The McKinsey Quarterly, 2002 Number 2
Like it or not, globalization is shaping the modern world. Yet international freight transportation, one of the industries that support the process, remains paradoxically old-fashioned. It is highly fragmented, with the top five container-shipping lines, for example, carrying only a quarter of the world’s ocean freight among them; prices on some routes bear little relation to supply and demand; and contracts are so casual that they might be sealed by a one-page fax or a handshake. Despite growth, since the mid-1990s, of more than 5 percent a year in the revenues of the air- and sea-freight sectors-to more than $130 billion annually by the end of the decade-both consistently underperformed the S&P 500 during the same period.
The proprietors of the many freight companies (including freight forwarders) that are state owned, in private hands, or owned by conglomerates have mostly been content to let them bump along the bottom. Are they doomed to stay there, or can the freight industry bring its practices and profitability more into line with its role in the modern world?
We believe that it can. Much of the trouble in the freight industry stems from the unusual amount of risk affecting its revenues. But the industry also satisfies the conditions required by the risk- and revenue-management concepts that, over the past 20 years, have markedly improved the performance of the fairly similar energy and airline industries. The big question facing freight businesses is how soon they too can apply such measures. As leading companies begin to take them up, others will have little choice but to follow, for those that apply them will also serve customers better.
More value at risk
Risk in freight transportation stems largely from three sources: changes in demand caused by the economic cycle, anomalies in the way contracts are drawn up, and uncertainty over prices.
Demand for freight transport generally tracks global economic cycles, so the freight industry could keep its margins steady by matching supply to demand. But individual companies can increase supply only in “lumpy” increments of productive factors with long lead times: ships and planes. Companies tend to order these simultaneously, when they think the world economy is set to grow. If, as can easily happen, they make a collective mistake about the cycle’s timing, they might all take delivery of new capacity just as demand drops.
The fortunes of air cargo in particular are tied to those of the world e conomy, partly because sea freight offers customers a cheaper substitute. One international freight forwarder, expecting prices to reach at least $3 a kilogram on the trans-Pacific air route at the end of 2000, chartered several large freight planes for six months under a fixed contract. But from mid-2000 to mid-2001, air cargo rates-reflecting the global economic slowdown and an increase in air cargo capacity-dropped by more than 30 percent, leaving the forwarder substantially exposed.
Contract practices in freight make future revenues extraordinarily unpredictable. Customers can reserve space on terms that in effect give them a free call option, for example. If the spot price for space falls below the forward price agreed upon between carrier and customer, the customer can simply rebook space on the spot market, without paying any penalty to the original provider. Even if the contract has a minimum-volume clause, such provisions are rarely enforced.
Equally curious, the industry gives volume-based discounts on all routes, regardless of capacity constraints. A customer seeking a lot of space on ships sailing from Hong Kong to Antwerp-a route in high demand-always pays less per 20-foot equivalent unit (the size of a typical freight container) than customers booking smaller volumes. Microeconomic theory suggests that on routes where capacity is scarce, rates should be higher.
Conversely, customers prepared to pay higher rates can’t be sure of getting space. A company with an urgent shipment but no space reserved on a full ship that is about to sail, for instance, can’t buy space on it at any price. Although customers often break their contracts with carriers when it suits them, carriers can’t do the same, in part because the freight industry has no system for compensating customers with nonurgent cargoes that could comfortably sail on the next available ship.
In the short term, prices for capacity can swing unpredictably for reasons such as overheating in a local economy or local supply difficulties, and these swings create gaps between spot and forward prices that customers may exploit by using the contract anomalies described above. But average pricing on some routes seems less rational. Prices for container shipping, to give one example, can drop far below what microeconomic theory predicts on routes with overcapacity. In 1999, when capacity in the westbound shipping lane from the Far East to Europe was, on average, 76 percent filled, the market price was $1,554 per 20-foot equivalent unit, close to the theoretical clearing price of $1,560. But on the return trip to Asia, with only 51 percent of capacity filled, the market price plummeted to $736 per 20-foot equivalent unit, compared with a theoretical clearing price of $1,450.
Carriers desperate to fill empty holds reduce prices to their marginal costs (terminal charges and documentation fees), preferring to treat the expense of sailing the ship as fixed, since “it has to go back to Asia anyway.” These carriers would keep prices firmer if they could, but a lack of central corporate control over pricing, and general overcapacity, create irresistible temptations for individual sales agents to drop prices to these “marginal” costs.
How risk and revenue management could help
Given the risks inherent in the freight transport industry, what could risk- and revenue-management concepts do to alleviate them? And how should these techniques be applied?
Risk management enables companies to quantify risks to their future revenues and costs. Those companies can then adjust prices and contracts so that such risks are shared fairly with customers and suppliers.
First, companies need to measure how much of their future revenues may be at risk from factors such as market downturns, no-shows, and exchange rate fluctuations and to estimate the probability that these events might occur. To do so, they must collect macroeconomic information such as historical price movements and price forecasts; apply this data, under different scenarios, to all contracts booked; and subtract from each contract’s book value the amount that is, on average, at risk. The difference between the book and risk-adjusted value can be alarming, pushing even apparently lucrative contracts into the red.
A freight company that knows the scale of its expected total revenue at risk can spread its risks more evenly and widely. For fairness’s sake, customers should shoulder some of the risk now borne solely by carriers and freight forwarders. So instead of giving free call options, for example, freight forwarders and carriers, in effect enforcing a penalty for no-shows, could offer discounts to customers who promised to pay for, say, two-thirds of the space reserved.1 Carriers could also offer a discount for forward contracts they can interrupt if spot prices soar-an approach that sounds harsh but recognizes that different customers have different requirements. Those with nonurgent shipments would be happy to pay less for an interruptible contract if they were guaranteed space on a later ship, while customers that suddenly needed space urgently would be pleased to know that it was available for a price.
To spread risks more widely, freight companies can adjust their portfolios to balance the risks that each type of customer represents. A carrier might, for example, discover that it expected pharmaceuticals companies to provide 20 percent of its revenues, at yields2 lower than those of high-technology companies, which were expected to account for 30 percent-but that only 15 percent of the value of the pharmaceuticals freight was at risk, compared with 50 percent of the value of the high-tech freight. If a company does a large proportion of its business with the unpredictable but high-yielding high-technology sector, it should try to move some to less risky sectors, such as pharmaceuticals.
To become more responsive to risk generally, freight companies need to adapt their business systems. The high fixed costs of air and ocean carriers make it hard for them to adjust their capacity to market downturns, for example. But if carriers made more of their fixed costs variable-by leasing some of their ships and aircraft, for example, or outsourcing more ground-handling activities-they could make adjustments with greater ease.
Revenue management enables companies to understand the fit between their capacity and future demand and to price and allocate capacity for different customer segments in a way that maximizes revenues. At present, most freight companies offer a standard service to all customers, with prices based on a vague understanding of the competitive situation. If those companies better understood the market’s dynamics and the needs of the different segments they address, they could vary their services and price each according to the value customers placed on it. Freight companies can vary their basic service-the provision of cargo space-by adding different delivery time guarantees, reservation requirements, and handling options. They can then offer an array of possibilities at different prices, charging the top premium for, say, late bookings requiring express delivery and special containers.
In air cargo, the largest companies already offer a range of premium and standard products, though the idea has yet to be applied to container shipping. The returns can be substantial: by introducing standard and express heavy-freight and parcel products, one leading air cargo carrier increased its yield by about 5 percent from 1995 to 1998 while the average indust ry yield fell by 15 percent. But to make demand management work, a carrier must be able to reserve capacity for higher-yielding products-an approach that requires the carrier to make accurate forecasts of demand for them on each route. On routes with limited capacity, carriers should offer high-yielding products almost exclusively, just as, say, on a 7 AM flight from London to Zurich most seats are reserved for full-fare passengers. Freight customers unwilling to pay full fare can buy standby capacity at the last moment if the forecasting has been a little off; otherwise they should wait.
The tide is turning
Risk and revenue management could improve the performance of freight transportation, much as they have already done in the energy and airline industries of many developed markets (seesidebar ”
Risk and revenue management: Before and after“). Thus far, however, freight companies have barely begun to apply these concepts-although that might be changing (seesidebar “Air cargo after September 11“).
There have, it is true, been a few attempts to implement them-such as revenue management in air freight-but many attempts have failed, and the transition has been slow and painful even for the successes. Why has the industry been so hesitant? Mainly because, until now, it could live without managing risk and revenue. In the energy and airline sectors, deregulation forced companies to improve their performance in order to survive. But passive owners have let many freight transportation providers get away with low rates of return, and strong growth in world trade over the past decade has promoted this kind of laxity: the revenues of freight companies have grown in step, masking their weak profitability. Individual companies have feared to make the first move, reasoning that if they denied low-yielding customers space in their holds, while their competitors continued to offer it, their customers would defect.
Moreover, the industry has found the disciplines of risk and revenue management difficult to apply for organizational reasons. First, the techniques require a delicate balance between central control and local entrepreneurial spirit, but freight companies generally have locally autonomous and fairly uncoordinated operations scattered across the globe. Second, most senior managers have moved up through operational positions and thus tend to see revenue management as an expensive and complicated IT tool,3not as a commercial principle that shifts the focus of the entire organization from volume to value.
Times are changing, however. The industry’s growth has slowed to single digits, forcing freight companies to focus more closely on the efficiency and profitability of their operations. Equity markets are putting more pressure for improved performance on publicly owned freight companies or their parents. In response, some companies, especially in container shipping, are consolidating, which has produced a group of potentially more competitive operators that will raise the level of the game for the rest.
Recognizing these pressures, companies have started to change. Some management positions are being filled by people who come from outside the industry or have experience in nonoperational functions. These managers have already implemented the concepts of risk and revenue management and better understand the required balance between central control and frontline autonomy. Meanwhile, the vendors of sophisticated revenue-management4systems are finally seeing an increase in demand from leading freight transportation companies. The investment required might exceed $10 million, and a single project could take two or three years to complete. But we believe that if the second wave of pioneers implements these tools successfully, making the organizational changes entailed by switching from a volume to a value mind-set, others will follow. After all, the first freight companies to adopt risk and revenue management successfully will not only capture the full value of their capacity but also, by definition, understand and serve the varying needs of their customers more satisfactorily, so customers are likely to defect to the first movers.
Risk and revenue management: Before and after
Risk- and revenue-management concepts should work well for the freight industry because the sector satisfies the conditions they require. First, the market can be divided into different customer groups with different needs, thereby revealing the consumer surplus waiting to be captured. Second, the level of uncertainty about demand and prices suits probabilistic forecasting tools. Third, capacity is perishable-an empty hold in a container ship can’t be stored for later use-and supply is inflexible, so tools to smooth revenue under fixed time and capacity limits are appropriate. Last, to be certain of paying very high fixed costs, companies must “fix” forward revenues.
Moreover, the use of both risk and revenue management reinforces their value to freight companies. The greater understanding of forward prices and value at risk that a company gains from risk management, for instance, feeds into the company’s decisions about allocating capacity on particular routes in order to maximize revenue. In many ways, the freight industry thus resembles the airline and power industries, in which both concepts have succeeded.
Although individual US airlines aren’t owned by the government, it effectively controlled their performance until the late 1970s by setting a single price for each route and decreeing which of the many carriers could operate where. But from the late 1970s on, the government relaxed the rules. American Airlines was the first to use rudimentary revenue-management techniques, for example: offering discounted fares to passengers who booked early, starting to reserve seats for higher-paying customers, and overbooking seats in the knowledge that some passengers would cancel at the last moment and that others would fail to show up. In this way, American Airlines claims to have been able to generate as much as $500 million a year in additional profits from the early 1980s onward.
Today few airlines fail to manage their revenue-most are on their third generation of revenue-management systems-and almost all offer fares at prices that vary according to conditions such as how far in advance tickets are bought, the length of the stay, and whether tickets can be refunded. Airlines that have not adopted these techniques have had some spectacular failures. After deregulation in the United States, for example, People Express was the first airline to offer tickets at discounts of 35 to 80 percent off prevailing rates. Traffic soared during the first three years, but the airline, unlike its competitors, failed to manage its revenue by restricting the number of seats it sold at deep discounts. As a result, and despite its popularity, it never covered its costs. It went into bankruptcy and had to sell its assets in 1986.
Until the early to mid-1990s, most power plants and electricity distribution networks in Europe were owned by the state, which managed the industry’s performance by controlling wholesale and retail prices and absorbed the risk of short-term mismatches between supply and demand. But the industry in many Euro pean countries has since been privatized. Newly independent companies have looked for ways to manage not only the risks the state once bore but also the risks represented by new competitors.
After privatization, the market set prices. Although these generally fell, it was their extreme volatility rather than their lower average level that prompted companies to manage this new exposure to price risk. Prices fluctuated not only seasonally, as expected, but also day-by-day, varying in 24 hours by as much as 300 times the average daily movement. Power generators were the first to develop risk-management products, such as contracts for uninterruptible service, for what had appeared to be a commodity industry. By refining these risk-management tools, the electricity generators managed largely to stabil-ize their revenues at levels that were acceptable to shareholders, customers, and industry regulators alike.
Risk management is hard to get right. While most of today’s leading energy companies now practice it well, some expensive mistakes were made at first, especially in Scandinavia, where three companies went bankrupt in the mid-1990s because their loosely controlled management teams lost too many bets on price movements that went the other way.
Air cargo after September 11
Could September 11 be the external shock that jump-starts modern revenue management in air cargo? It seems possible. A few carriers, such as Lufthansa Cargo, that have large freighter fleets had already begun to reduce their capacity before September 11, 2001, in response to the economic downturn. But in the weeks that followed the disaster, almost all airlines cut 10 to 15 percent of their capacity, mainly on North Atlantic routes. Besides announcing layoffs and receiving state aid, as expected, most carriers, surprisingly, managed to impose security surcharges reflecting the additional cost of counterterror measures. At the same time, some carriers, such as KLM Cargo, boldly increased their systemwide prices by 5 to 7 percent in view of continued strong demand in Asia and reduced supply over the North Atlantic, though during the third quarter of 2001, traffic for the most severely affected European and US carriers was 3 to 7 percent below the previous year’s levels.
It is too early to tell if the disaster has ended irrational pricing in air freight for good. But there are strong signs that the days of reducing prices below the level warranted by supply and demand could be over, at least for the larger, more sophisticated carriers.
Lucio Pompeo is an associate principal in McKinsey’s Zurich office, and Ted Sapountzis is a consultant in the Amsterdam office.
Sophisticated carriers are starting to experiment with such innovative products, whose commercial impact is now unknown. Lufthansa Cargo, for example, introduced two of them last year. One gives companies that commit themselves to pay for a certain volume on a route for the next 6 months a 5 percent discount on the standard rate; the other gives a 10 percent discount to customers making a similar commitment for the following 12 months.
In the freight transportation industry, yield means realized price per unit of transport capacity.
The first companies to apply these principles in effect copied the approach and IT systems used by airlines. The systems struggled because of a notable difference between the two industries: airlines sell space, seat by seat, to many different customers, whereas freight businesses generally sell up to 70 percent of their space in bulk contracts to just a few buyers with which they have long-term relationships. Not surprisingly, the copied systems didn’t wo rk.
Risk management relies on simpler computer tools that cost much less than revenue-management systems.