The economics of a chemical plant: costs, prices and profit: What makes the process industries different

1.1 A process plant earns its living from the value it adds between raw material and product

Brennan defines the process industries as those that transform raw materials into finished products at commercial scale, using physical, chemical and sometimes biochemical change inside engineered plants. Most of their output never reaches a shop directly. Hydrogen, ethylene and sulphuric acid are intermediate products: they feed a long chain of further plants before anything reaches a consumer.

Resources arrive impure and far from customers, so the industry is a network of purification, conversion and transport steps, each adding cost and, if well run, more value than cost.

Value added is the useful measure. Wittcoff defines it as sales minus the cost of raw materials and other bought-in inputs such as fuel, electricity and services. In the United States in 1999 the chemical industry shipped $419.7 billion of goods and added $229.3 billion of value, while petroleum and coal products shipped $168.1 billion but added only $42.4 billion. Chemical sales were less than twice refining sales, yet value added was 5.4 times as large. Wittcoff says that gap is why oil companies built chemical arms.

Brennan's illustrative company accounts show where value added goes. Of $1,550 million of sales, $1,020 million paid for materials and services, leaving $530 million of wealth created. Half went to employees, 15% to government, about 15% to lenders and shareholders, and 20% was reinvested as depreciation and retained profit.

Every project must pass one test: revenue must exceed operating costs and still leave an adequate return on the capital locked up in the plant.

• Brennan (2020), Ch 1 The scope of process industry economics · pdf pp. 8-10, 14-16
• Wittcoff (2004), Ch 1 The chemical industry · pdf pp. 42-43, 60

1.2 Capital intensity and continuous operation reward size and full plants

Brennan's first defining trait of the process industries is capital intensity, and it drives most of the economics. A world-scale alumina refinery, aluminium smelter, ethylene plant or pulp mill each needs more than US$1 billion, and a minimum scale is usually needed just to reach competitive cost.

Wittcoff traces how this came about. Before the Second World War, chemicals were often made in batches in general-purpose equipment: cheap to build, expensive in labour. Refiners pioneered large continuous plants, and after the war chemical makers copied them. A typical ethylene cracker grew from 70 million lb a year in 1951 to 1 billion lb in 1972. Brennan reports a 1.5 million tonne a year US ethylene plant commissioned in 2017.

The reason is partly geometry. The capacity of a tank or column rises with its volume, but its cost rises roughly with the steel surface that encloses it, so cost grows with about the two-thirds power of capacity (the square-cube law). Labour barely grows: doubling a unit may raise its labour cost only 10-20%.

Big continuous plants therefore have high fixed costs and low variable costs. Once a plant is built, the cheapest tonne is the next one, so producers run hard even when prices fall. Wittcoff points to the wide gap between the break-even price, where price equals total cost, and the shut-down price, where price falls below cash cost. That gap is a root cause of the industry's chronic overcapacity.

• Brennan (2020), Ch 1 The scope of process industry economics · pdf pp. 10, 16-17, 21
• Brennan (2020), Ch 2 Market evaluation and forecasting · pdf p. 56
• Brennan (2020), Ch 3 Capital cost estimation · pdf p. 59
• Wittcoff (2004), Ch 1 The chemical industry · pdf pp. 50, 58-59

1.3 Joint products and commodity pricing tie each plant to many markets

Many processes cannot make one product alone. A steam cracker splits hydrocarbon feed into a mixture, and the feed decides the mix. Brennan gives the figures: 125 tonnes of ethane make 100 tonnes of ethylene and only about 3 tonnes of propylene, whereas 330 tonnes of naphtha make 100 tonnes of ethylene plus about 52 tonnes of propylene, 14 tonnes of butadiene and 22 tonnes of benzene. The IEA puts the ethane route at about 1.2 tonnes of feed per tonne of high value chemicals (HVCs: ethylene, propylene, butadiene and aromatics) and says an ethane cracker yields about 50% more HVCs than a naphtha cracker, almost all of it ethylene.

Joint products complicate the arithmetic, because the cost of each is partly a matter of convention. Brennan notes that chlorine and caustic soda, or phenol and acetone, need a rule for sharing cost, perhaps by tonnage with an adjustment for extra processing. Minor saleable streams are usually treated as by-product credits and deducted from raw material cost.

Brennan's product matrix sorts chemicals by volume and by differentiation, meaning whether a buyer can tell one maker's product from another's. A commodity such as ethylene, ammonia or sulphuric acid is made in large volume to the same specification by everyone, so buyers choose on price, delivery and reliability. Pseudo-commodities such as polyethylene grades are made in volume but sold on performance. Fine chemicals like citric acid are low volume but made to standard specifications; specialty chemicals such as catalysts are low volume and sold on what they do.

• Brennan (2020), Ch 2 Market evaluation and forecasting · pdf pp. 27-29, 55
• Brennan (2020), Ch 3 Capital cost estimation · pdf pp. 95-96
• IEA (2018), The Future of Petrochemicals, Ch 2 · pdf pp. 31-32

See it in the table

Check your understanding

In Wittcoff's 1999 US figures, what made chemicals attractive to petroleum companies?

A commodity plant with high fixed and low variable costs sees prices fall below its total cost. What does the cost logic in Wittcoff's appendix suggest it does in the short term?

Per 100 tonnes of ethylene, how does propylene output compare between Brennan's naphtha and ethane crackers?