By JAMES FIELDING SMITH
James Fielding Smith is at the Henry County Public Service Authority, P.O. Box 7, Collinsville, VA
24078. E-mail address: jfsmith@neocom.net
PBB is more vulnerable to threats from fraud, falsification, and misrepresentation than
were previous budgeting systems. Vulnerability grows from complexity, dependence
on performance data, year-to-year changes in strategic focus, and the great expansion
of the numbers of kinds of specialists involved in the process. It appears to grow out
of PBB's widely recognized benefits. Transparency has been put forward as the primary
protective device against threats to PBB; however, this has problems with both
definition and operationalization. The fix may be a number of coordinated measures,
most importantly the inculcation of sound ethical standards throughout an organization
using PBB.
Growing from a desire to improve service and the efficient utilization of resources,
performance-based budgeting (PBB), sometimes also called performance budgeting,
has begun to penetrate all levels of government. The following definition from the
State of Maine captures both the meaning and flavor of PBB:
Performance budgeting links spending to results. Unlike line-item budgeting, it does not budget
around categories of spending . . . nor does it use the previous year's budget as a starting point.
Performance budgeting allocates resources based on achieving specific measurable outcomes.
Rather than asking 'what will be needed in personnel and other costs to carry out an assigned
task?' The first question is 'what is the outcome to be achieved?' Thus performance budgeting
first asks not 'how many people do we expect to put through a job training program?' or 'how
many permits have to be issued to protect this river?' but rather, 'how many people do we intend
to place in jobs?' or 'how many miles of river will be fishable and swimmable?' The number of
people placed in jobs and the miles of river that are fishable and swimmable are outcomes and
they are the basis of performance budgets.... Outcomes are defined through a strategic planning
process that considers the critical issues facing an agency, the agency's capabilities, and input
from stakeholders such as citizens, customers, legislators, and employees.'
From a handful of mostly state programs prior to 1990, PBB has spread to nearly all
states,^ the federal govemment, and many local governments, special use districts, and
non-governmental organizations.^ PBB may soon be the de facto standard for public
budgeting. Many perceive PBB as clearly superior in an atmosphere of resource constraint
for optimizing both the effectiveness and the efficiency of programs. Down
slope agencies feel the cumulative effects of PBB's enthronement in federal rules
governing transfer payments, grants, delegations, and devolution of powers or programs.
Less obviously, yoking budgeting and strategic planning so directly in PBB has
nearly doubled the direct constituency that chooses or implements budgeting systems.
Accepting as a given that PBB is superior to previous budgeting systems, an assumption
that probably deserves serious examination elsewhere, this article addresses
the premise that PBB is more vulnerable to threats from fraud, falsification, and
misrepresentation to earlier budgeting systems. An examination of the sources and
types of this vulnerability will lead to a proposed approach to preventing fraud, falsification,
and misrepresentation. This article will review the potential threat mechanisms,
the magnitude of the possible threats, and possible countermeasures.
SOURCES OF VULNERABILITY
Each advantage or benefit of PBB seems to introduce at least one new risk or threat to
data integrity, as is shown in Table 1 on the following page. The sources of vulnerability
fall roughly into four categories: (1) same as other budgeting systems but heightened;
(2) introduced from strategic planning; (3) unique to PBB; and (4) temporary, or
learning curve, problems.
Same as Other Budgeting Systems but Heightened
Any threat operating in line-item, program, performance, or other budgeting system
will also threaten data integrity in PBB. Although the more rigorous degree of priority
setting in PBB will make fewer data points or streams appear vulnerable and needing
protection, PBB's explicitly defined linkages of fiscal data to performance data can
enable amplification and ramification of errors, thereby heightening their effects. Managers,
budgeters, and planners dealing with PBB should try to build on past experience
in evaluating vulnerabilities and agency responses.
Organizational structure or behavior may harbor threats to data integrity in any
budgeting system. All budgeting systems suffer from the results of a lack of incentives
for team implementation; for example, the assignment of non-overlapping, strictly
individual responsibility for each task and subtask. This remnant of the managementby-
objective culture can affect data integrity in PBB by distorting performance measures
to follow organizational units or chains of command more exactly than do the
organization's processes. Despite its potential for promoting teamwork, PBB remains
vulnerable to parochialism. Lee, Melkers, and Willoughby, in assessing state budget
practices, have noted changes to traditional budgetary relationships to respond to
PBB's special requirements.''
Temporal relationships affect the applicability of long-term or trend analysis to any
budgeting system for detecting aberrations in budgeting. However, as budgeting has
evolved from line-item to PBB, the focus has shifted on categories labeled with major
missions and away from a stable list of line-items. As the major goals evolve and
change from budget cycle to budget cycle, systemic trends may be obscured by the
more obvious changes in higher categories. Another environmental example would be
the succession of major emphasis programs in water quality since 1972: primary
treatment, then, following at roughly four to five year intervals, secondary treatment,
nutrient removal, toxicity, and currently, metals and animal waste. Each new focus
brings very different outcomes and outcome measures, even if the most basic process—
here, processing permits—carries on throughout. Budgets inevitably change
from year to year as programs and subprograms grow, reach maturity, and end. Temporal
changes can greatly complicate year-to-year comparisons and long-term analysis.
Introduced from Strategic Planning
Just as PBB shares categories of threats with earlier budgeting systems, it also shares
threats inherent to strategic planning. Oversimplification and temporal inconstancy, as
noted above, along with parochialism apply but become particularly important in the
key strategic planning functions of stating vision, articulating a mission statement, and
specifying goals and objectives. Any problem affecting or even shading the operationalization
of these higher concepts represents a particular threat to PBB. For twenty-six
years the ultimate goal of the Clean Water Act has been to make all the waters of the
United States "fishable" and "swimmable," but these terms have undergone periodic
redefinition to fit changing economic realities, technological advances, shifts in programmatic
emphasis, national policies, and relations between EPA and the states. Each
redefinition would have affected PBB, had PBB been in use all this time.
Unique to PBB
PBB's own special sources of threat come most directly from PBB's complexity.
Everyone advertises PBB as procedurally transparent and logically accessible, and
these characterizations do apply to the basic concept. However, building, maintaining,
and managing a PBB system involve daunting levels of complexity. Each little niche
of complexity can become an opportunity for error, fraud, falsification, or misrepresentation.
These key data processes can be disrupted or distorted in any segment of
PBB: strategic planning, defining performance measures, reporting performance data,
reporting or apportioning expense data, classifying outcomes, defining of success,
developing budgets, implementing the budget correction process, analyzing performance
data, analyzing cost data, and other areas.
Because of this complexity and the number of major skills—virtually separate specialized
careers—involved, no one person or type of specialist can master the whole
PBB system. Teamwork by necessity could result from this but, on the other hand, no
one manager can master and oversee all the contributory processes.
PBB, when it makes fiscal performance links, creates a problem where none existed
before. Separately, performance auditing and financial auditing have developed to
serve the needs of their targeted substrates; however, performance auditing is significantly
less well developed as a technique and as a field than financial auditing. This is
true even in the private corporate sector where, under the term "management accounting,"
performance auditing has long been practiced. Combining financial and performance
auditing for PBB can create mismatches and lacunae, allowing threats, which
were not threats to the separate branches, to function. Comprehensive auditing for
PBB will require many new techniques and approaches. For example, consider the
challenges of erecting confidence limits around correlations within a model.
PBB works better than other budgeting systems because it drives out the superfluous
and not required, matching funding with real priority projects. However, by undercutting
traditional politics to do this, PBB seems likely to accentuate politicization of
goals within the organization. Internal politicization may, in turn, link up with traditional
politics. Regardless, politicized competition for funding may engender, at least
among the more farsighted participants, a high degree of politicization of selection and
wording of perfonnance measures, both in the appropriation and authorization phases
and in the public accounting and electioneering phases. Whether politicization in this
case equates to hyperbole will depend on leadership and the culture of the organization.
Specific threats to data integrity. Data integrity faces challenges from unintentional
errors as well as from intentional ones, and sometimes the two types get confounded.
Error also operates both on concept, e.g., theories, models, linkages, cause-and-effect,
and on data. Errors in concepts generally create more dastardly results partly because
they propagate themselves like computer viruses every time they are used in a calculation
or analysis. Confounding of conceptual errors and misrepresentation can also
occur. A rascal, wanting to twist the outcomes of PBB, might yield to the temptation to
create a distortion in embedded concepts. Such misrepresentation, if discovered, might
possibly appear to be a conceptual error, an honest mistake, or the result of a legitimate
difference in interpretation. Data point manipulation to obtain the same level of misrepresentation
almost certainly would exhibit both difficulty of detection and ease of
explanation once detected. Fortunately, errors in concept can easily be detected by
rather simple countermeasures, as will be seen below.
Errors in operationalization of measures are to conceptual error as tactics are to
strategy. Both intentional and unintentional errors can occur here but have less widespread
consequence to PBB integrity than conceptual errors. Ways in which this may
occur include the selection of wrong measures, the use of erroneous modeling assumptions
for complex indicators (e.g., using principal component analysis), the omission of key aspects, the use of inconsistent or inadequate baselines, or the application of
incorrect transformations or inflators. A special case of operationalization error results
from costing errors, especially those involving the misassignment or incomplete distribution
of costs. Abstract aggregation of variables creates performance measures that
are hard to understand out of context. For example, a state environmental agency
might have a goal of a 60 percent return to compliance within one year, which summarizes
thousands of facilities covered by over a dozen permitting and regulatory programs,
not even accounting for the many difficulties inherent in defining a key term
like "compliance."
Some activities can directly inject error into PBB data, as with any other complex
dynamic model or data system. In sample shopping someone selectively forwards data
or selectively discards data. This may result in areal, temporal, or quality misrepresentation
of phenomena. Data fudging involves the direct manipulation of the data to
change its import or meaning. All data points remain, but some or all may undergo
illegitimate change. Both sample shopping and data fudgings how the overall effect of
changing the characteristics of groups of data, e.g., means, medians, measures of
dispersion.
Temporary, or Learning Curve, Problems
PBB seems to ignite a search for completeness and specificity, which may initially
result in over complicated definitions, models, and performance measures. This may
obfuscate essential relationships or, as simplification sets in with program maturity,
lead to discontinuities in baselines and data interpretation.
Newness itself may create error. Most often resulting from a lack of adequate
training or from the overselling of PBB's benefits, this effect can express itself as
misplaced effort, misunderstood directions, procrastination, or approximation, any of
which can cause a differentially higher data error rate in a new system.
Uncertainty of consequences. All organizations newly using PBB, and perhaps
some veteran users, experience uncertainty about the consequences of missing performance
targets. The full range of uncertain consequences must be considered: more or
less funding, dumbed-down targets, loss of confidence, restructuring, morale, reprisals,
or punishing headlines, plus a large range of opportunity costs. Through such uncertainty,
PBB creates another layer of rewards, incentives, and disincentives, possibly
motivating distortions in data. The uncertainty concerning consequences may express
itself, however, in uncertain directions, distorting data either upward or downward,
depending on expectations. Uncertainty and politicization may interact, potentiating
the effects. The extent of uncertainty's effects will depend on whether the sign of the
effects is known accurately throughout the organization, which can take place through
policy, training, documentation, or informal diffusion. The contrarian notion that uncertainty
promotes honesty seems to be unverifiable, at least in the context of PBB.
MISREPRESENTATION
Among all sources of error, the misrepresentation of results undoubtedly causes the
most significant compromises of integrity. Misrepresentation occurs when someone,
who has at least enough authority to make a point stick, claims that the results really
mean something other than what systematic analysis can support. No level of data or
analysis can escape vulnerability to misrepresentation, and at the highest level misrepresentation
gets fed directly to the media and to decision making bodies such as
legislatures. Misrepresentation can have profound costs for an organization. The costs
operate in many dimensions: loss of credibility, program compromise, and future
appropriations, for example. Despite such risks, misrepresentation occurs depressingly
often, and PBB may goad desperate managers into trying to make their organizations
appear more competitive, more effective, or more efficient than reality supports.
Unfortunately, misrepresentation breeds. Managers, technical staff, customer support
staff, and clerical staff throughout an organization can be infected by the willingness
to countenance misrepresentation. Lingering effects of management by objective
may precondition some to be vulnerable to misrepresentation.
Reengineering applications. A special source of vulnerability comes from the high
degree of compatibility, or perhaps suitability, of using PBB as a primary tool during
the reengineering of organizations. PBB's unique powers to drive and reinforce
changes in vision, mission, and organization also have consequences for data integrity.
The high stress environment surrounding reengineering may increase incentives for
misrepresentation or distorted definitions of performance measures, thereby accentuating
problems in PBB implementation. This special linkage between PBB and
reengineering verges on co-dependency and merits careful monitoring.
In summary, PBB's vulnerabilities result from complexity, dependence on performance
data, and year-to-year changes in strategic focus, exacerbated by the great
expansion of the budgeting constituency to include strategic planners and many more
levels of managers. This expanded corps must deal not only with new career
subspecialties but also with newly important interactions among budgeting, strategic
planning, and operations.
COUNTERMEASURES
PBB's complexity invites, or perhaps demands, automation. Not only does automation
facilitate the necessary data flows and analysis, but it can also include programming
for error detection, inconsistencies, and other forms of QA/QC. However, automation
adds to the problem. It can hide manipulations and assumptions. It can give a false
sense of objectivity, sometimes carelessly equated with accuracy. It can reduce
thoughtful analysis while simultaneously promoting it.
Awareness
Awareness encompasses a range of countermeasures. Its most elemental form entails
disseminating knowledge of threats to the entire user community or to key subcommunities.
Many tools can contribute to awareness campaigns, from posters to position
performance expectations, and from caveats written in budgets to comments in budget
briefings. Of the fourteen states listed by NASBO as being primarily PBB-using, ten
had sites on the Internet as of March 17, 1998.^ Examination of nine of these showed
seven to have a statement of the importance of awareness of threats to data integrity,
but none went beyond the statement. There appeared to be a tacit assumption that the
mechanisms in place for protecting fiscal and performance data would suffice for PBB.
Competition and Cooperation
The intense levels of competition implicit in the priority setting process of PBB should
have the ancillary effects of countering threats to integrity. In this competitive atmosphere,
each rival mission and program can expect scrutiny from competitors. The
overall process should squeeze padding and excess, reducing the number and capacity
of internal hiding places for error. As a result, opportunities for successful misrepresentation
should shrink even though some competitors may succumb to the temptation
of sweetening their models or data up front.
Cooperation and teamwork can yield a de facto countermeasure against threats to
PBB integrity. Like most tools discussed herein, they could also subvert the process,
depending on leadership and organizational culture. Such an effort would involve a
conspiracy, a possibility both remote and outside the scope of this paper. Cooperation
among specialists and the resultant teamwork ought to yield shared knowledge and
mutual respect, which will support data and procedural integrity.
Openness
Openness consists of many elements, each of which can contribute to protecting PBB
integrity. The more steps and procedures done in the open, the greater the effort it will
take to propagate fraud, falsification, and misrepresentation. Accessibility of the process,
especially of strategic planning and fiscal-performance linkages, reinforces integrity,
partly through coincidental inspection, citizen scrutiny, and feelings of ownership
in the widest possible range of stakeholders. Clarity means a combination of understandability
and logic. It requires clear statements of consequences, or lack thereof,
and of what outcomes are "good."
Comprehensive Auditing
Since PBB combines existing budgeting, strategic planning, and operational monitoring
techniques, the same auditing and QA/QC tools already used with those techniques
are ready for application to PBB. The problems arise with trying to keep up with a
moving target—after all, PBB is most highly touted for its ability to deal with an
organization undergoing change—and with coping with the linkages and interactions
among fiscal and performance variables. Notwithstanding preexisting tools for auditing
and the challenge of programmatic change, comprehensive auditing presents the
same fundamental Demingesque problem when applied to PBB as when applied, perhaps
in the guise of systematic or intensive inspections, to any process; inspections, no
matter how intense or skillful, can only find the errors but have severely limited
abilities to find the sources of errors. Similar effort invested in inspections will yield
less result in quality or accuracy than a commitment to a process of continuous improvement,
which, in turn, involves awareness, commitment, innovation, and imagination.
Comprehensive auditing will always have a place in PBB to satisfy legal audit
requirements on the fiscal end and, increasingly, on the performance end. When automated,
comprehensive auditing forms the basis of most QA/QC routines and can
estimate random error in the system. On the other hand, systematic error will persist
unless rooted out by something like a continuous improvement process.
Politicization
The accentuated politicization that characterizes the priority setting phase of PBB
offers a remarkably strong tool for detecting and sometimes correcting errors in the
PBB system and data. In this case, politics acts as a public watchdog. This tool
probably acts most strongly when PBB deals with severely constrained total resources
and under conditions of openness.
Politicization may come into play in one rare but powerful way of limiting some
threats to PBB integrity: the intelligent exemption of certain types of programs from
PBB. Despite PBB having been sold as a universally applicable system, some programs
should go into sanctuaries so that the pressures of competition for resources and
performance evaluations do not have any opportunity to distort data or program management.
This draconian solution may have few legitimate applications. Enforcement
programs come to mind as a legitimate candidate because of their fragility, sensitivity,
and lack of control over workloads and the timing of outcomes. Timing's particular
importance arises out of cost-efficiency being the underlying principle of PBB and
time being a critical factor in computing any measure of cost efficiency on both the
cost and performance sides.
TRANSPARENCY
Qne last major category of corrective action remains: transparency. Transparency,
often put forward as the primary protective device, or even panacea, against threats to
PBB, has problems of both definition and operationalization. Transparency has attracted
many shadings of meaning ranging from "obvious" to "entirely in the back-
ground." Peter Droege offers this definition, "Transparency should denote in an ideal
situation: (1) public accessibility of data, (2) the actively managed interpretation of
process and structure, and (3) public announcements as part of an accountability program,
with facilitated discussions of assumptions, process, and implications of outcomes
or decisions."^ Don Krueckeberg says it "involves procedures which are easy to
follow, guided by rules that seem necessary and make sense, no hidden consequences."'
The present author proposes transparency to be a process in which the steps
are visibly or clearly linked to the mission or vision of the organization. Transparency
is a portmanteau that can be deconstructed into many of the concepts described earlier
in this article. Transparency is a desirable and maybe necessary condition, but it is not
a sufficient condition for data integrity.
PUBLIC AND PRIVATE SECTORS
A pair of critically important questions arise: "Why do organizations in the corporate
sector not encounter the problems identified in this article?" and "But if they do, why
don't they care?"* The same budgeting practices—both systems and corrective measures—
have been used widely in the private sector, where the participants cannot be
assumed to be more skillful or more ethical, without encountering the problems identified
in this article. A survey of articles published in Management Accounting between
October 1994 and December 1997 revealed only one article specifically on data integrity,
ethics, or fraud.^ Such findings in the most topical journal in the private sector
may indicate a very high degree of confidence in the integrity of data in PBB in the
private sector. One theory to explain this might be that private organizations assume
that competitive pressure ruthlessly seeks greater efficiency, which drives out major
inefficiencies such as data threats create.
Does this theory apply in the public sector? Consider this argument, which applies
only to the public sectors in democracies. Efficiency and democracy are fundamentally
incompatible, as democracy is never the most efficient way to run anything and its due
process safeguards inherently add inefficiencies to all other public processes. Public
programs are the tools of democracy. They may be forced to try to become more
efficient and otherwise to copy the private sector model, but they are still tied to
democracy. Bureaucracy is where democracy tries to reconcile this conflict. PBB is
bureaucracy's bravest try to date to approach the private ideal, responsibility budgeting.
But due process and other safeguards interfere with pure competition, limiting
how close this approach gets to the ideal. PBB in the public sector therefore needs
countermeasures applied to avoid or control the threats to its integrity.
Klitgaard presents another viewpoint, one which deals with international cooperation
against corruption, necessarily combining private and public issues.'^ He looks at
a case in which data integrity, misrepresentation, fraud, and other issues occur in the
hothouse of monopolistic systems under pressure from bribery or other inducements to
cheat. The democracy-competition theory, although not explicit, undergirds the con-
cept of corruption equals monopoly plus discretion minus accountability, which is
Klitgaard's central thesis.'^ Concluding that corruption is a systemic problem,
Klitgaard proposes a systemic strategy, a strategy that could be characterized as ethical
education.
ETHICS AND THE CULTURE OF PBB
The fix may be a number of coordinated measures, most importantly the inculcation of
sound ethical standards throughout an organization using PBB. The basic problem is
cultural. Although PBB can create another layer of rewards, incentives, and disincentives,
it still operates within the organization's culture. Melkers and Willoughby found
only seven out of the fifty states have formal incentives and disincentives in their
budgeting systems.'^
A comprehensive program based on ethics is needed. All transactions within PBB
should pass the test: is there any hidden incentive for error or disincentive for ethical
behavior? Internal recognition and rewards programs bear examination in this light, as
do personnel policies, especially hiring, promotion, and discipline. Emphasis should
openly go to corrective measures taken to eliminate sources of error. Essentially all
training programs should include units or goals supporting ethical behavior.
The interplay between individual gain and organizational advancement is usually
assumed to be synergistically positive. This may not be true. Consider any case in
which organizational growth and survival will depend on favorable performance measures:
this might cause organizational momentum—an effect pervasive throughout the
organization—to swamp any effects of individual leaders. This is a strong argument
for the inculcation of ethics throughout the levels of an organization. It works both
ways, to restrain a rogue leader or to restrain a rogue organization. Ethics also need to
be fostered among and between agencies, as any one agency will rarely encompass all
of a policy area. Trans-organizational ethics may present the greatest challenge of all.
Many organizations still labor under the burdens of systematized distrust, which is
the legacy of MBO and theory X management. This legacy seriously impedes culture
change, as would be needed to mobilize ethics to protect the integrity of PBB. On the
other hand, its effects on morale and unit identity are often recognized within an
organization and could be used to energize cultural change. Mindless cutthroat competition,
as fostered by MBO, is a serious threat to PBB integrity.
Klitgaard's anti-corruption strategy combines punishing prominent offenders ("fry a
few big fish"), involving "the people" in diagnosing corrupt systems, focusing on
prevention by repairing systems, creating reform incentives, showing cultural units
they are not alone in these problems, mobilizing creative competition, and providing
"tool kits" filled with best practices for fighting corruption.'^ Klitgaard and this article
agree that disseminating awareness and analytical capabilities widely within organizations
or cultures gives the best protection against such things as fraud, falsification,
and misrepresentation. To the extent that transparency promotes these, transparency
contributes to the protection of PBB. Transparency alone, however, does not suffice.
PROSPECTS
PBB's perceived strengths and advantages come largely from its superior facility for
tracking environmental and programmatic change. It forces the size of programs to
what is appropriate now, or in the next budget cycle, and it imposes life cycles on
programs, ending programs that have finished their purpose or outlived it and bringing
on programs to fulfill new mandates or needs. PBB has a strong analog in the phenomena
of ecological succession, where resource constraints and changes create a more-orless
predictable sequence of biological populations or communities.
Despite possible problems, PBB with its overwhelming advantages in terms of
productivity remains the best approach around. In addition, PBB has palpably greater
philosophical soundness than other budgeting systems. It expresses the philosophy "if
it's not paid for, it's not real." And if it's not real, it shouldn't be paid for.
Is PBB more susceptible to fraud, falsification, and misrepresentation compared too
previous other budgeting systems? Yes. Can it be protected? Yes. A number of corrective
and preventive mechanisms, including the systematic promotion of ethics throughout
the organization, can protect PBB.
James Fielding Smith is at the Henry County Public Service Authority, P.O. Box 7, Collinsville, VA
24078. E-mail address: jfsmith@neocom.net
PBB is more vulnerable to threats from fraud, falsification, and misrepresentation than
were previous budgeting systems. Vulnerability grows from complexity, dependence
on performance data, year-to-year changes in strategic focus, and the great expansion
of the numbers of kinds of specialists involved in the process. It appears to grow out
of PBB's widely recognized benefits. Transparency has been put forward as the primary
protective device against threats to PBB; however, this has problems with both
definition and operationalization. The fix may be a number of coordinated measures,
most importantly the inculcation of sound ethical standards throughout an organization
using PBB.
Growing from a desire to improve service and the efficient utilization of resources,
performance-based budgeting (PBB), sometimes also called performance budgeting,
has begun to penetrate all levels of government. The following definition from the
State of Maine captures both the meaning and flavor of PBB:
Performance budgeting links spending to results. Unlike line-item budgeting, it does not budget
around categories of spending . . . nor does it use the previous year's budget as a starting point.
Performance budgeting allocates resources based on achieving specific measurable outcomes.
Rather than asking 'what will be needed in personnel and other costs to carry out an assigned
task?' The first question is 'what is the outcome to be achieved?' Thus performance budgeting
first asks not 'how many people do we expect to put through a job training program?' or 'how
many permits have to be issued to protect this river?' but rather, 'how many people do we intend
to place in jobs?' or 'how many miles of river will be fishable and swimmable?' The number of
people placed in jobs and the miles of river that are fishable and swimmable are outcomes and
they are the basis of performance budgets.... Outcomes are defined through a strategic planning
process that considers the critical issues facing an agency, the agency's capabilities, and input
from stakeholders such as citizens, customers, legislators, and employees.'
From a handful of mostly state programs prior to 1990, PBB has spread to nearly all
states,^ the federal govemment, and many local governments, special use districts, and
non-governmental organizations.^ PBB may soon be the de facto standard for public
budgeting. Many perceive PBB as clearly superior in an atmosphere of resource constraint
for optimizing both the effectiveness and the efficiency of programs. Down
slope agencies feel the cumulative effects of PBB's enthronement in federal rules
governing transfer payments, grants, delegations, and devolution of powers or programs.
Less obviously, yoking budgeting and strategic planning so directly in PBB has
nearly doubled the direct constituency that chooses or implements budgeting systems.
Accepting as a given that PBB is superior to previous budgeting systems, an assumption
that probably deserves serious examination elsewhere, this article addresses
the premise that PBB is more vulnerable to threats from fraud, falsification, and
misrepresentation to earlier budgeting systems. An examination of the sources and
types of this vulnerability will lead to a proposed approach to preventing fraud, falsification,
and misrepresentation. This article will review the potential threat mechanisms,
the magnitude of the possible threats, and possible countermeasures.
SOURCES OF VULNERABILITY
Each advantage or benefit of PBB seems to introduce at least one new risk or threat to
data integrity, as is shown in Table 1 on the following page. The sources of vulnerability
fall roughly into four categories: (1) same as other budgeting systems but heightened;
(2) introduced from strategic planning; (3) unique to PBB; and (4) temporary, or
learning curve, problems.
Same as Other Budgeting Systems but Heightened
Any threat operating in line-item, program, performance, or other budgeting system
will also threaten data integrity in PBB. Although the more rigorous degree of priority
setting in PBB will make fewer data points or streams appear vulnerable and needing
protection, PBB's explicitly defined linkages of fiscal data to performance data can
enable amplification and ramification of errors, thereby heightening their effects. Managers,
budgeters, and planners dealing with PBB should try to build on past experience
in evaluating vulnerabilities and agency responses.
Organizational structure or behavior may harbor threats to data integrity in any
budgeting system. All budgeting systems suffer from the results of a lack of incentives
for team implementation; for example, the assignment of non-overlapping, strictly
individual responsibility for each task and subtask. This remnant of the managementby-
objective culture can affect data integrity in PBB by distorting performance measures
to follow organizational units or chains of command more exactly than do the
organization's processes. Despite its potential for promoting teamwork, PBB remains
vulnerable to parochialism. Lee, Melkers, and Willoughby, in assessing state budget
practices, have noted changes to traditional budgetary relationships to respond to
PBB's special requirements.''
Temporal relationships affect the applicability of long-term or trend analysis to any
budgeting system for detecting aberrations in budgeting. However, as budgeting has
evolved from line-item to PBB, the focus has shifted on categories labeled with major
missions and away from a stable list of line-items. As the major goals evolve and
change from budget cycle to budget cycle, systemic trends may be obscured by the
more obvious changes in higher categories. Another environmental example would be
the succession of major emphasis programs in water quality since 1972: primary
treatment, then, following at roughly four to five year intervals, secondary treatment,
nutrient removal, toxicity, and currently, metals and animal waste. Each new focus
brings very different outcomes and outcome measures, even if the most basic process—
here, processing permits—carries on throughout. Budgets inevitably change
from year to year as programs and subprograms grow, reach maturity, and end. Temporal
changes can greatly complicate year-to-year comparisons and long-term analysis.
Introduced from Strategic Planning
Just as PBB shares categories of threats with earlier budgeting systems, it also shares
threats inherent to strategic planning. Oversimplification and temporal inconstancy, as
noted above, along with parochialism apply but become particularly important in the
key strategic planning functions of stating vision, articulating a mission statement, and
specifying goals and objectives. Any problem affecting or even shading the operationalization
of these higher concepts represents a particular threat to PBB. For twenty-six
years the ultimate goal of the Clean Water Act has been to make all the waters of the
United States "fishable" and "swimmable," but these terms have undergone periodic
redefinition to fit changing economic realities, technological advances, shifts in programmatic
emphasis, national policies, and relations between EPA and the states. Each
redefinition would have affected PBB, had PBB been in use all this time.
Unique to PBB
PBB's own special sources of threat come most directly from PBB's complexity.
Everyone advertises PBB as procedurally transparent and logically accessible, and
these characterizations do apply to the basic concept. However, building, maintaining,
and managing a PBB system involve daunting levels of complexity. Each little niche
of complexity can become an opportunity for error, fraud, falsification, or misrepresentation.
These key data processes can be disrupted or distorted in any segment of
PBB: strategic planning, defining performance measures, reporting performance data,
reporting or apportioning expense data, classifying outcomes, defining of success,
developing budgets, implementing the budget correction process, analyzing performance
data, analyzing cost data, and other areas.
Because of this complexity and the number of major skills—virtually separate specialized
careers—involved, no one person or type of specialist can master the whole
PBB system. Teamwork by necessity could result from this but, on the other hand, no
one manager can master and oversee all the contributory processes.
PBB, when it makes fiscal performance links, creates a problem where none existed
before. Separately, performance auditing and financial auditing have developed to
serve the needs of their targeted substrates; however, performance auditing is significantly
less well developed as a technique and as a field than financial auditing. This is
true even in the private corporate sector where, under the term "management accounting,"
performance auditing has long been practiced. Combining financial and performance
auditing for PBB can create mismatches and lacunae, allowing threats, which
were not threats to the separate branches, to function. Comprehensive auditing for
PBB will require many new techniques and approaches. For example, consider the
challenges of erecting confidence limits around correlations within a model.
PBB works better than other budgeting systems because it drives out the superfluous
and not required, matching funding with real priority projects. However, by undercutting
traditional politics to do this, PBB seems likely to accentuate politicization of
goals within the organization. Internal politicization may, in turn, link up with traditional
politics. Regardless, politicized competition for funding may engender, at least
among the more farsighted participants, a high degree of politicization of selection and
wording of perfonnance measures, both in the appropriation and authorization phases
and in the public accounting and electioneering phases. Whether politicization in this
case equates to hyperbole will depend on leadership and the culture of the organization.
Specific threats to data integrity. Data integrity faces challenges from unintentional
errors as well as from intentional ones, and sometimes the two types get confounded.
Error also operates both on concept, e.g., theories, models, linkages, cause-and-effect,
and on data. Errors in concepts generally create more dastardly results partly because
they propagate themselves like computer viruses every time they are used in a calculation
or analysis. Confounding of conceptual errors and misrepresentation can also
occur. A rascal, wanting to twist the outcomes of PBB, might yield to the temptation to
create a distortion in embedded concepts. Such misrepresentation, if discovered, might
possibly appear to be a conceptual error, an honest mistake, or the result of a legitimate
difference in interpretation. Data point manipulation to obtain the same level of misrepresentation
almost certainly would exhibit both difficulty of detection and ease of
explanation once detected. Fortunately, errors in concept can easily be detected by
rather simple countermeasures, as will be seen below.
Errors in operationalization of measures are to conceptual error as tactics are to
strategy. Both intentional and unintentional errors can occur here but have less widespread
consequence to PBB integrity than conceptual errors. Ways in which this may
occur include the selection of wrong measures, the use of erroneous modeling assumptions
for complex indicators (e.g., using principal component analysis), the omission of key aspects, the use of inconsistent or inadequate baselines, or the application of
incorrect transformations or inflators. A special case of operationalization error results
from costing errors, especially those involving the misassignment or incomplete distribution
of costs. Abstract aggregation of variables creates performance measures that
are hard to understand out of context. For example, a state environmental agency
might have a goal of a 60 percent return to compliance within one year, which summarizes
thousands of facilities covered by over a dozen permitting and regulatory programs,
not even accounting for the many difficulties inherent in defining a key term
like "compliance."
Some activities can directly inject error into PBB data, as with any other complex
dynamic model or data system. In sample shopping someone selectively forwards data
or selectively discards data. This may result in areal, temporal, or quality misrepresentation
of phenomena. Data fudging involves the direct manipulation of the data to
change its import or meaning. All data points remain, but some or all may undergo
illegitimate change. Both sample shopping and data fudgings how the overall effect of
changing the characteristics of groups of data, e.g., means, medians, measures of
dispersion.
Temporary, or Learning Curve, Problems
PBB seems to ignite a search for completeness and specificity, which may initially
result in over complicated definitions, models, and performance measures. This may
obfuscate essential relationships or, as simplification sets in with program maturity,
lead to discontinuities in baselines and data interpretation.
Newness itself may create error. Most often resulting from a lack of adequate
training or from the overselling of PBB's benefits, this effect can express itself as
misplaced effort, misunderstood directions, procrastination, or approximation, any of
which can cause a differentially higher data error rate in a new system.
Uncertainty of consequences. All organizations newly using PBB, and perhaps
some veteran users, experience uncertainty about the consequences of missing performance
targets. The full range of uncertain consequences must be considered: more or
less funding, dumbed-down targets, loss of confidence, restructuring, morale, reprisals,
or punishing headlines, plus a large range of opportunity costs. Through such uncertainty,
PBB creates another layer of rewards, incentives, and disincentives, possibly
motivating distortions in data. The uncertainty concerning consequences may express
itself, however, in uncertain directions, distorting data either upward or downward,
depending on expectations. Uncertainty and politicization may interact, potentiating
the effects. The extent of uncertainty's effects will depend on whether the sign of the
effects is known accurately throughout the organization, which can take place through
policy, training, documentation, or informal diffusion. The contrarian notion that uncertainty
promotes honesty seems to be unverifiable, at least in the context of PBB.
MISREPRESENTATION
Among all sources of error, the misrepresentation of results undoubtedly causes the
most significant compromises of integrity. Misrepresentation occurs when someone,
who has at least enough authority to make a point stick, claims that the results really
mean something other than what systematic analysis can support. No level of data or
analysis can escape vulnerability to misrepresentation, and at the highest level misrepresentation
gets fed directly to the media and to decision making bodies such as
legislatures. Misrepresentation can have profound costs for an organization. The costs
operate in many dimensions: loss of credibility, program compromise, and future
appropriations, for example. Despite such risks, misrepresentation occurs depressingly
often, and PBB may goad desperate managers into trying to make their organizations
appear more competitive, more effective, or more efficient than reality supports.
Unfortunately, misrepresentation breeds. Managers, technical staff, customer support
staff, and clerical staff throughout an organization can be infected by the willingness
to countenance misrepresentation. Lingering effects of management by objective
may precondition some to be vulnerable to misrepresentation.
Reengineering applications. A special source of vulnerability comes from the high
degree of compatibility, or perhaps suitability, of using PBB as a primary tool during
the reengineering of organizations. PBB's unique powers to drive and reinforce
changes in vision, mission, and organization also have consequences for data integrity.
The high stress environment surrounding reengineering may increase incentives for
misrepresentation or distorted definitions of performance measures, thereby accentuating
problems in PBB implementation. This special linkage between PBB and
reengineering verges on co-dependency and merits careful monitoring.
In summary, PBB's vulnerabilities result from complexity, dependence on performance
data, and year-to-year changes in strategic focus, exacerbated by the great
expansion of the budgeting constituency to include strategic planners and many more
levels of managers. This expanded corps must deal not only with new career
subspecialties but also with newly important interactions among budgeting, strategic
planning, and operations.
COUNTERMEASURES
PBB's complexity invites, or perhaps demands, automation. Not only does automation
facilitate the necessary data flows and analysis, but it can also include programming
for error detection, inconsistencies, and other forms of QA/QC. However, automation
adds to the problem. It can hide manipulations and assumptions. It can give a false
sense of objectivity, sometimes carelessly equated with accuracy. It can reduce
thoughtful analysis while simultaneously promoting it.
Awareness
Awareness encompasses a range of countermeasures. Its most elemental form entails
disseminating knowledge of threats to the entire user community or to key subcommunities.
Many tools can contribute to awareness campaigns, from posters to position
performance expectations, and from caveats written in budgets to comments in budget
briefings. Of the fourteen states listed by NASBO as being primarily PBB-using, ten
had sites on the Internet as of March 17, 1998.^ Examination of nine of these showed
seven to have a statement of the importance of awareness of threats to data integrity,
but none went beyond the statement. There appeared to be a tacit assumption that the
mechanisms in place for protecting fiscal and performance data would suffice for PBB.
Competition and Cooperation
The intense levels of competition implicit in the priority setting process of PBB should
have the ancillary effects of countering threats to integrity. In this competitive atmosphere,
each rival mission and program can expect scrutiny from competitors. The
overall process should squeeze padding and excess, reducing the number and capacity
of internal hiding places for error. As a result, opportunities for successful misrepresentation
should shrink even though some competitors may succumb to the temptation
of sweetening their models or data up front.
Cooperation and teamwork can yield a de facto countermeasure against threats to
PBB integrity. Like most tools discussed herein, they could also subvert the process,
depending on leadership and organizational culture. Such an effort would involve a
conspiracy, a possibility both remote and outside the scope of this paper. Cooperation
among specialists and the resultant teamwork ought to yield shared knowledge and
mutual respect, which will support data and procedural integrity.
Openness
Openness consists of many elements, each of which can contribute to protecting PBB
integrity. The more steps and procedures done in the open, the greater the effort it will
take to propagate fraud, falsification, and misrepresentation. Accessibility of the process,
especially of strategic planning and fiscal-performance linkages, reinforces integrity,
partly through coincidental inspection, citizen scrutiny, and feelings of ownership
in the widest possible range of stakeholders. Clarity means a combination of understandability
and logic. It requires clear statements of consequences, or lack thereof,
and of what outcomes are "good."
Comprehensive Auditing
Since PBB combines existing budgeting, strategic planning, and operational monitoring
techniques, the same auditing and QA/QC tools already used with those techniques
are ready for application to PBB. The problems arise with trying to keep up with a
moving target—after all, PBB is most highly touted for its ability to deal with an
organization undergoing change—and with coping with the linkages and interactions
among fiscal and performance variables. Notwithstanding preexisting tools for auditing
and the challenge of programmatic change, comprehensive auditing presents the
same fundamental Demingesque problem when applied to PBB as when applied, perhaps
in the guise of systematic or intensive inspections, to any process; inspections, no
matter how intense or skillful, can only find the errors but have severely limited
abilities to find the sources of errors. Similar effort invested in inspections will yield
less result in quality or accuracy than a commitment to a process of continuous improvement,
which, in turn, involves awareness, commitment, innovation, and imagination.
Comprehensive auditing will always have a place in PBB to satisfy legal audit
requirements on the fiscal end and, increasingly, on the performance end. When automated,
comprehensive auditing forms the basis of most QA/QC routines and can
estimate random error in the system. On the other hand, systematic error will persist
unless rooted out by something like a continuous improvement process.
Politicization
The accentuated politicization that characterizes the priority setting phase of PBB
offers a remarkably strong tool for detecting and sometimes correcting errors in the
PBB system and data. In this case, politics acts as a public watchdog. This tool
probably acts most strongly when PBB deals with severely constrained total resources
and under conditions of openness.
Politicization may come into play in one rare but powerful way of limiting some
threats to PBB integrity: the intelligent exemption of certain types of programs from
PBB. Despite PBB having been sold as a universally applicable system, some programs
should go into sanctuaries so that the pressures of competition for resources and
performance evaluations do not have any opportunity to distort data or program management.
This draconian solution may have few legitimate applications. Enforcement
programs come to mind as a legitimate candidate because of their fragility, sensitivity,
and lack of control over workloads and the timing of outcomes. Timing's particular
importance arises out of cost-efficiency being the underlying principle of PBB and
time being a critical factor in computing any measure of cost efficiency on both the
cost and performance sides.
TRANSPARENCY
Qne last major category of corrective action remains: transparency. Transparency,
often put forward as the primary protective device, or even panacea, against threats to
PBB, has problems of both definition and operationalization. Transparency has attracted
many shadings of meaning ranging from "obvious" to "entirely in the back-
ground." Peter Droege offers this definition, "Transparency should denote in an ideal
situation: (1) public accessibility of data, (2) the actively managed interpretation of
process and structure, and (3) public announcements as part of an accountability program,
with facilitated discussions of assumptions, process, and implications of outcomes
or decisions."^ Don Krueckeberg says it "involves procedures which are easy to
follow, guided by rules that seem necessary and make sense, no hidden consequences."'
The present author proposes transparency to be a process in which the steps
are visibly or clearly linked to the mission or vision of the organization. Transparency
is a portmanteau that can be deconstructed into many of the concepts described earlier
in this article. Transparency is a desirable and maybe necessary condition, but it is not
a sufficient condition for data integrity.
PUBLIC AND PRIVATE SECTORS
A pair of critically important questions arise: "Why do organizations in the corporate
sector not encounter the problems identified in this article?" and "But if they do, why
don't they care?"* The same budgeting practices—both systems and corrective measures—
have been used widely in the private sector, where the participants cannot be
assumed to be more skillful or more ethical, without encountering the problems identified
in this article. A survey of articles published in Management Accounting between
October 1994 and December 1997 revealed only one article specifically on data integrity,
ethics, or fraud.^ Such findings in the most topical journal in the private sector
may indicate a very high degree of confidence in the integrity of data in PBB in the
private sector. One theory to explain this might be that private organizations assume
that competitive pressure ruthlessly seeks greater efficiency, which drives out major
inefficiencies such as data threats create.
Does this theory apply in the public sector? Consider this argument, which applies
only to the public sectors in democracies. Efficiency and democracy are fundamentally
incompatible, as democracy is never the most efficient way to run anything and its due
process safeguards inherently add inefficiencies to all other public processes. Public
programs are the tools of democracy. They may be forced to try to become more
efficient and otherwise to copy the private sector model, but they are still tied to
democracy. Bureaucracy is where democracy tries to reconcile this conflict. PBB is
bureaucracy's bravest try to date to approach the private ideal, responsibility budgeting.
But due process and other safeguards interfere with pure competition, limiting
how close this approach gets to the ideal. PBB in the public sector therefore needs
countermeasures applied to avoid or control the threats to its integrity.
Klitgaard presents another viewpoint, one which deals with international cooperation
against corruption, necessarily combining private and public issues.'^ He looks at
a case in which data integrity, misrepresentation, fraud, and other issues occur in the
hothouse of monopolistic systems under pressure from bribery or other inducements to
cheat. The democracy-competition theory, although not explicit, undergirds the con-
cept of corruption equals monopoly plus discretion minus accountability, which is
Klitgaard's central thesis.'^ Concluding that corruption is a systemic problem,
Klitgaard proposes a systemic strategy, a strategy that could be characterized as ethical
education.
ETHICS AND THE CULTURE OF PBB
The fix may be a number of coordinated measures, most importantly the inculcation of
sound ethical standards throughout an organization using PBB. The basic problem is
cultural. Although PBB can create another layer of rewards, incentives, and disincentives,
it still operates within the organization's culture. Melkers and Willoughby found
only seven out of the fifty states have formal incentives and disincentives in their
budgeting systems.'^
A comprehensive program based on ethics is needed. All transactions within PBB
should pass the test: is there any hidden incentive for error or disincentive for ethical
behavior? Internal recognition and rewards programs bear examination in this light, as
do personnel policies, especially hiring, promotion, and discipline. Emphasis should
openly go to corrective measures taken to eliminate sources of error. Essentially all
training programs should include units or goals supporting ethical behavior.
The interplay between individual gain and organizational advancement is usually
assumed to be synergistically positive. This may not be true. Consider any case in
which organizational growth and survival will depend on favorable performance measures:
this might cause organizational momentum—an effect pervasive throughout the
organization—to swamp any effects of individual leaders. This is a strong argument
for the inculcation of ethics throughout the levels of an organization. It works both
ways, to restrain a rogue leader or to restrain a rogue organization. Ethics also need to
be fostered among and between agencies, as any one agency will rarely encompass all
of a policy area. Trans-organizational ethics may present the greatest challenge of all.
Many organizations still labor under the burdens of systematized distrust, which is
the legacy of MBO and theory X management. This legacy seriously impedes culture
change, as would be needed to mobilize ethics to protect the integrity of PBB. On the
other hand, its effects on morale and unit identity are often recognized within an
organization and could be used to energize cultural change. Mindless cutthroat competition,
as fostered by MBO, is a serious threat to PBB integrity.
Klitgaard's anti-corruption strategy combines punishing prominent offenders ("fry a
few big fish"), involving "the people" in diagnosing corrupt systems, focusing on
prevention by repairing systems, creating reform incentives, showing cultural units
they are not alone in these problems, mobilizing creative competition, and providing
"tool kits" filled with best practices for fighting corruption.'^ Klitgaard and this article
agree that disseminating awareness and analytical capabilities widely within organizations
or cultures gives the best protection against such things as fraud, falsification,
and misrepresentation. To the extent that transparency promotes these, transparency
contributes to the protection of PBB. Transparency alone, however, does not suffice.
PROSPECTS
PBB's perceived strengths and advantages come largely from its superior facility for
tracking environmental and programmatic change. It forces the size of programs to
what is appropriate now, or in the next budget cycle, and it imposes life cycles on
programs, ending programs that have finished their purpose or outlived it and bringing
on programs to fulfill new mandates or needs. PBB has a strong analog in the phenomena
of ecological succession, where resource constraints and changes create a more-orless
predictable sequence of biological populations or communities.
Despite possible problems, PBB with its overwhelming advantages in terms of
productivity remains the best approach around. In addition, PBB has palpably greater
philosophical soundness than other budgeting systems. It expresses the philosophy "if
it's not paid for, it's not real." And if it's not real, it shouldn't be paid for.
Is PBB more susceptible to fraud, falsification, and misrepresentation compared too
previous other budgeting systems? Yes. Can it be protected? Yes. A number of corrective
and preventive mechanisms, including the systematic promotion of ethics throughout
the organization, can protect PBB.
