The Information and Communication Technology Survey (ICTS) is part of a comprehensive program designed to provide more detailed and timely information on capital investment by nonfarm enterprises. Results improve the quality of current economic indicators of business investments, as well as the quarterly estimates of gross domestic product. The data also provide facts about trends in capital expenditures useful for identifying business opportunities, product development, and business planning.

Due to budgetary constraints in 2013, the Census Bureau suspended the 2012 ICTS. Funds were appropriated for fiscal year 2014 allowing the Census Bureau to reinstate the ICTS for the 2013 survey year. However, to operate within funding levels provided for FY 2015 and FY 2016, the Census Bureau decided to suspend the 2014 and 2015 ICTS. The 2016 ICTS was not reinstated for FY2017 and the Census Bureau decided to permanently suspend the ICTS.

ICTS is a companion to the Annual Capital Expenditures Survey (ACES). ACES was first funded in 1991. ICTS was created later in response to economic data user and policymaker concerns about the lack of available data on e-business infrastructure investment by nonfarm businesses. Rapid changes and advances in information and communication technology (ICT) equipment have resulted in these assets having short useful lives and a tendency to be replaced much quicker than other types of equipment. Enterprises are expensing the full cost of such assets during the current annual period rather than capitalizing the value of such assets and expensing the cost over two or more years. In some cases this is due to the short useful life of the asset, and in other cases this is because enterprises have varying dollar levels for capitalization.

The ICTS data are critical for providing improved source data to the investment component of gross domestic product, capital stock estimates, and capital flow tables. The data will also be used to assess future productivity and economic growth prospects. In addition, the data provide facts about trends in such expenditures useful for identifying business opportunities, product development, and business planning.

The 2013 estimates in this report are based on data collected from a sample of 46,642 enterprises with employees (of which 17,880 selected with certainty along with 28,762 others). The sample frame for enterprises with employees was about 5.7 million enterprises with employees (of which 5.5 million are single establishment enterprises).

The industry categories used in the 2013 ICTS were comprised primarily of 3-digit and selected 4-digit industries from the 2007 North American Industry Classification System (NAICS). Industry combinations were developed through consultation with data users.

Three survey forms (ICT-1(S), ICT-1(M), and ICT-1(L)) were used for the 2013 ICTS. Each enterprise in the sample was sent one of the forms depending on the diversification of their operations. Recipients of these survey forms were asked to provide industry-level data for capitalized and noncapitalized purchases, and operating leases and rental payments for three types of ICT equipment (computers and peripheral equipment; ICT equipment, excluding computers and peripherals; and, electromedical and electrotherapeutic apparatus). In addition, enterprises were asked to provide industry-level data for capitalized and noncapitalized purchases and payroll for developing software, and noncapitalized software licensing and service/maintenance agreements. Capital expenditures data was requested on the 2013 ICT survey to maintain comparability for those types of equipment collected. Additional detail regarding the ICTS forms and instructions refer to the Survey Forms and Instruction section.

In accordance with federal law governing census reports (Title 13 of the United States Code), no data are published that would disclose the operations of an individual establishment or enterprise. Disclosure limitation is the process for protecting the confidentiality of data. A disclosure would occur if someone could use published statistical information to infer the identity or operations of a business that has provided information under a pledge of confidentiality. Disclosure suppression protects the confidentiality of individual businesses’ information by withholding (suppressing) the cell values in tables of aggregate data for cases where only a few businesses are represented or dominate the statistic presented.

The disclosure analysis for the ICTS statistics is performed on each data item. When the estimate for a specific data item cannot be shown without disclosing information for individual enterprises, the publication of that data item is suppressed. The process of suppression does not change the higher level aggregate totals, so the integrity of the data is not adversely affected.

The following abbreviations and symbols are applied in this publication:

(D) The data is withheld (suppressed) to avoid disclosing data for individual enterprises. The data may be included at higher level totals where the individual enterprise’s data is not disclosed due to aggregation.
(NA) The data is Not Applicable.
(NS) The difference in the two estimates being compared not statistically significant at the Census Bureau standard of 90% confidence. This is also used for the estimate of percent change, where a NS notation means the estimate of percent change is not statistically different from 0%.
(Z) The estimate is greater than 0, but still rounds to zero in the published units. The estimate is less than half of the published unit of the table
(X) The estimate of the quality, using Relative Standard Error, is not able to be calculated reliably. This is due to how sampling variance is estimated when most, or all, of the sampled enterprises that had a chance to not be sampled did not report positive values for that item.

Note that an estimate of 0 means that there are no contributors to the estimate, other than those reporting a value of 0. The corresponding RSE is shown as a dash (-). An RSE of 0.0 means that all of the contributors to the corresponding cell estimate were selected with certainty, and the estimate has no sampling variability, though may still contain nonsampling error.