We have thus as yet been unable to detect reliable evidence that the credit is having a positive impact on total R&D expenditures. Our analysis is still too incomplete, however, to sustain any definite conclusion on this issue.

While companies interested in tax benefits would understandably wish to see the credit extended, analytical considerations indicate that the credit will become increasingly effective in 1983, 1984 and 1985 if it is not expected to be extended. For if firms do not anticipate that the credit will be extended, the loss of future credits as a result of increasing current expenditures will be reduced by one-third for expenditures undertaken in 1983, by two-thirds for expenditures undertaken in 1984 and entirely for expenditures undertaken in 1985. By that last year, for firms with tax liability and with qualified R&D above base but less than 100 percent above base, the effective rate of credit would be raised to the full 25 percent.

Without judging the issue as to whether any R&D tax credit is socially desirable, certain revisions would seem likely to improve the one currently in effect.

First, while the 50 percent limitation does not apparently relate to a large proportion of R&D, its negative incentive effects can be considerable where it does come into play. To the extent that our major concern is for increasing R&D expenditures and not for distributing the benefits of reduced taxes in some more egalitarian fashion, it would appear that the 50 percent base limitation might well be removed. If it is not entirely removed, at the least, future bases should take into account the limitation on the current base. Where the 50 percent limitation is currently in force, each dollar of current qualified expenditure in excess of base might better generate only 50 cents toward increasing future bases. This would eliminate the anomaly by which, for firms in this category, each dollar increase in R&D expenditures would generate only 12.5 cents in current credit while losing 25 cents in future credits.

A second significant revision, in the interest of increasing incentives, would entail eliminating or reducing the asymmetry as between firms with R&D above base and firms with R&D below base. This might of course be accomplished directly by levying a negative credit, or an additional tax, equal to 25 percent of reductions in qualified R&D spending. Since this would certainly be viewed with extreme distaste by firms affected, a more practical solution would be to combine such a tax for declining R&D with a general credit on all R&D over, say, 80 percent of base. Thus firms with decreasing R&D would lose some of the general credit and would hence be discouraged from such reductions, but might never have to pay a net tax relating to R&D.

The most important and fundamental revision, however, would relate to altering the moving, company-specific nature of the base. It is this feature which results in firms generally losing dollar for dollar in future credits the amounts that they gain in current credits from increasing their R&D expenditures. Under the current law, firms whose R&D is increasing and expected to increase over base within the zero to 100 percent range, and who have and expect to have tax liabilities to be offset, in effect gain only the interest savings of postponing taxes over a three year period, rather than eliminating them.4

Several methods are available which would alter this provision in the interest of increasing the effective incentive. First, the company-specific base could be retained but left fixed at, say, the average of 1980, 1981 and 1982 qualified R&D expenditures (with possibly some exceptions or adjustments for new firms). This would imply rapidly growing amounts of R&D credits over the years but would mean that the effective rate of credit would be equal to the statutory rate for firms which have current tax liabilities and are not subject to the 50 percent base limitation. It would also generally offer an incentive to firms to minimize their reductions in R&D.

Somewhat more complicated to administer, but less costly in terms of tax revenues, would be a provision to start with a company-specific base, say of R&D expenditures of 1980, 1981 and 1982, but to adjust that base in accordance with national or industry movements in R&D. Thus if a firm were in an industry where R&D in 1983 grew by 10 percent, its base in calculating its R&D tax credit for 1984 would be raised by 10 percent. The credit for each firm would hence relate to increases in R&D spending over a base which is both company-specific but adjusting to movements of R&D among companies in similar situations. Unless the industry were unduly small in relation to the size of the firm--and this could be avoided by appropriate grouping of industries-each firm would have a full 25 percent effective credit as an incentive to increase its expenditures. Its increased expenditures would contribute to raising the base and reducing future credits for everybody but would

* These interest savings would, however, prove a perpetual stream if the indicated conditions remain in effect.

have a trivial effect in raising its own base and reducing its own future credits. By having the base increase on the basis of industry experience rather than companyspecific experience, we would achieve the maximum incentive impact along with the minimal Treasury tax loss associated with an incremental credit.

Finally, the impact of the R&D tax credit would be increased by making it refundable or a direct subsidy rather than a credit. This would mean that the considerable number of firms in the economy not paying federal income taxes would also have an incentive to increase R&D spending. With the vastly increased depreciation deductions implicit in the accelerated cost recovery system, the investment tax credit and credits for foreign taxes, we have reached the point where many existing credits, let alone new ones, can be realized only by selling or merging firms. Unless we are to restore the corporate income tax by eliminating most or all of the existing incentives, deductions or loopholes, depending on one's view, new credits such as those for R&D would have to be refundable, regardless of taxes, to avoid significant losses of whatever incentive effects they have.

Alternatively, and this may well be a better path to take, the corporate income tax as such might be eliminated, presumably with shares in real corporate earnings attributed to the individual stockholders. The business tax incentives or "tax expenditures” might be eliminated and we could readily move to direct subsidies of those private expenditures and activities which we find it in the social interest to encourage.

TABLE 2.1.-OFFICE OF TAX ANALYSIS RETURNS FOR 1981: BASE, QUALIFIED R&D, CREDIT WITHOUT

LIMITATIONS, LIMITATION LOSSES, TENTATIVE CREDIT, TAX LOSS, CREDIT CURRENTLY CLAIMED, AND RATE OF GROWTH OF QUALIFIED R&D OVER BASE, BY INDUSTRY

(Dollars in thousands)

Agriculture

$400
Agncultural services... 600
Oil and gas drilling ... 1,330
Oil and gas field
Services

1,380 Mining

1,400 Construction.

1,500 Meat products.

2010 Dairy products.. 2,020 Preserved fruits and

vegetables. 2,030 Grain milling 2,040 Bakery, sugar,

confections. 2,050 Alcoholic beverages. 2,080 Other food; tobacco.... 2,090 Weaving mills

2,228 Knitting mills.

2,250 Other textiles

2,298 Apparel..

2,300 Logging and wood products

2,400 Furniture

2,500 Paper

2,600 Newspapers and

periodicals 2,700 Books and other

publishing ... 2,735 Commercial printing... 2,799 Industrial chemicals ... 2,815 Drugs...

2,830 Soap products 2,840 Paints.

2,850 Agricultural and

169

53 26 862

1 52 48 353

402

692 1,679

168

119 1,173 1,198

42 72 892 79 38 287 93

360 620 787 89 81 886 1,105

42 32 43 38 259 44 56

949

16,041 17,256

171 1,221 1,551

TABLE 2.1.-OFFICE OF TAX ANALYSIS RETURNS FOR 1981: BASE, QUALIFIED R&D, CREDIT WITHOUT

LIMITATIONS, LIMITATION LOSSES, TENTATIVE CREDIT, TAX LOSS, CREDIT CURRENTLY CLAIMED, AND RATE OF GROWTH OF QUALIFIED R&D OVER BASE, BY INDUSTRY —Continued

(Dollars in thousands)

Steel.

Refining

2,900 Rubber

3,050 Plastic

3,070 Leather.

3.100 Glass

3,225 Cement and concrete.

3,240 Other stone, clay, et cetera

3,298

3,370 Nonferrous metals. 3,380 Containers.

3,410 Hand tools,

hardware, cutlery... 3,428 Plumbing...

3,430 Structural metal 3,440 Forgings and

stampings.. 3,460 Engraving, coating, et cetera...

3,470 Ordnance.

3,480 Miscellaneous

fabricated metal.... 3,490 Farm machinery .. 3,520 Construction

machinery 3,530 Metalworking

machinery... 3,540 Special industrial

machinery. 3,550 General industrial

machinery.... 3,560 Office and

computing

machinery...... 3,570 Other nonelectrical

machinery ......... 3,598 Radio and television... 3,611 Household

appliances... 3,630 Electronic

components, et cetera.

3,670 Other electrical

equipment........ 3,698 Motor vehicles and

equipment.. ...... 3,710 Areospace..

3,725 Shipbuilding...... 3,730 Other transportation

equipment 3,798 Scientific

instruments;

clocks..... 3,815 Optical and medical goods.....

3,845

4,587

297

4,290

1,482

2,808

47

3,289

1,169

TABLE 2.1.-OFFICE OF TAX ANALYSIS RETURNS FOR 1981: BASE, QUALIFIED R&D, CREDIT WITHOUT

LIMITATIONS, LIMITATION LOSSES, TENTATIVE CREDIT, TAX LOSS, CREDIT CURRENTLY CLAIMED, AND RATE OF GROWTH OF QUALIFIED R&D OVER BASE, BY INDUSTRY —Continued

(Dollars in thousands)

Photographic

equipment.. 3,860 Other manufacturing .. 3,998 Transportation services

4,000 Communications 4,800 Utilities........

4,900 Grocery wholesaling ... 5,004 Machinery

wholesaling 5,008 Motor vehicle

wholesaling... 5,010 Other durables

wholesaling. 5,020 Electrical goods

wholesaling 5,060 Paper Wholesaling 5,110 Drug wholesaling 5,129 Apparel wholesaling ... 5,130 Chemical and oil

wholesaling. 5,160 Food and

miscellaneous

wholesaling 5,190 Hardware and

building materials... 5,220 Food retailing ... 5,300 Auto dealers and

service stations...... 5,500 Apparel stores. 5,600 Furniture stores.. 5,700 Restaurants

5,800 Drug stores.

5,912 Other retailing 5,995 Banking...

6,000 Credit agencies; brokers.

6,100 Life insurance 6,355 Other insurance.

6,359 Insurance agents......... 6,411 Real estate

investment cos 6,511 Lodging; personal services.

7,000 Advertising and

business services... 7,300 Repair services

(including auto) .... 7,500 Amusement and

recreation services.

7,800 Medical services. 8,000 Other services 8,999

TABLE 3.1.-R&D EXPENDITURE AND TAX CREDIT ESTIMATES FROM BUSINESS WEEK-COMPUSTAT

SEC COMPILATIONS

(In millions of dollars)

69

6 21 7

3 104 20

1 103 19 35 21 60

4,527

361

156 3,020 1,347

100 2,976 1,502 1,066

643 2,207 4,717

4,454

291

137 2,136 1,189

114 2,133 1,317

764

464 1,579 3,334

567 233 514 768 663 352

188 1,018

493 233 538

Aerospace.
Appliances
Automotive: Cars, trucks.
Automotive: Parts, equipment
Building materials.
Chemicals..
Conglomerates.
Containers
Drugs.
Electrical
Electronics..
Food and beverages
Fuel..
Into. Proc.: Computers.
Info. Proc. Office equipment
Info. Proc.: Peripherals, services.
Instruments.
Leisure time.
Machinery: Farm, construction.
Machinery: Industrial..
Metals and mining ..
Miscellaneous manufacturing.
Oil services and supplies.
Paper
Personal and home care..
Semiconductors.
Steel...
Telecommunications..
Textiles and apparel.
Tires and rubber..
Tobacco

4,512
345

150
2,598
1,266

121 2,508 1,460

922

553 2,079 3,845

709 315 605 883 735 404 212 1.152 642 256 607

177

15 7 5 14 17 41 11

5 28 29 56 30 19 25 14 36 58 17 16 43 13 84 19 16 21 9 5 15 11 10 1

769

406

683 1,013

31 2 9 6 1 47 '11

1 38 14 17

9 50 51 15

8 10 12 7 6 3 14 15 3 7 8 4 15 1 6 0

761

416

211 1,254

792 273 710

22 21 20 30 10 8 3 28 35

6 22 16

6 42

1 11 0

TABLE 3.2.- PERCENT INCREASES IN R&D AND TENTATIVE TAX CREDIT FROM BUSINESS WEEK

COMPUSTAT-SEC COMPILATION

Percent increases in R&D

Industry

Percent increase in

tentative credit, 1981

82

1980-81

1981-82