The Market Does Not Price Productivity

A better question is harsher and more useful. What if the market is not failing to value productivity. What if we are failing to make productivity believable as a controllable, repeatable capability that survives pressure?

The False Certainty. The assumption that productivity automatically becomes value

Productivity is treated like gravity in manufacturing. If you improve it, everything else should follow. Lower cost, higher margin, more cash, better resilience, higher valuation. It is the cleanest story in the operating world. It is also the story that keeps failing in the capital world. Not because productivity does not matter. It matters more than most teams admit, especially when labor is tight, volatility is normal, and pricing becomes contested. The failure is simpler. Most productivity programs improve results without improving credibility. They create performance, but they do not create controllability in the eyes of the market. That distinction is the gap between operators and investors. Operators see the mechanism. Investors see the risk that the mechanism is not durable. The market is not a scoreboard for effort. It is a discounting machine for uncertainty. It does not reward the presence of improvement. It rewards evidence that improvement is a managed property of the enterprise.

The Missing Variable. Controllability is what gets paid

Controllability is not a buzzword. It is a capital concept expressed in operating language. It means management can explain the drivers, measure the drivers, intervene on the drivers, and repeat outcomes across time, across plants, across product cycles, and across leadership changes. It means variance compresses. It means guidance becomes less fragile. It means cash conversion becomes less episodic. It means shocks do not rewrite the plan every quarter. Investors are not allergic to cyclicality. They are allergic to ambiguity. Cyclical businesses with disciplined operating systems often command respect because they behave like engineered systems inside volatile markets. Other businesses, sometimes with similar margins, get treated like weather. You can report it, but you cannot run it. This is why the same cost improvement can be priced two different ways. If the market believes the cost improvement is a one-time clean-up, it gets treated like a temporary lift. If the market believes the cost improvement is evidence of a repeatable operating capability, it gets treated like a structural change. Structural change is what expands multiples.

The tragedy is that most enterprises talk about productivity in a way that signals the first category. Projects, initiatives, programs, savings. Those words sound like episodic effort. They do not sound like controllable systems.

The Evidence. Productivity separation exists, yet the pricing gap stays

The Industrial Productivity Index work is useful because it forces a serious conversation. It separates companies by industrial productivity growth over time and shows a real gap between “Pathfinders” and “All others.” The spread is not small. It is the kind of separation that should have strategic consequences. In the 2018 to 2024 window, Pathfinders show dramatically higher IPI growth than the rest of the cohort. Yet when you look at market cap growth in the same summary view, the line does not behave the way the operating story expects. On the surface it looks like a contradiction. Productivity leaders, lower market cap growth. That line should not be dismissed. It should also not be used as a weapon against productivity. It is a forcing function. It forces the right questions about what is being measured, what the market is actually responding to, and what causal chain is missing between shop-floor advantage and investor belief. Three things are almost certainly true at the same time. First, market cap growth is not shareholder return. It ignores dividends, buybacks, and share count reduction. It can understate shareholder outcomes when capital returns are high. Second, productivity can improve margins and cash while still failing to expand the multiple if the market does not believe the improvement is durable or repeatable. Third, the market may already be pricing “controllability” through signals that are adjacent to productivity but not identical to it. Guidance accuracy. Volatility. Working capital discipline. Capital allocation. Evidence of an operating system that survives leadership rotation. Those are belief variables. So the paradox is not a refutation. It is an invitation to build the causal model properly.

Why language matters. The market reacts to belief updates, not effort

This is where most people get nervous because it sounds like spin. It is not spin. It is signal.

Markets move on belief updates. Belief updates come from evidence, and evidence is communicated through disclosures, guidance, operating commentary, and the consistency between what is said and what is delivered. Peer-reviewed finance research has repeatedly shown that language in disclosures and conference calls contains information the market responds to. Tone, readability, certainty, and the way management frames drivers are associated with returns, volatility, and other market reactions. That does not mean executives can talk their way into valuation. It means the market is listening for something specific, and that something is usually a proxy for controllability. Here is the practical translation. The market is not rewarding optimism. It is rewarding coherence. Coherence looks like this. Drivers are named. Levers are clear. Trade-offs are explicit. Outcomes are quantified. The logic is consistent quarter to quarter. When conditions change, management explains why results moved in causal terms, not in vague generalities. Over time, the organization earns a reputation for being able to run the system rather than merely describe it. That reputation becomes a discount-rate advantage. It is not glamour. It is trust.

The operating system is the bridge. It turns productivity into a believable capability

Most enterprises cannot prove controllability because they do not actually have it. They have pockets of excellence and bursts of improvement. They do not have an operating system that makes performance repeatable. This is why the Owens Corning framing is so instructive. It does not present TPM as a tool. It presents it as an operating system. It does not present digital as the outcome. It calls digital the enabler, not the outcome. It ties the operating system to an explicit annual productivity target in cost of goods sold. That is not marketing language. That is architecture. It tells investors, employees, and the board that productivity is not a campaign. It is a governed property of the enterprise. This is the heart of the argument. Productivity is not what the market struggles to price. The market struggles to price productivity when the enterprise cannot prove it is controllable. An operating system is how you prove it. Not by saying “we have one.” By showing the causal chain, the cadence, the metrics that actually drive outcomes, and the durability of results through time.

The causal chain. How productivity becomes investability

If you want a causal model that can stand up to a CFO, a board, and an investor, it has to connect mechanisms to money and then connect money to belief. That is the missing middle in most productivity narratives. Start with the mechanism layer. Productivity comes from loss removal. Loss removal comes from stability. Stability comes from disciplined operating systems that govern equipment, flow, quality, labor development, and problem solving cadence. That is where TPM, Lean, and the management system matter. Not as slogans. As control structures. Then comes the translation layer. Loss removal reduces conversion cost. Stability improves schedule attainment. Better schedule attainment reduces the need for buffer inventory and expediting. Lower inventory and fewer surprises release working capital. Released working capital increases free cash flow without relying on pricing. Higher free cash flow creates more strategic degrees of freedom. You can reinvest with intent or return capital with discipline. Then comes the belief layer. When free cash flow improves because pricing masked the problem, investors treat it as temporary. When free cash flow improves because conversion costs structurally decline and variance compresses, investors treat it as repeatable. Repeatable is controllable. Controllable cash flows get discounted less. Discounted less means multiples expand or compress less in downturns. This is the link most teams do not model. They model productivity to margin. They do not model productivity to variance. They do not model variance to credibility. They do not model credibility to multiple. That is where the market is actually pricing. If you want one sentence that boards understand, here it is. Productivity becomes value when it becomes credibility.

The pricing trap. Pricing can hide the truth, then punish you later

Pricing is not evil. Pricing is a lever. The problem is reliance. In the inflationary window, a lot of companies learned a dangerous lesson. They discovered they could protect margins without fixing conversion. Investors accepted it for a time because everyone was doing it and the macro regime supported it.

Then conditions tightened. Demand softened. Customers pushed back. Competitors got aggressive. The same companies were exposed because they had not built controllability. They had built a story. This is why pricing is not a substitute for productivity. It can temporarily protect earnings. It cannot create operational control. In some cases it signals the absence of control, especially when it is the only lever management can articulate with confidence. A real causal model has to include “pricing reliance” as a node, because pricing reliance is often a symptom of unaddressed conversion weakness. You do not have to moralize it. You model it, then you see what it predicts.

Incentives are not adjacent. Incentives are causal parents

Boards often treat compensation as governance hygiene. In reality, compensation is an operating architecture decision. It shapes what gets protected and what gets starved. If senior incentives overweight short-term EPS, leaders will rationally favor levers that show up fast. Pricing actions. Buybacks. Deferred maintenance. Cuts that erode training throughput. Reduced investment in the operating system. Those are not character flaws. They are predictable responses to incentive design. If incentives overweight multi-year operational stability, loss removal, cash conversion, and repeatable COGS productivity, leaders will rationally invest in the operating system because it becomes the best way to win. This is why an “incentive misalignment” measure belongs inside the causal graph. Not in a separate governance appendix. Inside the model. It is upstream of reinvestment rate, maintenance maturity, leadership behavior, and long-run credibility. The market already assumes this is true. Investors read incentive design as a clue about whether a company is being run for controllability or for optics.

What the COO Council can do with this. Change the unit of analysis

The highest-leverage move here is not another productivity playbook. It is changing what the Council asks for. Most conversations ask, “Did productivity improve.” The better question is, “Under what conditions does productivity become investable.” That second question forces a causal hierarchy.

It forces the right dependent variable. Total shareholder return and multiple behavior, not market cap alone. It forces the right mediators. Free cash flow conversion, working capital release, volatility compression, guidance credibility, capital allocation discipline. It forces the right moderators. Macro regime, sector tailwinds, baseline multiple, commodity exposure, demand volatility. Once you ask it that way, the work becomes buildable. You stop arguing about whether productivity “matters” and start building the model that explains when it gets paid and why. This is where LNS work becomes uniquely valuable. It already speaks the language of operating systems and management mechanisms. The Industrial Productivity Index work already produces separation. The Council can connect those to the missing belief layer and produce something the market actually recognizes. Not another index. A causal knowledge system.

The build. The causal knowledge document we actually need

A causal model that can survive a CFO has to be explicit about what it knows, what it assumes, and what evidence supports each edge. That means a node ontology that includes context, operating system maturity, mechanism metrics, financial translation, capital allocation, narrative specificity, and credibility history. It means edges that are written as mechanisms, not correlations. Downtime reduction causes throughput stability. Throughput stability causes inventory reduction and service level improvement. Inventory reduction causes working capital improvement. Working capital improvement causes free cash flow improvement. Free cash flow stability and credibility reduce perceived risk. Reduced perceived risk supports multiple expansion or reduces multiple compression in shocks. It means evidence objects, not vibes. A target like “2 percent annual productivity in cost of goods sold” is an evidence object. “TPM is our operating system” is an evidence object. Guidance accuracy across quarters is an evidence object. Maintenance capex behavior is an evidence object. You tag them. You timestamp them. You connect them to edges. It means you treat language correctly. Not as persuasion. As a belief signal that is only valuable when it tracks reality. When it diverges from reality, it becomes an early warning indicator, not a strength.

This is how you build a model that does not collapse into storytelling. You make storytelling accountable to evidence.

The better board question. What must be true for our productivity to be priced

If you want one question that changes behavior, use this. What would have to be true for the market to treat our productivity as inevitable. That word, inevitable, is the whole point. It forces the organization to stop celebrating improvements and start engineering controllability. It forces the COO to define the operating system, not the initiative list. It forces the CFO to track variance compression and cash conversion, not just margin. It forces the CEO to align capital allocation with capability-building, not just quarterly optics. It forces the board to design incentives that reward the causal chain, not the headline. It also forces honesty. Because if productivity is real but not priced, one of two things is true. Either the market is missing it, or the enterprise has not proven it is controllable. The second answer is more likely. It is also the one you can fix.

Closing. The market is not unfair. It is precise about what it needs

A productivity story that cannot explain how it repeats is not a value story. It is a performance anecdote. Markets do not pay premiums for anecdotes. They pay premiums for enterprises that look engineered. Enterprises where results are not a lucky quarter, but the predictable output of a controlled system. That is why the market does not price productivity in the way operators expect. It prices controllability, because controllability is what turns productivity into durable cash flows and credible guidance. It is what reduces fear during volatility. It is what keeps the multiple from collapsing when conditions tighten. The practical implication is demanding. It requires a shift from measuring outcomes to proving mechanisms. It requires causal models that connect operating reality to financial translation and then to belief formation. It requires leadership teams to stop speaking in the language of effort and start speaking in the language of control. That is not a communications upgrade. It is an operating upgrade.

When an enterprise becomes controllable, productivity is no longer a claim. It is a property. Then the market starts treating it the way operators always assumed it would.

References and sources

Industrial productivity and Pathfinder separation are sourced from LNS Research, IPI Results 2025 (Pathfinders versus All Others IPI growth over 2018–2024, and the paired median financial comparison view that includes market cap growth alongside operating margin and free cash flow). The operating-system example and explicit productivity target language are sourced from LNS Research, LNS IX 2025 Final (Owens Corning framing TPM as an operating system, positioning digital as an enabler, and stating a quantified annual productivity target in cost of goods sold). The claim that markets treat corporate language as analyzable, incremental information is supported by Loughran and McDonald, “When Is a Liability Not a Liability? Textual Analysis, Dictionaries, and 10-Ks,” Journal of Finance (showing that generic sentiment dictionaries misclassify financial language, motivating finance-specific word lists that are widely used in empirical work on disclosure tone and market outcomes). Wiley Online Library+1 The claim that disclosure readability is economically meaningful, and relates to earnings quality and persistence, is supported by Li, “Annual Report Readability, Current Earnings, and Earnings Persistence,” Journal of Accounting and Economics (reporting that harder-to-read annual reports are associated with poorer performance, and easier-to-read reports with more persistent positive earnings). care-mendoza.nd.edu+1 The claim that earnings-call wording carries incremental information the market reacts to is supported by Price, Doran, Peterson, and Bliss, “Earnings Conference Calls and Stock Returns,” Journal of Banking and Finance (finding that conference call linguistic tone predicts abnormal returns and trading volume, beyond other disclosed information). ScienceDirect+1 The claim that “market cap growth” can diverge from shareholder experience because shareholder return includes reinvested distributions is supported by Morgan Stanley Investment Management’s research note defining total shareholder return as a capital accumulation rate with reinvested dividends. Morgan Stanley The payout and total-return framing, including why dividends and repurchases matter to what owners actually receive, is further supported by CFA Institute materials on dividends and share repurchases. CFA Institute+1 The index-math treatment of total return, specifically that total return indices reflect price movement plus reinvested dividends, is supported by S&P Dow Jones Indices methodology and related documentation. S&P Global+1 McKinsey’s corporate finance work is used to support the narrow point that value creation is driven by operating cash flow and returns on capital, not by the mix of payout mechanics, even though the mix affects per-share optics. McKinsey & Company+1