Hitachi XBRL » Blog Archive » The Upshot of Japan’s XBRL Voluntary Program

Contributors

Bob Schneider is a Partner in Nihon Equity Research.com, which offers customized research of small- and mid-cap Japanese stocks to institutional investors. He worked in the equity research department of Morgan Stanley for six years, including three years as Supervisory Analyst and Senior Editor of their Tokyo office. More

Wilson So is the Director of Hitachi Consulting Corporation XBRL Business Unit, introducing Hitachi’s XBRL products and consulting services to both the commercial and government sectors and actively contributing to XBRL thought leadership. More

The Upshot of Japan’s XBRL Voluntary Program — and Why So Many Companies Signed On

Written by Toshinori Kobayashi Posted on June 6, 2008

Toshinori Kobayashi is director for enforcement of corporate disclosure at JFSA, the Financial Services Agency of Japan. He was responsible for the XBRL renovation project of EDINET, the JFSA’s electronic disclosure system.

In my previous post, I offered a brief overview of the process of introducing XBRL to Japan’s EDINET system. With many countries and related agencies now looking to add XBRL to their own reporting regimes, there’s been heightened interest in our second pilot program that was successfully conducted in the summer of 2007. Even though it’s been about a year since the program was instituted, we are still receiving many inquiries. I’d therefore like to address some of the key points about the program.

While the first pilot in January/February 2007 was an offline program for some 50 companies (chosen by Japan’s Financial Services Agency, or FSA) to check the taxonomy, the second pilot effort was a large-scale online program in which 1,233 companies voluntarily took part. For us on the EDINET team, the second program thus became a major event that propelled the XBRL effort forward. Its implications were threefold:

(1) Many companies showed a high level of interest in the adoption of XBRL to the EDINET system, with one-quarter of all listed companies voluntarily participating.

(2) The results gave us much confidence about the successful completion of the project. No major problems were detected in the EDINET system and the taxonomy themselves. Moreover, the taxonomy was found to cover over 95% of the tags needed for filings of the participating companies.

(3) Issues to be faced in the remaining half-year of the project became clear. Along with the positive results noted above, we did find some problems that needed to be addressed.

Let me focus on Point (3). As mentioned above, 1,233 companies participated. Among them, 962 companies (about 80%) were able to complete electronic transmission of their financial reports in the XBRL format. Up to now, when Japanese companies needed their financial reports prepared, over 90% contracted with specialty printers for support. In the second pilot program, 78% of participants received support from financial printers. While 97% who used financial printers were successful in electronically filing their statements, 82% of companies who didn’t receive help were unsuccessful.

Thus, we learned that one essential for ensuring the smooth introduction of XBRL-mandated statements in the current fiscal year was to have printing firms with the skills to handle XBRL filings available on a wide-scale basis for companies to use.

Regarding the companies that were unsuccessful, we have to recognize that this was their first attempt at such filings. Nevertheless, it became clear that filing XBRL reports is a significant challenge for companies without sufficient skills, which made us recognize that it would be necessary to adopt several measures quickly (as I will discuss later).

A survey conducted by the FSA of the firms that failed revealed two major reasons for their lack of success, each of which was reported by more than 30% of the companies. One was difficulty in using the software tools and insufficient information about obtaining those tools (the specific answers were “We didn’t understand the methods for producing XBRL data” or “We didn’t have tools or services for creating XBRL data”). The other was time constraints (i.e., “Because our people were busy with other work, we weren’t able to devote sufficient personnel and hours to preparing the statements” or “It was taking more hours than we had allotted for the task”).

As expected, when all companies in the pilot program were asked about the issues they faced in preparing statements, they mentioned the problems of tools (i.e., “the cost of tools and services”, or “difficulty in using tools”). Others cited “difficulty in determining the account title (since many choices were possible),” or “the large amount of preparation materials and the big manual” and “the manual was hard to understand.” Interestingly, some companies mentioned the difficulty in coming to an agreement with the auditor on the selection of account titles.

Companies which successfully completed the filing were also asked about the estimated amount of work involved (i.e., number of hours X number of people) in creating the statements. For the first year of preparing the statements, 22% of respondents said the amount of work would be twice that for the HTML statements used up to now, and 50% said it would be 1.5 times to 2 times. Thus nearly three-quarters of successful firms said they estimated that in the first year the amount of work involved would be 1.5 times or more than previously. On the other hand, some 14% said the amount of work would be unchanged; no companies said there would be a decrease.

Looking toward the second and subsequent years, however, only 6% of companies estimated there would be twice as much work as there was in preparing HTML statements, and the proportion who said 1.5 times to twice as much effort fell to 28%. Those anticipating the amount would remain the same was 50%, while 4% said it would fall.

Finally, it is worth noting that, if we analyze the data submitted, 30% made some sort of error in FRTA rules, and 30% made an error in FRIS rules.

The problems from the pilot program listed above had been generally anticipated. However, rather than EDINET system and technology issues, the question that was most urgent was how we at the FSA could support the reporting companies so that the transition to XBRL would go quickly and smoothly. Given this imperative, in the second half of the fiscal year, we took the following steps — concomitant with the development of the system and taxonomy — to ensure smooth operation of the new EDINET system:

(1) Improvement and Completion of Guidelines
We had been providing three types of guidelines for users: (1) creation of a taxonomy tailored for each company; (2) creation of instance documents; (3) selection of the right account item. We made these guidelines easier to understand based on the feedback from users in the pilot program.

(2) Extensive Meetings on XBRL
We had already conducted meetings on XBRL with companies in the past, but we arranged 30 more in January-February 2008 to help reporting entities gain a deeper understanding of interactive data. In total, 4,000 companies participated.

(3) Outreach to Software Vendors
We exchanged information closely with software vendors and pushed them to develop tools that were easier to use. Specifically, we provided ideas about tools handling FRTA/FRIS violations in a better way (in order to reduce the number of violations).

There are various challenges to overcome to introduce XBRL successfully to a system such as EDINET. While we tend to devote much of our energy to technical issues, such as taxonomy creation, a smooth XBRL introduction is realistic only when the business world is completely prepared. We also need to devote our efforts to the needs of actual users. It was this point that gave the pilot program particular significance.

Let me now focus on the very high level of participation in the second pilot program, which I noted exceeded 1,200 companies. The high number has surprised observers outside Japan, and to be honest, it surprised me, too. Initially, we thought at best 400 to 500 companies would participate. Even that level seemed very high, given the few dozen in the SEC’s voluntary program (VFP). Participation in the Japanese program was completely voluntary. There were no restrictions, either direct or indirect, placed on those who didn’t want to participate. At the same time, there were no limitations placed on those who wanted to participate.

Why, then, did over 1,200 companies voluntarily choose to join? The answer to this question – on which we have received many inquiries from overseas agencies – appears to be embodied in the following four points:

(1) The Prospect of an XBRL Mandate in April 2008
Companies pondering whether or not to join the pilot program knew well that XBRL would be adopted for financial reporting in April 2008. For many firms, this expectation gave them the sense that participation in the pilot program was a necessity. The appeal of the pilot program to reporting entities was that it would be an online test of the system and taxonomy, which would give them practice and experience in XBRL filings.

(2) Aggressive Outreach by the FSA and Others
The FSA aggressively sought to recruit companies to the program, conducting ten meetings for the purpose. In addition, we also relied on the Keidanren, the Tokyo Stock Exchange, the Japanese Institute of Certified Public Accountants, the Investment Trusts Association, and other associations to spread the word. The printing firms noted earlier also were important in informing their client companies about the program.

(3) Enlisting Software Frims and Printing Companies for Support
Detailed information on the pilot program was provided to software vendors and printing companies early on. The former were able to provide cost-free tools for pilot program use and the latter were able to offer support services, thereby reducing the burden to participating companies.

(4) Filed XBRL Data Was Not Made Public
Most of the filed data was not made public (with the exception of that for ten companies that agreed to release it for reference purposes). In addition, since it was a pilot program, we were lenient in accepting data that was not of the highest quality. Thus, companies didn’t need to be concerned that their reputations might suffer from the publication of XBRL data of insufficient quality, which stemmed from a lack of experience with interactive data.

Needless to say, the environment and terms under which any country or agency introduces and mandates XBRL can vary greatly from those in Japan. Nevertheless, I hope that some of the details and implied suggestions I’ve related about the Japanese experience will be useful to those who are seeking to introduce XBRL in their specific jurisdiction.

Add to Del.icio.us | Digg this

This entry was posted on Friday, June 6th, 2008 at 6:59 pm and is filed under General, . You can follow any responses to this entry through the RSS 2.0 feed. Both comments and pings are currently closed.

Comments are closed.

Contributors

The Upshot of Japan’s XBRL Voluntary Program — and Why So Many Companies Signed On

Subscribe

Search Posts

Recent Posts

Archives

Categories

Links of Note