A substantial portion of the United Kingdom's .11.international transactions in long-term capital is comprised of inflows and outflows of direct investment in foreign subsidiaries and branches, whether by acquisition of share and loan capital, retention of profits, changes in branch indebtedness, or by changes in intercompany accounts. These transactions have historically been very large. On average, since 1961 U. K. firms have invested abroad in direct investments about 11.5%o as much as they have invested at home. On a per capita basis, the U. K. is one of the world's largest foreign investors. These flows are of considerable importance for the balance of payments. The quarterly value of the deficit on direct investment account has averaged ?28 million since 1961, compared with an average surplus on current account of ?34 million, and the variance of the balance on direct investment is about 1 1% of the variance of the balance on current account. Section II describes a model of direct investment flows. This model is tested in section III. The final section assesses some of the implications of the estimated model.

C URRENT controversy among economists on systems for recording and classifying Government transactions has focused largely on the choice between the Cash Budget on the one hand and the income and product account on the other; the Administrative Budget seems to have few friends in the profession. A good part of the controversy over budget concepts appears to exist mainly because the major types of Federal economic influence are not being distinguished sharply enough in discussion of the currently available measures of Government activities. In particular, GNPdeficit advocates look for the income and spending effects of Federal transactions, while the Cash Budget defenders seem to be looking principally for liquidity and financial market effects. Lending effects, meanwhile, have been largely left out of the discussion; they are usually mentioned only as an intrusive element of the cash deficit relative to income and spending effects. When lending is recognized as a separate and, at times, sizable force in the economy, and when the spending vs. liquidity distinction is kept in mind, we begin to have a basis for deciding which of the Federal budget figures to use in any particular discussion. What is proposed here is a framework within which each of the budget concepts can be integrated but still distinguished, permitting whatever focus of discussion is desired without foreclosing on the benefits of alternative perspectives. The framework is an adaptation of the account for the Federal Government sector in the Flow of Funds national accounting system developed at the Federal Reserve, and now available on a quarterly, seasonally adjusted basis. This adaptation is based on the following premises: first, no single number can possibly tell all or even very much about Federal economic influences, and we need a set of several figures even to begin to see a picture. Second, the accrual version of Federal transactions in income and product accounts tells more of non-financial influences than the cash version. Third, Federal financial transactionsboth lending and borrowing-are important channels through which the Government can and does influence economic developments and should be included explicitly in any accounting of Government activities. Fourth, lending is different from both borrowing and spending in influence and cannot reasonably be combined with either. Fifth, however, simplicity of presentation is essential if any new version of Government accounting is to achieve fairly widespread use. The form shown in the attached chart tries to meet the preferences and objectives stated above. It delineates in a minimum number of lines the major types of Governmental activities as they affect distinctively different types of economic activity: payments for goods and services and for transfers, receipts from tax revenues, Government lending, Government borrowing from the banking system, and borrowing from the public. A. The upper panel of the chart shows the Government's direct contribution to the spending stream and the amount it taps from this stream through taxes. Changes in the gross size and composition of both spending and revenues are at least as important for economic analysis as a net deficit or surplus on any accounting basis and deserve separate reporting. The figures plotted are those in the national income account, that is, they are the accrual version of Government spending and receipts. It would be definitely desirable, if space permitted, to distinguish spending for goods and services from transfer payments and to distinguish personal from business taxes, but as a * Views expressed here are those of the authors, and do not necessarily reflect the views of the Board of Governors of the Federal Reserve System. ressed here are those of the authors, an do sa il reflect the views of the Board of Governors

This paper estimates the degree of stickiness in aggregate consumption growth (sometimes interpreted as reflecting consumption habits) for thirteen advanced economies. We find that after controlling for measurement error, consumption growth has a high degree of autocorrelation, with a stickiness parameter of about 0.7 on average across countries. The sticky consumption growth model outperforms the random walk model of Hall (1978) and typically fits the data better than the popular Mankiw (1989) model, though in a few countries, the sticky consumption growth and Campbell-Mankiw models work about equally well.