Manage a Living System, Not a Ledger
Lean accounting is part of the management problem besetting US industry
By H. Thomas Johnson
Professor of Business Administration
Portland State University
"Lean accounting" embodies a fatal flaw common to all "lean" activities such as lean manufacturing, lean enterprise, lean leadership, and lean sustainability. The adjective "lean" merely denotes particular features of a business activity that presumably improve the activity's performance, but do not change the principles that shape how it is carried out. Such features usually focus on reducing every conceivable form of waste, including excess variation (mura), excess burden (muri), and excess consumption of time, energy, and material (muda).
But the improved business performance promised by these features can never be more than temporary, because "lean" does nothing to change a host of finance-driven practices that inevitably increase instability and diminish long-run business performance. Examples of such practices include optimizing the efficiency of individual operations at the price of reducing efficiency in the whole; building work-arounds to keep work moving instead of solving problems that cause delay and interruptions; and sacrificing quality and lead time to increase output and meet standard cost-variance targets.
Causing those destructive practices is the assumption that financial information not only defines the purpose of business, it also provides the primary means to control the financial outcomes of a business. Until that idea is abandoned and the practices it spawns cease, there is no reason to believe that "lean" in any of its forms, especially lean accounting, can improve the long-run performance of businesses or the economy as a whole. This becomes evident when one examines why companies seem unable to match Toyota's financial performance.
A key reason why American companies fail to emulate Toyota's long-term financial results is their belief that managers can use financial targets as "levers" to control those results. Of course, Toyota and all American companies use basically the same accounting systems to compile measurements for reporting financial results. Unlike Toyota, however, an American company also assumes it can use the same linear and one-dimensional financial information to understand, explain, and control those results, even though the results emerge from nonlinear and multidimensional operations. This difference in thinking between Toyota and American companies reflects two views of reality, only one of which provides a sound basis for running a business.
Following the pre-1900, Newtonian view of reality as a collection of independent parts that move only under the influence of external forces, business educators have trained generations of American managers to believe that a company's overall financial result is the sum of individual results in each and every part of the organization. Supposedly, the results in all the parts add up linearly to produce the organization's total result.
If such a linear cause and effect relationship connects the parts to the whole, then to change the whole by a desired magnitude requires no more than changing one or more of the parts by the same magnitude. In other words, to reduce operating costs and increase profit by $1 million requires no more than reducing payroll in that amount by steps such as laying off personnel, reducing wages and salaries, or outsourcing work to a lower-wage region. It doesn't matter how the reduction is accomplished, since removing a dollar in one part of the business has the same impact as removing a dollar in any other part. All that is required is to change the financial size of some part or parts by the magnitude of the desired total financial change.
Toyota's management does not seem to think this way. Following a view of reality as a network of interdependent self-organizing parts, they seem to believe that the company's long-term financial results emerge from a particular pattern of relationships among all parts of the organization, including customers, suppliers, employees, shareholders, and the communities that sustain the organization. In that context, the company's long-term financial results emerge from countless nonlinear feedback loops in a complex, self-adaptive, and self-corrective living system.
That long-term result is not a mechanistic sum of individual results produced independently by separate parts; it emerges holistically from the system of relationships among a community of interconnected and interdependent parts. Thus, changing spending in the amount of $1 million in some part of the system will not generate an equal change in overall cost and profit in the long run. Indeed, unmeasurable nonlinear feedback loops might cause the long-term change in cost and profit to be a multiple of $1 million, and in the opposite direction. For example, the money saved by replacing high-paid experienced employees with lowwage contract workers in another country is likely to result in more defects, longer lead times, and increased customer dissatisfaction.
Toyota seems always to have known this, because their strategy for reducing cost and increasing profit is not to cut costs or pump up revenues; rather, it is to continuously improve the system of relationships among all parts of the business, in ways that eliminate waste and increase flow. In other words, they believe that improving the system, not simply moving or changing its parts, is the surest way to improve long-term financial results.
It's hard for Americans to understand the idea that a business organization cannot improve its long-run financial results by working to improve its financial results. But the only way to ensure satisfactory and stable longterm financial results is to work on improving the system from which those results emerge. That is what Toyota has always done, and that is what most discussions of "lean" ignore. Unfortunately almost all "lean" initiatives in American companies fail to see that American-style financial "control" procedures and Toyota-style management don't mix, even with "lean" accounting practices in place.
The fact that Toyota makes virtually no use of management accounting targets (or "levers") to control or motivate operations is no doubt an important reason why Toyota's financial performance is unsurpassed in its industry. Toyota focuses its operations on continuous system improvement through endless rapid problem solving. And they emphasize genchi genbutsu, or "going to the place," to see where a problem occurs, firsthand. They don't rely on second-hand reports or tables and charts of data to achieve a true understanding of root cause. Instead, they go to the place (gemba) where you can watch, observe, and "ask why five times." This attitude shows a deep appreciation that results (and problems) ultimately emanate from, and are explained by, complex processes and concrete relationships, not by abstract quantitative relationships that describe results in simple, linear, additive terms.
It should not be surprising, then, to realize that managers in a Toyota plant, unlike their counterparts in American organizations, do not refer to accounting documents such as standard cost-variance budgets to discuss the state of current operations. As I write in my most recent book Profit Beyond Measure, the Toyota accounting system treats daily plant operations essentially as a "black box" that it does not enter. Accountants, of course, record everything that goes into the plant, and all the products that come out. But within the plant they don't track the flow from incoming resources to outgoing finished product. Everything one needs to know about the transformation that takes place inside the plant is inherent in the flow of the work itself. In the Toyota Production System, the work provides all the information needed to control its state.
All information needed to control a Toyota plant's operations is in the work simply because all work flows continuously, at a balanced rate, through virtually every operation—from the beginning to the end of the manufacturing process. The work has been carefully designed so that one can quite literally "see" its current state.
Is it on time to meet the day's orders? If not, how much additional time will be needed? Have defects or other errors occurred along the way? Are components to final assembly being replenished on a timely basis? Has any undue inventory accumulated anywhere? Are problems being identified and addressed according to standard procedures? Such questions, and hundreds more, can be answered at every moment in every step of the process throughout the plant. No accounting system can alert managers as well or as fast if anticipated costs and revenues will not be achieved. Any "exceptions" that managers might need to address to keep financial results on track are visible in real time as the work is being done, not days, weeks, or months later in a report from the accounting department.
If traditional management accounting practices are the key problem preventing American businesses from emulating Toyota's performance, what should companies do? Many proponents of lean accounting suggest that companies should reform management accounting by doing things such as activity-based value-stream costing, direct costing, cash-flow accounting, value-added capacity analysis, and more. These proposals should cause a sense of deja-vu among those old enough to recall the proposals put forth some 20 years ago to gain better control over burgeoning overhead costs with activity-based cost (ABC) information. ABC seemed like a good idea at the time, but in retrospect it was a good answer to the wrong question.
We see better today—when we understand more fully what Toyota does—that reducing manufacturing overhead costs requires a better way to organize work, not better cost information. The question that proponents of ABC should have been asking was how to organize work to eliminate the causes of overhead activity, not how to trace costs of overhead activities to products in more discriminating ways. Perhaps now is the time for companies interested in becoming "lean" to reframe the question that management accounting control systems are supposed to answer. It's time to recognize that management accounting controls are a good answer to a wrong question, that if the question were properly reframed, management accounting controls probably would not be a valid answer.
The question most companies now ask is how to control the financial results of business operations. Inherently, this question assumes that financial results are a linear sum of individual contributions from separate parts of the business. Accounting control information seems the logical way to show how those contributions, and changes in them, add up to the organization's overall financial results. But if we assume that financial results emerge from complex interactions and nonlinear feedback loops in the interrelated parts of a natural living system, then attempting to control those results with linear accounting information is not only erroneous, but possibly destructive to the system's operations in the long run. In that case, the new question should be: how does one control the financial results emerging from operations that abide by the principles which govern a natural living system?
An early answer to this question was provided in the 1930s and 1940s by Walter Shewhart and W. Edwards Deming, both trained in mathematical physics, and both experienced in using state-of-the-art statistical tools in business and government. One of their lasting contributions was to devise a scientific way to estimate "control limits," within which a business system's results would normally fall until one of two steps were taken that altered the limits. One step was to ignore all but abnormal variation in results and work to improve the system itself, thereby narrowing the control limits and improving long-term performance. The other step, a less desirable but more common way of managing, was to try to improve long-term performance by intervening in the system every time results varied from a desired target. The inevitable consequence of the second step, Shewhart and Deming proved, is to widen the system's control limits, and thus impair its long-term performance.
In essence, Shewhart and Deming likened a well-designed business system to a living system in nature. Its results vary over time, but the range of variation has limits. In a human social system such as a business, however, managers can improve performance by taking steps to reduce that range of variation. The key to performance improvement, then, is to nurture the system that produces results, not to drive the system to achieve targets that fall outside its normal performance limits. In his early work, Deming articulated 14 principles that defined what he meant by nurturing the system. Those principles included things such as: create constancy of purpose; constantly improve systems by reducing variation; cease dependence on inspection; do not base purchases on price alone; do not reward individual performance; institute training; eliminate management by objectives; and more.
This is precisely the approach that Toyota relies upon to manage its operations. Toyota lives by a set of deep underlying principles that, after observing their system on many study missions to their plants in the 1990s, I define in Profit Beyond Measure as "managing by means" (MBM). The essence of MBM, which compares Toyota's system to a living system, is that satisfactory business results follow from nurturing the company's system (the "means"), not from manipulating and wrenching its processes to achieve predetermined financial results (a mechanistic strategy popularly known as "managing by results"). In a recent—and excellent— synthesis of Toyota's system principles, Jeffrey Liker, professor of industrial and operations engineering at the University of Michigan (Ann Arbor), articulates a similar concept in his book The Toyota Way with the phrase, "creating the right process will produce the right results."
This sentiment is central to the Toyota organization's deep-seated belief in the importance of defining the properties their operating system should manifest, and then having everyone in the organization work to continuously move the system toward those properties. Frequently one hears Toyota people use the label "True North" to refer to system properties such as safety (for employees and for customers), moving work always in a continuous flow, one order at a time on time, with no defects, with all steps adding value, and with the lowest possible consumption of resources. The assumption is that the more that every process in the system manifests the properties of True North, the better will be the company's long-term performance.
These three approaches to managing operations—the Shewhart-Deming approach, managing by means (MBM), and the Toyota Way—all suggest how different it is to nurture the system that produces a company's financial results than it is to force the system to produce a desired result beyond its current capabilities. The latter strategy is, of course, followed by virtually all large companies in the US today, especially the large publicly traded companies whose top managers are pressured to deliver results demanded by financial markets and other outside interests. Many of those companies now pursuing lean initiatives in the expectation of achieving performance like Toyota's will not, or cannot, forego pressure to drive operations with management accounting "levers of control." That makes the likelihood of their performing at Toyota's level nearly zero.
Managers who look primarily at financial information to judge the performance of a business are certainly working in the dark, unless I am mistaken and the operations they manage actually behave according to mechanistic principles. But anyone who is aware of modern life science can never again view a human social organization, such as a business, as anything but a natural living system. That being the case, it stands to reason that the key to favorable long-term financial performance is to design and run operations according to living-system principles. Such principles resemble Deming's 14 points, the principles of MBM, or The Toyota Way. Only if a company can describe its operating system in terms of such principles can it know whether or not the system is improving.
Financial quantities cannot reveal if a system is improving or not. Accounting information, lean or otherwise, cannot be used to "motivate a lean transformation." If a company requires cost information of any kind to show the "savings" from "going lean" it is lost, and will never get there. Requiring cost information to justify taking the steps that are necessary to emulate Toyota discourages people from continuously removing sources of delay and error that stand in the way of moving closer to achieving living-system principles.
Instead, to avoid problem solving they will create work-arounds such as rework loops, forks, and inventory to keep work moving (even if it is not continuously flowing) in the hope of eliminating unfavorable unit-cost variances. In other words, the demand to justify operational decisions with cost information causes people to forego root-cause problem solving and, instead, to build "cost-effective" work-arounds that violate sound system principles. Eventually, the system principles are forgotten, and managers spend increasing amounts of time working to improve the efficiency of the work-arounds.
No company that talks about improving performance can know what it is doing if its primary window on results is financial information and not system principles. No amount of financial manipulation will ever improve long-term results. Performance in the long run will improve only if managers ensure that the system from which the performance emerges adheres more and more closely to principles resembling those that guide the operations of a living system. All companies that intend to perform like Toyota should recognize that they must follow a systemic path marked by living-system principles. They will never get there by trying to motivate and direct "lean" initiatives with "lean accounting" and management accounting "levers of control."
H. Thomas Johnson
H. Thomas Johnson is Professor of Business Administration at Portland State University (Portland, OR). This article draws from his essay Lean Dilemma: Choose System Principles or Management Accounting Controls—Not Both in Lean Accounting and Performance Measurement: the Relentless Pursuit of Perfection, Joe Stenzel, editor, (New York: John Wiley, 2007), Chapter 1. His most recent book, Profit Beyond Measure: Extraordinary Results Through Attention to Work and People (New York: The Free Press, 2000), was awarded the 2001 Shingo Prize for Excellence in Manufacturing Research. Further information on Johnson's career and his other books is available at: http://www.sba.pdx.edu/faculty/tomj/tomj.htm. He may be reached by e-mail at: firstname.lastname@example.org .
This article was first published in the December 2006 edition of Manufacturing Engineering magazine.