Pricing Strategies in Commodity Markets

Author(s): Jerry Metcalf

Source: Interfaces, Vol. 12, No. 4 (Aug., 1982), pp. 57-63
Published by: INFORMS
Stable URL: http://www.jstor.org/stable/25060286
Accessed: 08-12-2017 11:40 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
http://about.jstor.org/terms

INFORMS is collaborating with JSTOR to digitize, preserve and extend access to Interfaces

This content downloaded from 190.213.241.97 on Fri, 08 Dec 2017 11:40:46 UTC
All use subject to http://about.jstor.org/terms
 INTERFACES Copyright ? 1982, The Institute of Management Sciences
Vol. 12, No. 4, August 1982 0092-2102/82/1204/0057$01.25
PRICING STRATEGIES IN
COMMODITY MARKETS
Jerry Metcalf
Weyerhaeuser Company, Tacoma, Washington 98401

Abstract. There are a great many products that do not have the luxury of brand
identification. They are the great unwashed, the commodity products for which there is
virtually no loyalty among their customers. These products are mainly sold on the basis of
price and availability with service and terms being significant but less important factors. A
quick scan of the futures exchanges will provide a list of the major products in this
category. All products are characterized by being traded in markets which are fairly effi
cient and quite volatile in terms of price. It is this volatility, especially on the down side,
which creates the problems on which we hope to shed some light.

Because this is a real-world application, let us stage a situation which may be

uncomfortably familiar. Our hero is a product manager with pricing authority. His
firm is engaged in the business of distributing, say, commodity widgets which are
sold from an inventory. Times have been good, and our hero's inventories, while
turning at a good rate, are on the large side. The plot thickens when he arrives at
work one Monday morning to find that there is a blizzard in the East, floods in the
West, a sharp increase in interest rates, and no customers. The market goes into free
fall, and he soon finds that the market prices of his products are well below their
average inventory cost. Our hero now has a very large problem and an intense desire
to change lines of work. What to do? It does not appear that prices will recover any
time soon, so simply riding it out is not an option.
In this situation, one has three choices of action. We will look at all three and
develop a model for analyzing which alternative would be most appropriate. There
will be no talk of nifty parabolic utility functions; rather, we will look to a basic
trade-off analysis. Our objective is not to push back the frontiers of science but to
communicate a quite practical technique which will be immediately useful.
The three choices of action in this situation are shown below:

Do Nothing. Continue to price at the market and sell at a rate which the
market allows.
Dump. Lower prices (below the market) to stimulate the sales rate and blow
out the high-cost inventory as soon as possible.
Hold Back. Price above the market and accept a lower sales rate. Don't get
down and dirty on the high volume (very low margin) business but hang back
and pick off the lower volume mixed business with better margins.
The goal is to choose the course of action which maximizes the total future
return (or minimizes the loss) to the company. Our hero does not want to fall into the
trap of reacting to short-term conditions in his profit and loss statement. His horizon
is sufficiently long to discount short-term losses if the ultimate return warrants such a
strategy. Our hero is a real hero, isn't he?

MARKETING; INVENTORY/PRODUCTION?POLICIES, PRICING

INTERFACES August 1982 57

This content downloaded from 190.213.241.97 on Fri, 08 Dec 2017 11:40:46 UTC
All use subject to http://about.jstor.org/terms
 In order to understand which course of action makes the most sense, let's look at
the trade-offs. The base case will be the "Do Nothing" case. You really don't "do
nothing," but you don't buck the market. You sell at prices and rates determined by
the market. If it means selling below plan rate, so be it. The key variable in the
alternative analysis is the movement of prices above or below the market and the
resultant impact on the sales rate. Let's restage the situation, then step through the
analysis.

The Situation
Prices have crashed. The average inventory cost is well above the current mar
ket values and prices are not forecast to recover any time soon. You feel you could
probably make a normal, positive margin on material brought into inventory today,
but it will take a while to sell off the old, high-cost inventory.
If you chose to blow out the inventory, you would lower prices below the
current market level to stimulate the sales rate. Thus, the old, high-cost inventories
would be moved out sooner (in less time), and you can begin making your normal
margin on the resupplied volume. Note that we are viewing inventories on a first-in,
first-out (FIFO) basis. The trade-off is the money spent in lowering prices below the
market on the existing inventory versus the money gained from the additional period
of time in which sales can be made at a positive margin.
To understand the concept, it helps to graph what is happening over time (refer
to Figure 1). The size of the inventory does not change over time, but the old
inventory (declining line) is sold off and the new inventory is resupplied. Because
inventory is represented on a FIFO basis, the old is completely disposed of before the
new inventory is accessed. Our industrial hero would be indifferent when the costs
equaled the gains.
FIGURE 1.
30 r
25

i 20

10

-2_I_I_I_L.
Time 1 Week

[
200
AIC
| 19? I
? 180 | Realization
170 r
-L.
Week 50 51 52 1 2 3 4 5 6 7 8 9 10 11 12
Time

58 INTERFACES August 1982

This content downloaded from 190.213.241.97 on Fri, 08 Dec 2017 11:40:46 UTC
All use subject to http://about.jstor.org/terms
At the point of indifference
ID = ICRM,
where
/ = Inventory volume (in units),
D = Change in price (above or below market in $/unit),
C = % change in sales rate ratio (e.g., ?.20 = 20% drop),
R = Price under consideration (in $/unit),
M = Gross margin ratio (e.g., .05 = 5%).
Dividing both sides by / yields
D = CRM,
which can be rewritten as

C = DIRM, R = DICM, or M = DICR.

Now that we have these handy-dandy little formulas, let's see wh
learned. Our hero plugs in some numbers; he assumes that a $5 drop i
$200 to $195) will stimulate his sales rate by 20% and he expects a 5%
That is, he should get a change in revenue of $5 to offset the $5 chan
= .2X195X.05 = $1.95 ^ $5. This policy would have cost $5 per
back $1.95; not a good policy if one wishes to remain in business over
time. This circumstance is quite common among commodity products
to be so low, and movements in market price so large, that huge
required to pay back any price-cutting strategies. Dumping strate
viable.
Holding strategies are merely the inverse, and the same analysis can be used.
One must be cautious, however, because the elasticities may not be symmetrical.
Effective price increases over market may have a much more dramatic effect on the
sales rate than do price decreases. What, you ask, if the market isn't forecast to hold?
What if you think prices are going to take that long, slow slide to oblivion? In the
face of a forecast of declining prices, what is the strategy?
The same basic analysis applies with the addition of a new term. Not only do
changes in price affect the rate at which inventories are sold, but sales made today
will have a better realization than those made next week. The expression which
describes the situation is shown below:

D = CRM+ CIK
25(1 4- O '
where
K = The average dollar change in the market price over
the period (for example, a $3 per week drop),
5 = The sales rate per unit time when pricing at the
market.

The derivation of this expression is shown in the appendix.

One other common circumstance is to have prices and volumes fall off as the
market dies. The objective under these circumstances would be to reduce inventories

INTERFACES August 1982 59

This content downloaded from 190.213.241.97 on Fri, 08 Dec 2017 11:40:46 UTC
All use subject to http://about.jstor.org/terms
to a level that is appropriate given the new sales rate. The new term is the savings
generated as a result of the lower carrying costs. The expression is shown below:

D=CRM + .5Art(l-V),
where
A = Average inventory cost,
r = Interest rate for inventory carrying costs,
V- Ratio of change in inventories; e.g., .80 = a 20%
reduction (V=N/0, where N is the new and O the
old inventory volume),
t = Time required to accomplish the desired change in
inventory volume.
Here again, the reader is referred to the appendix for a derivation.

Discussion and Conclusion

The models presented here have direct and immediate applications in a number
of industries. While you may not know the exact elasticity of a product, you can still
set reasonable bounds. Operating managers especially have considerable experience
to draw upon in estimating the effects of price changes on sales rates. Even if these
parameters are unknown, one can compute the change in sales rate necessary to
produce a viable strategy. If this estimated rate falls outside the bounds of normal
experience, one can be assured the strategy is risky. Our objective was not to provide
a highly accurate model, but to produce a decision rule with wide application and
easy implementation.
There are a number of implications of following these strategies. One is soon led
to a posture where sales volumes are allowed to fluctuate with the market. This may
be no problem for a wholesaler with no production facilities, but achieving steady
takeaway for a mill would pose a conflict. The cost, however, is readily computed.
Inventory reductions are generally best accomplished by restricting resupply rather
than engaging in heavy-duty price competition. Given the low margins typical of
most commodity items, it is very difficult to compensate for price cuts with increased
volume ? it is, however, not impossible.
Increased market shares are, in general, desirable. Using the techniques shown
above, we can begin to evaluate the cost of gaining market share through price
leverage. You can easily spend a considerable sum gaining a market share which
depends to large measure on your willingness to continue to offer low-ball prices. In
a commodity market there are few loyalties. The question is the same. Is the value
lost in reducing price regained in the increased volume?

60 INTERFACES August 1982

This content downloaded from 190.213.241.97 on Fri, 08 Dec 2017 11:40:46 UTC
All use subject to http://about.jstor.org/terms
 APPENDIX
The revenue which is derived from the sale of an inventory during a declining
price market is shown below (assuming a linear price change):

Revenue =-7,
where
Pj = Beginning price ($/unit),
P2 = Ending price ($/unit).

The ending price at the "market" sales rate would be

The ending market price would be higher if the sales rate were accelerated because
less time would have passed before the inventory was depleted. This can be ex
pressed as

p\ ? p _ _ is
2 l s(\ + o
where

P\ = The new ending price.

The change in total revenue resulting from an increased sales rate, D2, is shown
below (for example, revenue would increase if the volume were sold off sooner
before prices had a chance to decline further):

-?['"-''-?rT?*]-i['--'--5*]
I2CK
25(1 + q *
At the point of indifference, the changes in revenue should be offsetting. The losses
from selecting a particular strategy of pricing should equal the gains. In this case,
there are two sources of change: changes in price (compared to market) which affect
the sales rate and the change in the market price itself. The complete expression
would be
ID = ICRM + D2.
Substituting D2 above yields:
I2CK
ID = ICRM + ??? .
25(1 +C)

INTERFACES August 1982 61

This content downloaded from 190.213.241.97 on Fri, 08 Dec 2017 11:40:46 UTC
All use subject to http://about.jstor.org/terms
 Dividing through by / yields:

D = CRM+ CIK
25(1 + C) '

Under the scenario where one wishes to reduce inventories in conjunction with a
contemplated change in prices away from market, the following would apply. The
saving in inventory carrying costs is

(O-N)
Saving = *??- Art,
where
O = Beginning inventory volume (in units),
N = Ending inventory volume (in units).

Our original statement, rewritten in terms of N and O, is

OD = OCRM.
We can add the new cost term representing the savings in carrying costs:

OD = OCRM + ^'^ Art

2
N
D = CRM 4- .5 Art - .5 - Art.
O

Because we find it easier to think in terms of a percentage (or rat

inventories, we will represent NIO by V, the ratio change in inven
D = CRM + .5Art(l-V).

62 INTERFACES August 1982

This content downloaded from 190.213.241.97 on Fri, 08 Dec 2017 11:40:46 UTC
All use subject to http://about.jstor.org/terms
 A Weyerhaeuser Compan
Western Trading Center
P.O. Box 1645
Tacoma. Washington 98401
(206) 927-7900

February 9, 1982

Dr. R. E. D. Woolsey
Editor-in-Chief, Interfaces
Colorado School of Mines
Golden CO 80001

Dear Dr. Woolsey:

The strategies described in this paper were implemented in a

business at Weyerhaeuser Company during the first half of
difficult to point to exact results due to the many confounding
marketplace, we were happy with the outcome. Comparisons to
and to periods with similar market conditions indicate
performance and better returns on assets employed.

The technique proved easy to implement and the results could

minimum of effort. We have included the information into the
our distribution manager's.

Sincerely,

D. M. Brown
Product Sales Manager
DMB:la50/202/b2
Enclosure

cc: Chron copy

File copy

INTERFACES August 1982 63

This content downloaded from 190.213.241.97 on Fri, 08 Dec 2017 11:40:46 UTC
All use subject to http://about.jstor.org/terms