Given the current economic situation, cities and counties are looking for technologies to reduce administrative costs and improve the efficiency of their operations. One technology tool is purchasing cards (p-cards), which many cities and counties are using to streamline operations and reduce costs associated with the procurement function of government. Since p-cards arrived on the market in the early 1990s, many governmental entities have reported significant benefit from their use, such as reducing or eliminating the paperwork associated with requisitions, purchase orders and invoices for thousands of small-dollar transactions. The General Services Administration(GSA), for example, estimates that federal government use of p-cards generates cost savings ranging from $54 to $92 per transaction (U.S. GAO 1996) and reduces the time required to process paperwork transactions by two to six weeks (U.S. GSA 2006). More recently, market surveys have indicated that use of p-cards yields savings of about $70 per transaction (Palmer, et al. 2007). [Reference sources are listed at the end of the article.]

The ability of p-cards to reduce administrative costs reflects a re-thinking of the value of various processes and controls over low-value purchases. Analyses have indicated that the cost of managing low-value purchases in the traditional paper-based, purchase order-driven format is disproportionately high in relation to the cost of the goods purchased. In other words, organizations are finding that the cost to process a low-value transaction often exceeds the cost of the good or service itself. Furthermore, if left unchecked, the blizzard of low-value purchase paperwork can lead to lost discounts, late fees, excessively long wait times for goods and poor supplier relations due to concern over payment.

This article will present survey research findings to identify current and future trends and the economic impact of the use of p-cards in city and county government.

Benchmark norms

Exhibit I shows the organizational and p-card spending norms for both small (less than 1,000 employees) and large (1,000 or more employees) city/county respondents to the 2009 survey. [Survey methodology described on page 56.] Exhibit I indicates that average (median) monthly p-card spending at large cities/counties is $545,023 ($350,484) compared to $151,383 ($100,000) at small counterparts. However, because large cities/counties have significantly more employees, monthly p-card spending per employee is higher at small cities/counties ($318) than at their larger counterparts ($145). In part, the higher spending per employee at small cities/counties reflects more liberal distribution of cards to employees. And, while p-card spending as a percentage of municipal budget averages 1.4 percent across all respondents, small city/county p-card spending averages 1.76 percent of the municipal budget while large city/county spending averages 0.94 percent of the municipal budget.1

Large and small cities and counties are modestly different in terms of average monthly spending per card ($1,016 at small and $1,324 at large cities/counties), driven by a higher average transaction amount at large cities/counties ($297, versus $242 at small cities/counties). Monthly transactions per card, however, are similar across the two groups (4.21 at small and 4.45 at large cities/counties). On average, about 43 percent of transactions under $2,500 at small cities/counties, and about 50 percent at large cities/counties, are paid by p-card. Among transactions of $2,500 or greater but less than $10,000, small cities pay for about 25 percent of their purchases with p-cards, compared to 19 percent at large cities.

Past and future trends

Notwithstanding a severely challenging economic situation in North America, 65 percent of small cities and counties (and 72 percent of large cities/counties) reported an increase in p-card spending in the 2007-2009 time frame. Across all city/county respondents, the average p-card spending growth rate over the two-year period from 2007 to 2009 was 22.9 percent.

Going forward, survey responses indicate that 72 percent of small cities and counties, and 75 percent of large cities/counties, expect higher p-card spending by 2012. However, while most cities/counties expect to increase p-card spending over the next three years, the growth rate will be significantly lower than in the 2007-2009 time frame. Across all cities and counties, p-card spending is expected to increase by only 3 percent in 2010 and, on average, about 7 percent per year through 2012.

Exhibit I. Small and Large City and County Purchasing Card Program Performance Statistics, 2009* (All Numbers are Averages Except Where Indicated Otherwise)
Small Cities and Counties, 2009 Large Cities and Counties, 2009
Number of employees 476 3,763
Program Performance Measures
Number of purchasing cards 149 412
Card-to-employee ratio 31.3% 10.9%
Average monthly p-card spending $151,383 $545,023
Median monthly p-card spending $100,000 $350,484
Transactions under $2,500 placed on p-card 43% 50%
Transactions between $2,500 and $10,000 placed on p-card 25% 19%
Monthly p-card spending per employee $318 $145
Annual p-card spending as a percent of annual city/county budget 1.76% 0.94%
Cardholder Activity Measures
Monthly transactions per card 4.21 4.45
Spending per transaction $242 $297
Monthly spending per card $1,016 $1,324
Active cards in a typical month 80% 76%
* Small cities are those with less than 1,000 employees; large cities are defined as those with more than 1,000 employees)

Impact on efficiency and service

The most widely recognized benefit of p-card use is cost savings per transaction. Exhibit II shows that the average administrative cost of procuring and paying for a good or service via the traditional purchase-order process is reported by city and county respondents to be about $111 per transaction. In contrast, the average cost associated with a p-card transaction is estimated to be $29 — a net savings attributable to p-card use of about $82 per transaction (a 74 percent cost reduction). To put the transactional cost savings into the larger context, consider that in 2009 an average large city/county organization spent $6.54 million on 22,000 p-card transactions.2 Considering the respondent-reported savings of more than $82 per transaction, one can conclude that p-cards generated overall transaction cost savings of about $1.8 million for the large city/county, which equates to about a 28 percent price reduction. At the macro-economic level, if all U.S. cities and counties paid for 1.25 percent of their budget with p-cards (the average for all cities/counties), then about $6 billion in transaction cost savings would accrue to the benefit of local taxpayers.3

For most cities and counties, another key benefit of the p-card is a reduction of 15 days (from 19.4 to 4.4 days) in the procurement cycle time (the time elapsed from the placement of an order to the date goods are received).

P-card data also plays an important role in an organization's ability to obtain discounts. Almost 18 percent of city/county respondents report that their organization uses p-card spending data to obtain a higher discount for goods or services from a vendor.

Best practices for city/county agencies

A principal goal of this survey is to gain insight into the drivers that push transactions to the p-card in order to generate cost savings and improvements in efficiency for the card-using organization. To better understand these drivers, we identified "best practice" p-card programs as those that have reported at least one top quartile (and no bottom quartile) metric across four key p-card program performance measures, including:

  • Percentage of under $2,500 transactions paid by p-card,
  • Percentage of $2,500 to $10,000 transactions paid by p-card,
  • P-card spending as a percent of annual budget, and
  • P-card spending per employee.

A second "needs improvement" group is also assembled from city and county respondents. This group is the reverse image of best practices, specifically p-card programs in which at least one of the four performance metrics is found to be in the bottom quartile and none in the top quartile. The activities of this group will be used to define underperforming city and county p-card programs.

Exhibit III reveals that, despite being similar in terms of headcount and age of p-card program, best-practice p-card programs report:

  • Nearly four times the average monthly p-card spending as the needs-improvement group ($524,578 versus $149,350),
  • Annual p-card spending as a percent of city/county budget that is five times higher than the needs-improvement group (2.85 percent versus 0.57 percent),
  • (On average) nearly twice as many p-card accounts (351 versus 195),
  • A significantly higher p-card-to-employee ratio (23.4 percent versus 11.3 percent)4, and
  • Average monthly cardholder spending that is about twice that of the needs-improvement group ($1,495 versus $766), driven by a higher average transaction amount ($335 versus $216) and more active monthly utilization of the p-card (4.46 versus 3.55 transactions).4

Exhibit III. Organizational and Purchasing Card Program Differences between the Best Practice and Needs Improvement Groups (All Numbers are Averages Except Where Indicated Otherwise)
Best Practice Group Needs Improvement Group Percent Difference
Number of employees 1,498 1,729 15%
Age of program (years) 8.00 7.44 -7%
Program Performance Measures
Number of plastic purchasing cards 351 195 -44%
Purchasing card-to-employee ratio 23.4% 11.3% -52%
Average monthly p-card spending $524,578 $149,350 -72%
Median monthly p-card spending $370,552 $89,000 -76%
Transactions under $2,500 placed on p-card 75% 26% -65%
Transaction between $2,500 and $10,000 placed on p-card 48% 3% -94%
Monthly p-card spending per employee $350 $86 -75%
Annual p-card spending as a percent of city/county budget 2.85% 0.57% -80%
Cardholder Activity Measures
Monthly transactions per card 4.46 3.55 -20%
Spending per transaction $335 $216 -36%
Monthly spending per card $1,495 $766 -49%
Active cards in a typical month 81% 74% -9%

Other recommendations

In addition to card distribution policies and internal guidelines that may limit card use, other decisions and policies within a local government differentiate best-practice (BP) from needs-improvement (NI) programs. By engaging in specific activities, BP local government entities get greater value from their p-card programs. Let's look at those activities.

Increasing spending limits. BP city/county p-card programs provide cards to employees with notably higher per transaction and monthly spending limits than NI organizations. The average per transaction spending limit at a BP city/county is $2,880, but only $1,382 at NI counterparts. Similarly, the average monthly spending limit at BP city/county is $7,875, which compares to $3,923 at NI counterparts.

Purchasing a wider variety of goods and services. Employees in BP city/county p-card programs use their p-card to pay for a wider variety of goods and services and, within every spending category, purchase a relatively higher amount of the good or service. This implies that NI cities/counties are unnecessarily limiting the types of goods and services that can be bought with p-cards.

Use of innovative p-card accounts. BP city/county p-card programs are more likely to use a variety of p-card options to meet their spending needs, such as "ghost accounts" and "electronic accounts payable" (EAP) accounts, and are able to use them in a more flexible manner.5 Survey responses indicate that 23 percent of NI but 32 percent of BP organizations use ghost accounts; further, only 3 percent of NI but 21 percent of BP program use EAP accounts.

Management support and communications. BP city/county p-card programs are more likely to indicate that a strong "business case" had been made to employees about the benefits of p-card use and that a top management sponsor gives vocal support to the p-card program.

Assessing potential and optimizing p-card program. BP city/county p-card programs are more likely to take steps to assess organizational potential for card spending, including to:

  • Analyze AP check payments to identify merchants who are to be encouraged to accept p-cards for payment (53 percent versus 18 percent ), and
  • Review purchase requisition traffic to identify employees who need p-cards (58 percent versus 20 percent).

Strategies to optimize p-card spending. BP city/county p-card programs are more likely to craft strategies to optimize card spending including to:

  • Target specific commodities or services for p-card payment (79 percent versus 38 percent),
  • Target specific vendors for p-card payment (60 percent versus 39 percent),
  • Require purchasing to refuse to process requisitions for purchases that can be paid by p-card (36 percent versus 18 percent).

Training and communications policies. BP city/county p-card programs are more likely than NI counterparts to:

  • Provide self-study p-card training materials (70 percent versus 43 percent),
  • Support p-card program administrator attendance at p-card user conferences to identify new ways to use p-cards (67 percent versus 37 percent), and
  • Have an ongoing method of communicating p-card information (80 percent versus 46 percent).

Fraud and misuse

The occasional misuse of p-cards continues to be fodder for news reports. In reality, the loss due to p-card misuse and policy violation is on average 0.006 percent of total p-card spending.6 For an average large city/county organization that spends $6.54 million annually, the average annual dollar loss is $392 (0.006% X $6.54 million). In contrast, as discussed earlier, the administrative cost savings of an average large city/county organization attributed to p-cards is $1.8 million.


Survey responses indicate that, although the economy has been in a downturn over recent years, city/county p-card spending has increased by about 23 percent in the two-year period between 2007 and 2009. Going forward, most cities and counties expect higher p-card spending, although the growth rate will be less impressive. Across all cities and counties, p-card spending is expected to increase by only 3 percent in 2010 and, on average, about 7 percent per year through 2012.

Survey responses also indicate that the average city/county transaction cost savings from procuring and paying for a good or service with p-cards is $82. In addition, cities and counties report that use of p-cards reduces the average procurement cycle time by 15 days, a savings that will reduce the need to incur the costs associated with high levels of inventory. Importantly, there is no evidence in the survey response indicating that the benefits of p-card use are negated by excessive policy violations, fraud or employee misrepresentation of p-card spending. Indeed, the financial impact of policy violations, fraud, and employee misrepresentation is insignificant in relation to p-card spending, and the vast majority of respondents believe that, in comparison to other payment methods, p-card spending is associated with a similar or lower likelihood of unwelcome incidents.

Survey methodology

In October of 2009, a 40-page web-based "2009 Purchasing Card Benchmark Survey" was released to 6,774 purchasing card program administrators at organizations that were either customers of one of 19 major card issuers or members of the National Association of Purchasing Card Professionals or the National Institute of Government Purchasing. A total of 1,915 responses were received for a response rate of 28.3 percent. Occasionally, one or more respondent may have given an incomplete response resulting in a different number of responses for different questions. Analysis of any given question is based on usable responses to each question. In addition, unusual "outlier" responses to specific questionnaire items when appropriate have been purged to facilitate a meaningful understanding of the data. All major purchasing card issuing brands are represented in the survey response.

A breakdown of survey respondents by organizational type is as follows: 24 percent are public corporations, 23 percent are private corporations, 16 percent (or 306 respondents) are city or county governmental units, 11 percent are universities, 6 percent are federal and state governmental agencies, 11 percent are school districts, and 9 percent are not-for-profit organizations.

Related Stories

About the authors

Mahendra Gupta is Virgil Professor of Accounting and Management at Washington University, St. Louis, Mo.; Miljan Markus is senior research associate for RPMG Research Corp.; and Richard J. Palmer is chair, Department of Accounting and MIS, Southeast Missouri State University, Cape Girardeau, Mo.


  1. By contrast, Federal Government p-card spending comprised 0.57 percent of its budget in FY 2009.

  2. Annual p-card spending at large city/county organizations is derived by multiplying the average monthly spending of $545,023 (Exhibit I) by 12. The number of annual p-card transactions is calculated by dividing annual p-card spending by the average amount (of $297 from Exhibit I).

  3. This figure is derived by multiplying annual spending by U.S. Cities and Counties (of $1.59 trillion as per by the average percentage of city/county budget spent on the p-card (1.25 percent). This product was then divided by the average transaction amount of small and large cities/counties ($270) to estimate the potential number of U.S. p-card transactions. This number is multiplied by the $82 per transaction cost savings reported by survey respondents.

  4. The difference in p-card distribution is, in part, a phenomenon associated with the internal rules of card distribution within the city/county. While only 3 percent of best-practice cities/counties restrict card use to one person per department, 23 percent of their needs-improvement counterparts impose this restriction on their program.

  5. Ghost card accounts are defined in the survey as "a p-card account number held in trust by a vendor who charges to the account at the buyer's instruction" (such as a travel management firm). Electronic account payable describes "non-plastic card accounts used to pay for invoiced goods and services (whether set up as a rotating pool of card accounts, ghost card accounts funded only to pay invoices from suppliers, or another arrangement similar in purpose)."

  6. Survey responses indicated that, in comparison to other payment methods, 96 percent of city/county respondents believe that p-card spending is associated with a similar or lower likelihood of policy violation. Further, 81 percent and 92 percent think p-card spending to be associated with a similar or lower likelihood of fraudulent or misrepresented spending, respectively.


Gupta, Mahendra, Richard Palmer and Christoph Sprenger, "The Progress of P-cards in North America: Current Data and an Historical Perspective," Cost Management, Volume 21 (No. 1), January/February (2007), pp. 22-28.

U.S. Gen. Accounting Office. (1996). Acquisition Reform: Purchase Card Use Cuts Procurement Costs, Improves Efficiency. GAO-96-138.

U.S. General Services Administration. 2006. How did it all get started?, (last visited Oct. 1, 2006).