers, the alternative that restores oysters the fastest will have the highest net benefits.

LITERATURE CITED

Berry, P. A. 1992. The retail demand for oysters: a study of fresh and canned oysters and oyster stew. Master's Thesis. Department of Agricultural & Resource Economics, University of Maryland College Park, MD. 98 pp.

Keithly, W. R. & H. Diop. 2001. The demand for Eastern oysters, Crassostrea virginica, from the Gulf of Mexico in the presence of Vibrio vulnificus. Mar. Fish. Rev. 63:47-53.

Lin, C. T., J. W. Milon & E. Babb. 1991. Determinants of subjective food safety perceptions: A case study of oysters in the southeast. Journal of Agribusiness Spring 1991:71-85.

Lipton, D. W., E. Lavan & I. E. Strand. 1992. Economics of mollusk introductions and transfers: The Chesapeake Bay dilemma. J. Shellfish Res. 11:511-519.

Lipton, D. & J. Kirkley. 1994. A Profile of the oyster Industry: Northeastern United States. Virginia Sea Grant Marine Advisory No. 54 VSG-94-08.72 pp.

Muth, M. K., D. W. Anderson, S. A. Karns, B. C. Murray & J. L. Domanico. 2000. Economic impacts of requiring post-harvest treatment of oysters. Research Triangle, NC: Research Triangle Institute report for interstate shellfish sanitation conference. 97 pp.

National Research Council. 2004. Nonnative Oysters in the Chesapeake Bay. The National Academies Press. Washington, DC. 325 pp.

Park, H., W. N. Thurman & J. E. Easley, Jr. 2004. Modeling inverse demands for fish: empirical welfare measurements in Gulf and South Atlantic fisheries. Mar. Res. Econ. 19:333-351.

DOUGLAS LIPTON, Corresponding author. E-mail: dlipton@arec.umd.edu

Department of Agricultural & Resource Economies, University of Maryland, College Park, Maryland
TABLE 1.
Results from Chesapeake Bay inverse oyster demand model.

 Standard
 Coefficients Error t-statistic

Intercept 4.198857 0.488164 8.601327 **
Chesapeake -7.5E-08 1.12E-08 -6.69841 **
Mid-Atlantic -5.4E-08 1.56E-08 -3.46849 **
New England 2.6E-09 3.675E-08 0.07116
South Atlantic 3.74E-07 7.08E-08 5.291183 **
Gulf of Mexico -5.1E-08 1.42E-08 -3.52745 **
Pacific Coast 1.6E-07 2.78E-08 5.740317 **
Dummy -0.57923 0.170346 -3.1033 **
Vibrio vulnificus -1.4E-08 8.48E-09 -1.70666 *

[R.sup.2] = 0.80; n = 55; ** significant at
95% confidence level; * 90% confidence level.

TABLE 2.
Median responses of oyster industry experts for wholesale
oyster prices resulting from a restored fishery.

 C. virginica C. ariakensis

Shucked, price per gallon $38.00 $38.00
Wild, price per hundred $20.00 $18.00
Cultured, price per hundred $20.00 $17.50

TABLE 3.
Industry expectations of profits based on product form.

 Product Form C. virginica C. aviakensis

Shucked (wild), profit per gallon $6.50 $8.00
Shucked (cultured), profit per gallon $6.50 $7.00
Halfshell (wild), profit per hundred $5.00 $5.00
Halfshell (cultured), profit per hundred $5.00 $5.00

TABLE 4.
A scenario for a sustainable oyster industry in Chesapeake Bay.

 Harvest Shucked Halfshell
 (million Value Value Value
Species bushels) ($ million) ($ million) ($ million)

C. virginica 3.2 $65.1 $85.1 $43.2
C. ariakensis 3.2 $65.1 $94.8 $28.5
COPYRIGHT 2008 National Shellfisheries Association, Inc.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2008 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Lipton, Douglas
Publication:Journal of Shellfish Research
Article Type:Report
Geographic Code:1USA
Date:May 1, 2008
Words:4519
Previous Article:Post-settlement survival and growth of the suminoe oyster, Crassostrea ariakensis, exposed to simulated emersion regimes.
Next Article:Establishing a baseline for management of the rock scallop, Spondylus calcifer (Carpenter 1857): growth and reproduction in the upper Gulf of...
Topics:


Related Articles
Long-term projections of eastern oyster populations under various management scenarios.
A comparison between a suction dredge and a traditional oyster dredge in the transplantation of oysters in Delaware Bay.
Fisherman choice and incidental catch: size frequency of oyster landings in the New Jersey oyster fishery.
Consumer rating of the Suminoe oyster, Crassostrea ariakensis, during home cooking.
Preface.
Suminoe oyster (Crassostrea ariakensis) culture in Korea.
Crassostrea ariakensis in Chesapeake Bay: growth, disease and mortality in shallow subtidal environments.
Long-term trends in oyster population dynamics in Delaware Bay: regime shifts and response to disease.
A comparison of the macrofaunal communities inhabiting a Crassostrea virginica oyster reef and oyster aquaculture gear in Indian River Bay, Delaware.
Prospect of oyster culture in Pakistan: pathology assessment of two commercially important oyster species.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters