spending analysis purposes. For 1983 and subsequent tax years, the base period includes 3 years plus the incremental spending analysis for the tax year at issue. Further, the difficulty of this task would be greatly compounded if the company under examination were a first time user of the credit seeking to apply carry forward credits earned in prior years. For example, a company could be eligible to claim a tax credit in tax year 1983 but might not have a tax liability against which to apply the credit until tax year 1990. Any effective future examination of that claim thus would entail an analysis of records dating back to the 1980 to 1982 base period-a most challenging task.

Taxpayers have also experienced administrative problems with the credit. Representatives for 30 of the 92 companies included in our sample said that they had experienced difficulties in deciding whether_certain research and experimentation activities qualified for tax credit purposes. For example, they questioned whether certain modifications to an existing product qualified and whether certain software-related activities were allowable. Thirty-one companies told us that, once they had determined that they were engaging in qualified research activities, they had experienced difficulties in deciding whether various expenses associated with these activities could be included when computing the tax credit. This was particularly the case with indirect overhead and administrative wage and supply expenses. Further, 27 companies said that they had experienced difficulty in computing base period expenses. The companies pointed out that their accounting systems were not designed to segregate the required cost data. Therefore, they had found it necessary to reconstruct data and refine their accounting systems.

In any case, however, the aforementioned difficulties apparently were not insurmountable. According to the corporate representatives who had experienced difficulties, they generally sought to resolve questions through reliance on in-house legal and regulatory interpretations and/or private sector consultants. Some corporate representatives told us that the vagueness of the law and accompanying regulations encouraged them to claim the credit when in doubt. Some further noted that such a stance is appropriate given the lack of clarity in the law, the low risk of being selected for examination, and the low risk that a penalty will be applied in a situation where final regulations have not been issued.

In summary, Mr. Chairman, we know that a wide variety of companies have claimed the research and experimentation tax credit for many different activities. According to the Treasury Department, the primary tax year 1981 beneficiaries of the tax credit, in dollar terms, were a relatively small group of major corporations. In our judgmental sample of firms, manufacturing companies were the primary users of the credit.

Some available statistics indicate, at least on the surface, that expenditures for research and experimentation have increased since enactment of the tax credit. However, some of the increase may be attributable to inflation and/or to taxpayer reclassification errors. Other information indicates that the tax credit may not have been the only impetus for increased expenditures for research and experimentation. There also are questions as to whether a 25 percent tax credit is sufficient to stimulated a significant increase in research and experimentation, especially in high-risk areas. And, the credit provides little in a the way of an immediate incentive to spend increased funds on research and experimentation for firms that do not expect to incur a current year tax liability.

Concerning costs, we know that some corporations already have claimed millions of dollars in tax credits and that Treasury expects that the tax credit will result in foregone federal tax revenues of more than $7 billion through 1989. Also, both IRS and taxpayers have experienced some administrative difficulties with the credit.

Thus, we have some data and information on the effectiveness of the tax credit, but some of that data and informatio can lead to conflicting conclusions. Given that, we cannot now recommend that the credit be made permanent.

We recognize, however, that as an alternative to making the tax credit permanent, the Congress may decide to extend the expiration date. Should this be the case, we believe that the Congress should include an important proviso-that Treasury and/or IRS systematically gather data on the tax credit so that a more informed decision on its effectiveness can be made in the future.

If a decision is made to extend the credit, the Congress may also want to consider making some policy revisions. For example, the Congress may want to consider whether the size of the credit is appropriate. The Congress may also want to consider whether the law should take inflation into account, so as to limit the tax credit to real increases in spending on research. Finally, the Congress may want to clarify the law where possible. For example, we found that there is a need for clarification as to whether certain computer software-related expenses qualify for tax credit pur

poses. And, other witnesses may point out some additional areas where clarification of the law may be appropriate.

This concludes my prepared statement, Mr. Chairman. We would be pleaced to respond to any questions.

Chairman RANGEL. You have spent a lot of time talking about the amount of the credit and amount of spending.

How could you determine whether it was actual research or not? Mr. FINCH. We basically accepted the company's word for that, sir.

Mr. GUNNER. Mr. Chairman, that will have to await the result of IRS examinations. That is when IRS will be getting into the extent of actual spending on research versus changes in accounting methods or changes in what is classified as research.

I think the IRS audits now coming on line will yield results on that aspect.

Chairman RANGEL. If you are taking the word of the taxpayer, then you are not prepared to testify whether or not this credit serves any purpose at all except to reduce tax liability.

Mr. GUNNER. I think the piece of information that the companies gave us, the ones that said the credit doesn't do much for them, is perhaps the most important piece of information that we were able to develop.

Chairman RANGEL. So the best part of your testimony relates to those people who don't use it?

Mr. GUNNER. The remaining information is subject to IRS verification.

Mr. Chairman, we do not have authority to ask the companies for records.

Chairman RANGEL. I am not asking you for your authority. I am asking in terms of common sense. Someone writes off $10 million. They say they were engaged in research.

We want to ask what did you do besides spend the money? What research? What were the findings as a result of spending the money? What benefits did you receive? Was there a better product involved?

If someone put his brother-in-law in the room and tells him to think and then writes it off, all you do is say, "Well, they spent the money. It is up to IRS to determine whether or not they came up with any new ideas."

Is that right?

Mr. FINCH. That is basically the case, Mr. Chairman.

Chairman RANGEL. We hope that you expand your sample and that the IRS can be more cooperative in giving you additional information. Our mission is to find out whether we are wasting money. We trust the taxpayers, of course, but it doesn't seem like it is making a lot of sense that people are receiving the credit, but America is not receiving the benefit from it.

So, keep us apprised as to what additional information you are able to pull together.

We thank you.

Mr. FINCH. Thank you, sir.

Chairman RANGEL. We will get all the answers now from the Department of Treasury, Ronald Pearlman, Acting Assistant Secretary for Tax Policy.

I want to thank the administration for exposing us to this way of improving our economy and making us more competitive and moving toward reducing the deficit.

I hope that we can put your entire statement, Mr. Pearlman, into the record and perhaps somewhere in your remarks you can tell us what benefit we received from the credit.

STATEMENT OF RONALD A. PEARLMAN, ACTING ASSISTANT SECRETARY (TAX POLICY), DEPARTMENT OF THE TREASURY Mr. PEARLMAN. Thank you, Mr. Chairman.

I will try through some data that we have developed to report to you as best we can on our findings. I am appreciative of the opportunity of being here this morning to present the views of the administration on the research and experimental tax credit.

The need for and the appropriate form of the credit is a matter of continuing interest to the administration as well as to the subcommittee. We have an interagency working group that has been established, has been ongoing for some months, studying various aspects of the credit.

What I would like to do this morning is summarize our views today based on the data that we have available.

Chairman RANGEL. That will be fine, Mr. Secretary. Then it will help me and give me more time to ask some of the questions which I think the members would be interested in.

Mr. PEARLMAN. We think that the most important question with respect to the effectiveness of the credit, Mr. Chairman, is whether the definition of expenses qualifying for the credit has been drawn in a way to target the credit most efficiently.

The administration remains committed to this credit and favors extending it for an additional 3 years through 1988. As part of the extension, however, we strongly support tying the credit more closely to those private activities which are likely to result in technological innovation.

We think through a more carefully drafted definition the credit can provide an important incentive for taxpayers to undertake what we all would consider is essential research and experimentation.

Chairman RANGEL. Are you talking about targeting the credit? Mr. PEARLMAN. We are talking about targeting. We are talking about a refinement of current definition of R&E.

Chairman RANGEL. Would that language be in your testimony? Mr. PEARLMAN. Yes. I will try to review with you the principles. The written statement goes into much more detail.

The administration is determined to encourage the continued growth of domestic R&E. We believe that technological innovation is necessary, indeed essential, to the long-term vitality of the economy and that can only come from a substantial commitment to research and experimentation.

Without that commitment we believe our competitive position in the work economy will inevitably decline. Now, the problem with the concept of research and experimenation is that it describes and encompasses a broad variety of research activities by commercial as well as by not-for-profit corporations.

We have, for example, research embracing everything from very basic research tied to general scientific rather than specific commercial activities and research activities that do relate directly to commercial product development.

We believe that broad Government support of research and experimentation is particularly essential in the area of basic research, that is, research which by its nature cannot be expected to be self-supporting.

Although there are advances from basic research which ultimately will have commercial applications, and indeed that is one of the objectives of such research, those commercial applications are frequently too remote to permit private investors to support that research on a profitable basis.

For this reason, the Federal Government over the years has traditionally taken a lead role in funding basic research primarily through grants to universities and other educational nonprofit organizations. We believe that Government support of R&E also is appropriate, however, with respect to certain forms of research with direct commercial application.

A business will invest in R&E only to the point that the expected commercial returns are at least equal to that of alternative investments. Although opportunities for commercial profit ordinarily may reflect society's desired allocation of resources, the return to the private investor from R&E may not reflect society's true benefit from resulting innovation.

When market rewards for R&E expenditures do not match society's need for technological innovation, then it falls appropriately on Government to supply an additional incentive.

Now, under the current research and experimentation credit, the key is the definition of qualified research expenses. That definition contained in 44F of the Internal Revenue Code is defined by cross reference to a definition of research or experimental expenses under section 174 of the code which preceded the enactment of the credit.

The legislative history of the credit indicates that the existing, preexisting administrative interpretations under section 174 were intended to apply for purposes of section 44F. Regulations under section 174 define the term research or experimental expenditures to mean research and development cost in the experimental or laboratory sense and includes costs incurred to develop experimental or pilot models.

The current regulations under section 174 do not precisely identify the type of expenses qualifying as research or experimental expenditures.

Under section 174 this lack of precision may not be critical since many expenditures are deductible currently either as research or experimentation expenses or as ordinary business expenses under the general rule of section 162 of the code.

The same imprecision, however, in defining qualifying expenditures is not acceptable for purposes of the credit since it enables taxpayers to argue that virtually all preproduction costs associated with developing product qualify them for the credit. A definition which fails to focus on innovation dilutes the incentive for taxpayers to undertake truly innovative research.

Our experience with the credit is based on a very limited sample. It is based on 6 months of 1981 and it is based on preliminary tax return data for 1982.

I am going to present that data to you, as I indicated, in summary form orally, but I would caution the subcommittee as strongly as I possibly can to refrain from drawing firm conclusions from this data because only a limited period of time is involved.

Chairman RANGEL. That did not stop you from recommending an additional 3 years.

Mr. PEARLMAN. Indeed it is for that reason, Mr. Chairman, that we recommend an extension of the credit for an additional 3 years as distinguished from a permanent extension.

We believe that the current data is not adequate to derive conclusions that would go into designing a credit on a permanent basis, but we believe that as data begin to be picked up on a more definitive basis over the next several years we will have the kind of data that will enable Congress to more critically and accurately evaluate the effectiveness of the credit.

The available data that we do have, and I might point out that we did review the tax returns in which research and experimentation credits were claimed, both in 1981 and 1982, although as I noted, the data for 1982 are limited because not all of that data is yet in our system, the data that we do have do indicate that in connection with the use of the credit, that the credit was claimed primarily by large corporations.

The 12,000 corporate taxpayers claiming credit in 1981 represented 39 percent of total corporate assets and receipts, less than 1 percent of all corporations filing 1981 returns.

Now, the greatest use of the credit by large corporations did not come as a surprise to us since those corporations typically would be ones that would have more extensive research and experimentation programs with relatively greater resources to devote to R&E.

One of the things that we did find that was interesting and we thought positive was that there were basically four industry categories in which most of the R&E credits were claimed and that a high percentage of the claimed credits were indeed being claimed in industries which would be typically associated with high technology.

In tables that have been attached to our written statement we detail those industries and amounts of the credits claimed by those industries.

In connection with the growth of R&E expenditures, our data for 1981 indicated that corporate taxpayers reported an average growth rate in research and experimentation expenditures from the pre-1981 period to 1981 of 40.3 percent.

Twenty percent of the corporate returns with tentative R&D credits reported growth rates in R&E in excess of 200 percent and an additional 18 percent reported growth rates between 100 and 200 percent.

For the most part, the industries reporting relatively high increases in R&E expenses received a relatively small fraction of all R&E credits averaging less than 2 percent each of the total tentative credits.