Innovation Case: Roaring Out of Recession

Unit 3 of Professor Ranjay Gulati's Innovation — Strategic Thinking for Turbulent Markets: Roaring Out of Recession and Investing in Growth Through Uncertainty.

This session has no single case company. Instead, it's two articles by the same author, written thirteen years apart. The first is a study of 4,700 companies that answers the question "what do companies that come out of a recession stronger actually do during it?"; the second circles back to a harder question — "why, even knowing what to do, do most people still not dare to do it?" Belden began its transformation in its most profitable year; these two articles are about how to spend money in your least profitable one. Opposite directions, same principle.

Here goes — enjoy:

Case Summary

December 2008: the Great Recession is in full swing. Three scholars launch a year-long project to answer a question that everyone talks about but almost no one has studied empirically: what strategy should you use during a recession in order to win after it? There's plenty of folk wisdom — Procter & Gamble, Chevrolet and Camel cigarettes survived the Great Depression by advertising heavily, stories we've all read many times — but very little data.

They took 4,700 public companies from S&P's Compustat database and ran them through the last three recessions: 1980–82, 1990–91 and 2000–02. For each one they looked at three periods: the three years before the recession, the recession itself, and the three years after. They used six balance-sheet items to measure where companies shifted their resources during the recession — headcount, cost of goods sold as a share of sales, R&D spending, SG&A, capital expenditure, and the stock of fixed assets. Only changes in the top or bottom third relative to the industry average counted as "significant."

The results are glaring. 17% of the companies didn't survive the recession — they went bankrupt, were acquired, or went private. Of those that survived, 80% had not regained their pre-recession growth rates three years after the recession ended; 40% hadn't even returned to their absolute pre-recession levels of sales and profits. Of the growth leaders before the recession, 85% were knocked out of that position during it. Only about 9% of companies did better after the recession than before it, outperforming their peers by at least 10% in both sales and profit growth.

The winners were neither of the two types people would expect. Drawing on the theory of Columbia University psychologist Tory Higgins — that people have two modes of self-regulation, "promotion focus" and "prevention focus," one seeking gains and the other avoiding losses — the authors sorted companies into four groups. Prevention-focused: cut almost everything, and invested more than rivals in nothing. Promotion-focused: invested more than rivals in at least one area, and cut more than rivals in none. Pragmatic: both cut (laying off staff or reducing cost of goods sold) and invested (in at least one of sales, R&D, capex or fixed assets). Progressive: reduced cost of goods sold but didn't lay off more people than peers, while investing more than rivals both on the market side — sales and R&D — and on the asset side — capex and fixed assets.

The probability of each group "breaking away" after the recession — leading peers by more than 10% in both sales and profit growth — was 21%, 26%, 29% and 37% respectively. In the three years after the recession, the average annual growth of prevention-focused companies was 6% in sales and 4% in profits; promotion-focused, 8% and 6%; progressive, 13% and 12%. After the 2000 recession, the 200 largest companies averaged US$12 billion more in sales than before the recession. The prevention-focused ones added only US$5 billion, with profits up US$600 million; the progressive ones added US$28 billion, with profits up US$6.6 billion — eleven times as much.

Then comes a set of contrasting stories.

Sony and HP: two directions in the same recession. In the two years after 2000, Sony cut its workforce by 11%, R&D by 12% and capital expenditure by 23%. Its profit margin rose from 8% in 1999 to 12% in 2002, but sales growth fell from an average of 11% a year before the recession to 1% afterwards. It went on to lose e-readers to Amazon, game consoles to Microsoft and Nintendo, and OLED TVs to Samsung. In December 2008 it announced yet again that it would cut US$2.6 billion in costs, lay off 16,000 people and postpone a badly needed LCD TV plant in Slovakia.

HP went the other way. As then-CEO Carly Fiorina put it:

"In blackjack, when the odds are improving, you double down. We're going to double down."

So it bought Compaq for US$25 billion — the largest acquisition in the company's history — raised R&D by 9%, spent US$200 million on brand advertising and US$1 billion on rolling out information technology in developing countries. These moves stretched the organization and spread management's attention thin. When the recession ended, HP found it couldn't catch up with the profitability of IBM and Dell: in 2004 HP's EBITDA margin was 8.4%, IBM's 16.8% and Dell's 9.3%.

Office Depot and Staples are the cleanest pair. Office Depot laid off 6% of its staff, but its operating costs didn't really come down; it launched a sales incentive program, yet sales growth still fell from 19% before the recession to 8% afterwards, five percentage points below Staples. Staples closed a batch of underperforming stores but grew its headcount by 10% — to support the new premium categories and services it was launching — while keeping operating costs under control. Its sales doubled from US$7.1 billion in 1997 to US$14.6 billion in 2003; Office Depot's went from US$8.7 billion to US$13.4 billion, up by half. In the three years after the recession, Staples' profit margin was about 30% higher than its old rival's.

Target and TJX both expanded during the recession, with opposite outcomes. In the 2000 recession, Target raised its marketing and sales spending by 20% and its capital expenditure by 50%, growing from 947 stores to 1,107, and from 30 SuperTargets to 118. Rather than build its own e-commerce, it partnered with Amazon to sell its goods; it collaborated with designers like Michael Graves, Philippe Starck and Todd Oldham, cementing its "cheap chic" image; it was also one of the 12 retailers that founded the WorldWide Retail Exchange, using a B2B electronic marketplace to drive down purchasing costs; and it folded its Dayton's and Hudson's department-store brands into the better-known Marshall Field's. During the recession its sales grew 40%, its profits 50%, and its profit margin went from 9% to 10%.

TJX — parent company of T.J. Maxx and Marshalls — added 300 stores to its 1,350 between 2000 and 2002, increasing floor space by 25% and nearly doubling capital expenditure. Rivals were contracting, and real estate was cheap. Sales did beat peers by four percentage points, but three years later its profit growth was 9% below its peers', whereas before the recession it had been on par. There was only one reason: it simply scaled up its old model without changing it.

The article ends with a nine-box grid. Three defensive moves (cutting staff, operational efficiency, both) and three offensive moves (market development, asset investment, both). The best box is "operational efficiency × investing in both markets and assets": sales growth of 13.0%, profit growth of 12.2%. The worst box isn't "cut only" but "cutting staff × investing in both markets and assets": sales growth of 3.3%, profit growth of minus 5.2%.

Thirteen years later, the same author comes back to talk about mindset. The 2023 article opens with Alaska Airlines. In early 2020, US airline passenger traffic plunged 96% year on year. It did everything its peers did: froze hiring, cut executive pay, renegotiated supplier payments, suspended buybacks and dividends, slashed capex. But that December, as rivals were cancelling aircraft orders, it ordered 68 Boeings plus options for 52 more — the only major US airline to place a big order that year — and used the opportunity to consolidate its mixed Airbus-and-Boeing fleet into an all-Boeing one. In the words of CEO Ben Minicucci:

"We knew supply chain issues would slow aircraft deliveries, but this deal put us first in line. After all, in this business, you can't grow without airplanes."

In 2021–22 it exercised those 52 options, locked in another 105 aircraft through 2030, and in 2022 concluded five labor agreements, including one with the pilots' union. In 2022 its pretax margin was among the highest in the industry.

Gulati says the 2010 study only answered "what to do." It didn't answer what kind of mindset allows a handful of leaders to dare to retrench and invest at the same time. He offers three: sensemaking (getting moving even in the fog), a bootstrap ethic (start-up-style frugality), and stakeholder balance (spreading the pain across everyone). Then come the stories of Edward Jones, Firefly, Panera and Levi Strauss — which I'll save for the next section.

The article closes on a question, borrowing a popular line from writer Vivian Greene: would you rather crouch down and wait for the storm to pass, or learn to dance in the rain?

The professor's three questions for this session:

1. What are the foundations of building competitive advantage in turbulent markets?
2. How do companies in different industries use creative strategies to grow both revenue and profits in tough markets?
3. How exactly did these companies put change into practice?

My Thinking

Both of the fastest roads lead to mediocrity: cutting costs across the board, and flooring the accelerator

Start with the prevention-focused. Their logic is airtight: the recession is here, survive first, cash is king, stop everything that can be stopped. The article says these companies have a 21% chance of breaking away after the recession — the lowest of the four groups.

Why the lowest? The authors list four things. An organization that only looks at losses sets its goals low, so innovation and cost-saving both become a matter of squeezing out the last drops; it doesn't learn to run more efficiently, it just "does the same with fewer people," so quality drops first and customer satisfaction follows; cost-cutting decisions get centralized in the finance department and applied across the board, and no one can tell which expenses are the seeds of post-recession growth; pessimism spreads, with centralization, tight control and the constant threat of another round of cuts — everyone is thinking about how to survive personally.

And the fastest cut turns out to be the most expensive. Only 23% of progressive companies cut staff, versus 56% of prevention-focused ones. Companies that relied on layoffs alone had an 11% chance of breaking away. Layoffs lower this quarter's costs; what you pay back is the next cycle's hiring costs, morale, and the window for expansion you missed. Office Depot cut 6% and its operating costs didn't fall; Staples added 10% more people and kept operating costs under control — because what it closed was stores, what it changed was the model, and what it added was people.

Now the promotion-focused. Fiorina's "double down" sounds like courage; the data gives it a 26% probability. Optimistic organizations attract optimistic people and develop a kind of positive groupthink, in which dissenting voices get sidelined and early signals like shrinking customer budgets get ignored. Worse still: prevention-focused companies can push costs as a share of sales down by three percentage points, while promotion-focused ones can't push them down by even one. When they finally have to cut, every department believes it's vital to the company, everyone blames everyone else, no trade-offs get made, and decisions freeze.

My own reading is that Sony and HP made two sides of the same mistake. One treated cost as the goal; the other treated optimism as a strategy. Both forgot to look at what customers, in this recession, had stopped buying — and what they still needed.

Reefing the sails isn't abandoning the voyage: the money you save in process should go where customers will come back

What the progressive companies do can be put in one sentence: their defense is selective; their offense is comprehensive. They cut cost of goods sold by rethinking supply chains, organization and processes, not by cutting people; and when they invest, they invest on both the market side and the asset side at once — R&D, marketing, and plants and equipment bought while they're cheap.

One number in the nine-box grid deserves a second look. The worst box isn't "cut without investing" but "lay people off, then invest across the board" — profit growth of minus 5.2%. Why? Because the people who would carry out the investment are gone. You've bought the equipment, opened up new markets, increased the R&D budget, but the people who could turn all that into products and sales are the ones you showed the door three months ago. Defense and offense aren't two ends of one line, a compromise between "cut a bit more" and "invest a bit more"; they're two dimensions, and they come in a sequence.

The contrast between Target and TJX says something else: cheap assets are not an advantage in themselves. Both bought stores after 2000, and TJX bought even more. The difference is that Target simultaneously changed how it purchased (the B2B exchange), its brand (designer collaborations, merging its department stores) and its channels (partnering with Amazon rather than building its own), while TJX simply doubled down on centralized buying and flexible distribution. Three years after the recession, one had a margin going from 9% to 10%; the other's profit growth trailed its peers by 9%.

Alaska Airlines demonstrates the sequence most clearly. It first completed every defensive move — froze hiring, cut pay, stopped dividends, renegotiated payments — made sure its balance sheet was strong enough, and only then placed that order for 68 planes. Offense is earned by defense; it isn't a substitute for it. And the order was itself a defensive move: consolidating a mixed fleet into all-Boeing permanently lowered its operating costs.

So where do you invest? The two articles offer only one filter: the customer. The 2010 article writes that progressive companies "stay closely connected to customer needs — the most powerful filter for investment decisions." In 2008 Target saw same-store sales slipping as Walmart's "everyday low prices" hit home with consumers. Instead of discounting across the board, it noticed something more subtle: consumers were cutting their "wants," not their "needs." So it doubled the size of its food sections, expanded its private labels Market Pantry and Archer Farms, and shifted the emphasis of its slogan "Expect more, pay less" to the second half. By 2008, food had become a US$1.8 billion business for Target.

You reef the sails so the ship can keep moving forward — not so it can sit still and wait for the wind to pass.

When you're lost, even an old map can save your life — get moving first, then correct course

The organizational psychologist Karl Weick tells a story that he himself admits may be apocryphal. A unit of soldiers was lost for days in snowstorm-swept mountains. Someone found a map in his pocket, morale soared, and they used it to find their way back to base. Only later did they discover that the map was of a different mountain range.

Weick's conclusion: when you're lost, any old map will do. The map saved them not because it showed the right way, but because it got them moving. Once you move, the terrain under your feet starts giving you feedback, and the route can be corrected.

This is what Gulati calls sensemaking. Difficult situations paralyze people, but we don't need to have everything figured out before taking the first step. Resilient leaders fix the ultimate goal, take a deliberate first step, and adjust as they go. In fact, the 2010 article ends with a similar line: few progressive leaders enter a recession with a master plan; instead, they encourage their organizations to discover what works and then combine those discoveries into a portfolio of initiatives.

Edward Jones is the best example of this mindset — and it happens to be a wealth management firm. In the mid-2010s it saw that its clients were aging and that its traditional model — advisers selling products to long-term investors — was under threat from cheaper, more convenient digital competitors; industry surveys showed consumers were broadly dissatisfied with the financial advice they got, most of it narrow product advice rather than holistic wealth management. In 2019 it launched a multi-year transformation called "human-centered holistic wealth management": advisers would understand clients' needs beyond money, and the firm would invest in systems, processes, talent and training.

Just as the transformation got under way, the pandemic hit. It didn't stop; it let investment grow faster than revenue. In late 2022 and early 2023, as the economic outlook darkened again and peers were all cutting costs and laying people off, some people inside once again asked whether to pause. CEO Penny Pennington's answer: not investing is the more dangerous option.

But the approach changed. It postponed hiring 1,000 employees; it strengthened its financial planning and analysis capability so that every new investment needed a business case and success metrics; it prioritized by value — upgrading the IT systems for its fixed-income business first, then those for equities. The destination didn't move; the steps got smaller, and each one came with a measuring stick. In her own words: "We invest in good years, and we invest in bad years."

My own reading is that Weick's old map and Edward Jones's FP&A don't contradict each other at all. One says "don't wait for a perfect map before setting off"; the other says "once you've set off, every step has to be measurable." Put together: the destination stays fixed, the steps can change, and every step can be measured.

The most common stupidity is forgetting what you set out to do

Nietzsche said that the most common form of human stupidity is forgetting what one is trying to do. Gulati uses this line to introduce the second mindset. Panicked executives, in pursuit of lower costs, often cut away the company's future along with them — some call it the "efficiency trap," others the "death spiral."

Drucker's old saying gets put to use again here: efficiency is doing things right; effectiveness is doing the right things. When cutting costs, the question isn't "how much more can we save?" but "after we save it, can we still deliver the products, services and processes that will get us to our destination?"

Gulati calls this mindset the bootstrap ethic: living like an early-stage start-up. Obsessive about customer value, and stingy about everything else. The filter is still the same one: does this project, this team, this product line make us better at serving our target customers? If yes, keep it and invest more; if no, cut it.

Firefly is a company that puts advertising screens on top of taxis and ride-hailing cars. When stay-at-home orders hit in 2020, nobody was looking at outdoor ads and revenue went almost to zero. CEO Kaan Gunay cut staff in sales, marketing and sales support, cut everyone's pay by 20%, renegotiated contracts with fleets and drivers, and moved operations and engineering to Turkey, where engineers are cheap but first-rate. Then, with the money saved, he hired a seasoned chief revenue officer and kept investing in engineering and product, because what customers wanted in 2021 and 2022 was better ad optimization, forecasting and planning tools; and, taking advantage of market pessimism and low valuations, he acquired a US peer in 2021 and a UK peer in 2022.

Panera is an example from the other end: a big company. In recent years the restaurant industry has faced slowing sales, labor shortages and rising costs all at once. The bottleneck CEO Niren Chaudhary saw was general-manager turnover — when a store manager leaves, the customer experience goes with them. So he doubled down on people, against the grain: substantially raising base pay, giving managers more training, and tying more of their compensation to store performance, so that the best managers could earn six figures. Where did the money come from? Menu prices rose faster than inflation, while new products (flatbread pizza) and promotions for price-sensitive customers (the US$10.99-a-month Sip Club unlimited drinks subscription) kept traffic up. The result: revenue growth that outpaced the industry, rising market share, and employee turnover down 10%.

The real crux of this session is in a passage in the middle of Gulati's article: there are three obstacles to playing offense and defense at the same time. Organizationally, adding and cutting happen simultaneously, and reallocating budgets triggers infighting — the fix is to use customer focus as the basis. Strategically, short-term efficiency and long-term investment are hard to tell as one coherent story — the fix is an overarching purpose. And culturally, there's loss aversion: people are wired to fear losing, and a crisis magnifies that fear. His fix is a single sentence: explicitly cultivate courage as a cultural value, make courageous behavior the norm, and have leaders go first. Minicucci said "we wanted to be aggressive"; Pennington said "not investing is the more dangerous option" — in both cases the CEO said it out loud first, and only then did the organization dare to follow.

A chef who can make sweet and sour lets every party hurt a little less, instead of making one party bear it all

Gulati's late colleague Sumantra Ghoshal — a professor at London Business School and the founding dean of the Indian School of Business — had a metaphor: the best leaders are like Chinese chefs, because they can make sweet and sour. They know how to look after stakeholders whose demands are completely different, all at the same time.

This is hard enough in good years, and harder in a crisis. Once the fear of scarcity takes hold, groups inside and outside the company become tribal, cling to their own interests and resist change — which makes both cutting costs and investing harder. Left unchecked, decisions become timid, unimaginative and short-sighted.

What resilient leaders do, Gulati calls practical idealism: ask certain groups for short-term concessions, but limit their pain, while attending to longer-term needs. Here's what Levi Strauss did in 2020: suspended dividends, cut 15% of its workforce, saved US$100 million — that was defense; but it extended health insurance for departing employees, added benefits for those who stayed, gave suppliers nearly US$2 million in grants and helped them find financing, and didn't stop a single sustainability program. CFO Harmit Singh said you can't "do it today and then forget about it tomorrow because a recession is coming."

It could have saved more. It chose to keep the people and suppliers it would need in the recovery. I love what Singh said next:

"There's no right answer, there's no formula, and I'm sure someone could find fault with what we did. But if I can sleep at night, I think we're in the right place."

At Edward Jones, Pennington talks about "financial balance" among three Cs: clients, colleagues, community. In 2023, even as she tightened spending, she kept investing in new tools for clients, kept paying advisers bonuses (smaller ones), and kept putting tens of millions of dollars into "financial fitness" programs for high school students.

This is the one genuinely new dimension the 2023 article adds to the 2010 one. The 2010 study looked at six balance-sheet items, and stakeholders weren't among them. Thirteen years later the author added this one, and my understanding is that it's because he had seen that relationships cut in a recession can't be bought back in the recovery.

In thirteen years the method hasn't changed; what's changed is whether you dare

Reading the two articles together, the most interesting thing isn't what they say but the distance between them.

The 2010 conclusion was a nine-box grid, a set of probabilities, a playbook of "what to do": save through efficiency, invest across the board, filter by customer. Thirteen years later the same person came back, didn't revise a single number — 9% is still 9% — but wrote at a different level: sensemaking, bootstrap, stakeholder balance. He admits himself that the earlier research "was largely confined to the strategic and tactical level and didn't explore the mindset that enables leaders to embrace protective and forward-looking strategies at the same time."

My own reading is that this distance is itself the answer. The method has been public since 2010: HBR published it, consulting firms followed up with several rounds of research, every CEO could read it. Yet only 9% of companies pulled it off — the bottleneck was never the method, but whether people get bent out of shape by fear. BlackRock's Mark Wiedman said in 2023 that crises send executives into "panic mode — they don't really want to lead, they don't want to lean into anything new."

In Book Five of the Meditations, Marcus Aurelius wrote: the impediment to action advances action; what stands in the way becomes the way. Gulati also cites the Stoics at the end of his article: adversity isn't just a threat to be contained, but a unique opportunity for personal and organizational growth. Put those two lines side by side and they say the same thing as the nine-box grid.

If I were asked in class which type my company is today, and what I'd cut first and protect first in the next recession — my answer would be: cut first everything that amounts to "doing the same with fewer resources" — repetitive manual work, meetings that don't produce decisions, tools that don't change what the client gets; push all of that into systems, which lowers costs permanently. Protect two things first: the people sitting across from clients, and the system that remembers what clients have said. They're Minicucci's "airplanes" — you can't grow without them, and if you let them go in a recession, you can't buy them back at the original price when demand returns.

The price of this choice is a thinner buffer and a slower road to fatter margins. I accept that. Staples' numbers tell me the asymmetry runs in this direction.

Belden began its transformation in its most profitable year; these two articles are about continuing to invest in your least profitable one. Put the two ends together and they amount to the same sentence: the best time to transform and invest is never when it feels safest.

Takeaways for Arcus42

Reading these two articles, what came to mind wasn't families — it was us.

Gulati writes about CEOs in a recession, but a start-up has no "non-turbulent period" — resources are always fewer than the things you want to do, and every quarter you're deciding what to cut and what to invest in. So for us, that nine-box grid isn't a contingency plan; it's daily life. I want to treat the points below as Arcus42's own basic principles for allocating resources in any period — not just advice for a recession, and not just advice for families.

First, cutting and investing are two lists, not one dial. The easiest mistake is to think of resource allocation as a single line — "let's be a bit more conservative this quarter" or "a bit more aggressive." What progressive companies do is keep two lists open at once: the defensive one written very narrowly, the offensive one written very broadly. For us, every quarter should answer two questions at once: which expense can be brought down permanently? Which investment should be bigger than everyone else's? Answer only one and you're prevention-focused or promotion-focused, with odds below 30% either way.

Second, save on process, not on the people facing clients — and don't get the order backwards. A start-up's biggest cost item is people, and the quickest way to save is also people. But in a business built on trust, the people facing clients are the product. The costs that can truly be brought down permanently lie in repetitive analysis, drafting, organizing and meeting notes — the part systems should take over. First let systems take over the repetition, then look at headcount. The worst box in the nine-box grid is exactly the one that gets the order backwards: send the people away first, then invest.

Third, every expense has to pass just one test: does this help us understand this family better? Gulati's filter is "customer needs." For us the question is more specific: does this tool, this process, this partner, this new service line make our understanding of this family deeper and our response faster? If yes, keep it and invest more; if no, cut it. It has another benefit — it turns budget arguments into a question everyone can answer. When we do asset allocation for families, we advise them to use the same question to screen out "wants" and protect "needs"; when spending on ourselves, there's no reason to switch questions.

Fourth, something cheap only counts as an investment if it changes the model; otherwise it's just scaling up. TJX and Target both bought stores when they were cheap; one saw profits fall, the other saw them rise, and the difference was whether they also changed how they made money. For us, tools, licenses and talent will all have their "cheap" moments; the test isn't price, but whether it changes our marginal cost of serving a family. If it doesn't, we're scaling up an old model — revenue can go up, profit won't.

Fifth, the destination stays fixed, the steps can change, and every step can be measured. Uncertainty most easily produces two kinds of paralysis: waiting until things are clear before moving, or moving without looking. Edward Jones took the middle road: the destination didn't move; what needed postponing was postponed, what needed strengthening was strengthened, every investment came with an evaluation metric, and everything was prioritized by value. For us, before every new AI capability or new service goes live, we should write down one sentence — "how will we judge whether it worked?" — and do the most profitable line first.

Sixth, let every party hurt a little less, instead of making one party bear it all. A small company's stakeholders are clients, shareholders, the team and partner advisers. Practical idealism doesn't mean letting one of them bear everything when things get tight, but having everyone give a little and everyone stay at the table. Especially the partner network — lawyers, tax advisers, private banks, investment banks — the easiest thing to "optimize" away and the hardest to buy back.

Seventh, courage has to be voiced by the founder first, and backed by her own decisions. Pennington said "not investing is the more dangerous option" in the very quarter when her peers were all laying people off. Its effect isn't to inspire, but to let everyone in the organization know: at this company, moving forward is allowed. Only the founder can say it — and she has to prove it in the quarter when revenue is tightest, with money that still gets invested anyway.

Final Thoughts

Before class, the line I agree with most is: "offense is earned by defense; it isn't a substitute for it."

---

These are pre-class reading notes, compiled from the assigned Harvard course readings. All data cited comes from the original texts and does not constitute investment advice regarding any company or industry.