Retained Earnings and Global Imbalances

Cian Allen

Paper prepared for the 16th Conference of IAOS OECD Headquarters, Paris, France, 19-21 September 2018

Session 1C, Day 1, 19/09, 11h00: General issues related to dealing with globalisation

Cian Allen [email protected] Trinity College Dublin

Retained Earnings and Global Imbalances

DRAFT VERSION 31/08/2018 PLEASE DO NOT CITE

Prepared for the 16th Conference of the International Association of Official Statisticians (IAOS) OECD Headquarters, Paris, France, 19-21 September 2018

Retained Earnings & Global Imbalances

Cían Allen⇤

Trinity College Dublin

August 2018

[VERY PRELIMINARY - DO NOT CITE]

Abstract

This paper shows that old-fashioned national accounting principles can significantly cloud the interpretation of key macroeconomic statistics in highly globalized economies. Measuring cross- border investment income in a globalized world is fraught with difficulties. A key concept in the balance of payments is the difference in the recording of retained earnings between foreign direct investors and foreign portfolio investors. Retained profits are attributed to direct investors, whereas only distributed earnings (via dividends) are attributed to portfolio investors. In the past few decades, a growing share of corporate profits have been retained within the firm and not distributed to shareholders. At the same time, cross-border holdings of firms have increased dramatically, amplifying the distortions created by this principle. We argue the recording of income should be independent of the timing of distribution of profits in order to better capture the underlying resources available to countries. We show that treating foreign direct and portfolio investors in a consistent manner can lead to a significant redistribution of (recorded) income between highly globalized economies.

JEL-Code:F31,F32,E21 Keywords: global imbalances, current account, corporate proﬁts, ﬂow of funds

⇤Cían Allen ([email protected]), PhD candidate at Trinity College Dublin under the supervision of Philip R. Lane, whom I thank for ideas and guidance. Part of this article was drafted while I was visiting UC, Berkeley. I am grateful to Pierre-Olivier Gourinchas for making this possible. This research is supported by the Government of Ireland Postgraduate Scholarship programme from the Irish Research Council. All remaining errors are my own.

1 1 Introduction

The past few decades have seen a considerable rise in retained corporate profit in advanced economies. At the same time, global firms and global investors have accumulated a growing share of the ownership of domestic firms. These trends have made the measurement of cross-border investment income considerably more difficult, in particular as statistical conventions have struggled to adapt to globalization. This paper focuses on a particular implication of these trends, the distortions of cross-border investment income due to the attribution of retained earnings to direct investors and not to portfolio investors. We reapportion the undistributed income to portfolio investors using the national accounts and construct corrected measures of the current account and gross national income. This leads to a significant redistribution of income between highly globalized economies. Our correction can give policymakers a better measure of the underlying net resources available to the country, an essential tool for prudent policy (Obstfeld, 2012). It also sheds light on the share of valuation changes on foreign assets (stock-flow adjustment) due to current production, i.e. retained earnings, and other capital gains due to asset price changes for instance. Both the national accounts (UN, 2008) and balance of payments (IMF, 2009) attribute earnings to direct investors as they are earned whereas portfolio investors are attributed profits as they are paid out in the form of dividends.1 Hence, part of the earnings of portfolio investors will not show up in the property income of the current account and in the Gross National Income (GNI). This means that if the distribution of portfolio investment is not uniform across countries, there will be an under/overstatement of the current account and the GNI across countries. We follow Lane (2017) in defining the best measure of the underlying net resources available for a country is one that is independent to the way firms decide to payout their earnings. With this in mind, there are different corrections possible. One direction is not to impute any flows and to treat domestic, foreign portfolio and foreign direct investors in a similar manner. However, a move away from market values might not be adapted to capture the net resources of countries. In addition, attributing all of the valuation gains and losses to the flow variables is not satisfactory as we are not distinguishing between part of the valuation changes of wealth that are due to production (retained earnings) and windfall gains (a price change due to expectations changes or unforeseen changes in productivity or growth).2

1The underlying rationale for this is based on direct investment representing control of the firm and a lasting relationship. 2There is an important conceptual difference between increases in the share price due to expected retained earnings and unexpected shocks to the share price. Both show up in revaluation account, but the former is not of a windfall nature. A large share of the changes in the revaluation account is holding gains of the windfall nature. However, in the past few years, retained earnings have grown more prominent. A windfall profit to shareholders comes because the net present value of expected future income streams is suddenly higher. However, retained earnings also increase

2 The correction we put forward is in between these two directions. Using the share of portfolio holdings in domestic outstanding equity, we reapportioning the share of retained earnings on domestic profits to foreign portfolio investors. This correction treats portfolio and direct investors in a similar manner. We then calculate a modified investment income and of gross national income, closer to the underlying net resources available to countries. This will also allow us to disentangle the valuation changes in portfolio investment due to (expected) retained earnings and (unexpected) capital gains due to other factors. To make things clear, consider the simple following example. Suppose a firm in country A is 50 percent owned by a direct investor in country B, and 50 percent held by multiple portfolio investors from country C (none of these investors reach the 10 percent threshold of direct investment). Sup- pose this firm makes a profit of 100 and decides to retain all of this profit.3 Investors in country B will get attributed the retained profits proportional to their holdings, i.e. 50, whereas investors in country C will not (everything else equal the value of their holdings should, however, increase in proportion). Therefore, in terms of cross-border flows, country B will receive investment income of 50 (improving the current account by the same amount), whereas country C will record no flows in the current account. (its net foreign asset position will, however, improve by 50). In addition, the gross national income of country B will increase by 50 (whereas there is no change to country C). Finally, the net lending balance of the corporate sector of country A will increase by 50, as will its GNI. It is important to note that this should be "temporary" inequality in nature. At some point, these earnings should go to the shareholders. It should also be noted that the ultimate impact on the balance sheets of direct or portfolio investors should not be different, as the holdings of portfolio investors will increase in proportion to the retained earnings of their holdings, everything else equal. As we can see, the inconsistencies in treatment between types of foreign investors may cloud the interpretation of the measured net investment income of the current account, gross national income and gross saving between countries.4 These measurement issues are compounded by the fact that a large share of direct investors are multinational enterprises with significant foreign ownership. Thus, if these domestic multinationals retain earnings, it will contribute to domestic savings, whereas if they distribute the earnings, the funds will flow to foreign investors. Our contribution is linked to the growing literature focusing on corporate saving, see Poterba (1987), Eggelte et al. (2014), Gruber and Kamin (2016), Cesaroni et al. (2017) and Chen et al. the share price, but the shareholders expected the retained earnings and already discounted them in the share price. Merging these two concepts could be problematic. The large share of the revaluation account are holding gains of the windfall nature. However, in the past few years, retained earnings have grown more prominent. 3We assume away any taxes or payments to bondholders for simplicity, hence interest on debt instruments claims 4This example illustrates the inequality between countries because of the difference in treatment between foreign investor types. However, earnings attributed to domestic investors are also on a cash basis, creating similar implications for the distribution of income between sectors within countries.

3 (2017). It is also related to the debate surrounding the US’s external position, see Gourinchas and Rey (2007). In addition, it follows Dalgaard et al. (2000) in calling for satellite accounts to get a satisfactory description of income distribution in the national accounts. It is also related to the recent papers that highlighted the distortionary effect global firms have had on the national accounts and balance of payments. Guvenen et al. (2017) show MNEs distort measures of value added and productivity, leading to an underestimation of the GDP of the home country of the MNE and a corresponding overestimation of GDP in countries hosting MNE. Lane (2015) shows financial engineering maybe at the source of the some of the deterioration in the measured current account of the UK and Avdjiev et al. (2018) documents the footprint of MNEs in balance of payment statistics. This paper is complementary to the independent and contemporaneous work by IMF (2018) and Fischer et al. (2018) who also correct the current account for retained earnings on portfolio equity. These papers differ from this one, as they use stock market data in order to estimate the correction, whereas this paper uses the national accounts, encompassing both listed and non-listed firms.5 The rest of this paper is structured in the following manner. First, we present the key national accounts identities and present our reapportioning method. Next, we document the rise in retained profits and foreign ownership in the corporate sector across advanced economies. In section 4, we present our adjusted measures of gross national income and the current account. Section 5 offers some conclusions and possible extensions.

2 Conceptual Framework

Let’s begin by taking a look at the basic national accounts and balance of payment concepts that we use in this paper. We will then illustrate the impact on key macro measures of the inconsistencies in the treatment between diﬀerent investors in the national accounts. Finally, we will expose our method for reapportioning part of domestic retained earnings to portfolio investors.

2.1 National Accounting

We are interested in national account concepts related to the total economy and to the corporate sector in particular. In relation to the total economy, as one of the contributions of this paper is to study the net investment income (NINC) of a country, we see that it shows up in the calculation of both gross national income, which is equal to gross domestic product plus the net investment income from

5Using stock-market data can be problematic for highly globalized economies, as large ﬁrms are not necessarily listed in the country in question, like in Switzerland and Ireland for instance. The IMF’s annual consultation of Switzerland mentions an overestimation of the current account due to the asymmetrical treatment of foreign ownership. They ﬁnd overestimations ranging from 3 percent of GDP in 2014 to 7 percent in 2006. It is not clear if they just correct for the debit side of portfolio holdings or also for the credit side like this paper.

4 abroad, and the current account, which is equal to the trade balance (exports minus imports, X-M) plus the net investment income from abroad (NINC).6

GNI = GDP + NINC (1) CA =(X M)+NINC The net investment income balance can be decomposed into the income received from assets minus the income paid on liabilities for the diﬀerent asset classes (foreign direct investment, FDI; portfolio investment, PI; and other investment, OTH), in the following manner:

NINC =(INCFDIA INCFDIL)+(INCPIA INCPIIL)+(INCOTHA INCOTHL) (2)

Direct investment is the relationship between a resident enterprise and a non-resident foreign investor which owns 10 per cent or more of the ordinary shares or voting power of an incorporated enterprise (or equivalent). In practice, it is the relationship between affiliates (subsidiaries and parents) or associates, i.e. multi-national corporations (MNEs). Investors are said to have a certain influence over decisions due to their lasting relationship. On the other hand, portfolio investment is all the cross-border transactions and positions involving debt or equity securities, other than those included in direct investment or reserve assets (IMF, 2009). The distinction between the type of investors is crucial when it comes to the timing of the allocation of earnings in the national accounts. Earnings of direct investment enterprises which are not distributed as dividends to direct investors are called reinvested earnings. The IMF records reinvested earnings as being distributed to direct investors in proportion to their equity ownership in the enterprise and then being reinvested into the same enterprise. They are recorded in the current account as income on direct investment and in the financial account as a transaction in equity. In the case of portfolio investment (and domestic investment), retained earnings are not assigned to shareholders as income. They are recorded as the saving of the enterprise and the increase in the value of the enterprise is recorded in the national accounts as a revaluation. Net investment income can further be decomposed into equity and debt. If we look just at the equity component, the net investment income on FDI equity and PI equity. We see that

NINCFDIEQ = NDIV FDIAEQ + NREFDIEQ (3) NINCPIEQ = NDIV PIEQ

The underlying rationale is to allocate saving to shareholders. The earnings of an enterprise accrue

6To be more precise it is the trade balance plus net factor income, that includes income from labor and capital. In practice, net labor income is relatively small in advanced economies.

5 to investors as they are earned. As the earnings are available to the enterprise for its use, they are deemed to be reinvested in the enterprise. Portfolio investors are seen to have little influence on the management of an enterprise and therefore have little input into the enterprises’ saving decisions. However, this creates inconsistencies, as this principle is not extended to cross-border portfolio investment and domestic resident-to-resident investment relationships. These inconstancies show up in the savings of firms, in gross national income and the current account. These distortions are magnified by the fact that these retained earnings accrue to MNEs, as a large share of the ownership of MNEs is by foreign investors. The current treatment of net investment income means we attribute saving to firms that are owned by residents or portfolio investors but not to firms with direct investors. The amount of saving that is recorded for an enterprise depends on the type of investors that own the enterprise. To see this, let us look at the sector accounts of the corporate sector. We usually view the role of the financial system as channeling household savings to firms in order to finance productive activities. The disposable income of firms is its savings. If the corporate sector does not use these funds to invest in fixed (or current) assets (or for capital transfers), the sector is a net lender vis a vis of other sectors.7 In the national accounts, the profits of firms are measured by the corporate operating surplus (GOS).8 Part of this return is used to pay interest to lenders, and to pay corporate taxes. The residual is the accounting profits that can either be distributed to shareholders in the form of dividends or retained in the firm. The later retained earnings constitute the corporate sector’s saving. Corporate saving represents an internal source of funds available to firms to either invest in non-financial or financial assets or to reduce debt.

GOS = GS + NDIV NREFDI + other (4) Finally, an additional implication of the current treatment of net investment income concerns the valuation adjustments to external wealth. We can decompose the change in the net international investment position (NIIP) between t and t 1 in the following manner:9

NIIPt NIIPt 1 = CAt + SFAt (5) 7However, given national accounts conventions, it is important to bear in mind that only domestic investment is taken into account. Invest abroad, in the form of FDI for instance, is not taken into account (and is recorded in the ﬁnancial account of the BOP). 8Gross operating surplus of the NFC sector is income after subtracting compensation of employees and taxes on production and imports less subsidies from gross value added. 9 We have FAt = (CAt +KAt +EOt). FAt is the ﬁnancial account balance, KAt pertains to the capital account balance and EOt is the net errors and omissions. For simplicity, we assume KAt and EOt are equal to 0.

6 where SFAt is the Stock-Flow Adjustment term, is often used as a proxy for revaluation changes.

The Stock-Flow Adjustment term is composed of a valuation term, KGt, the net capital gain on the existing holdings of foreign assets and liabilities, and a term capturing net other non-ﬂow changes to the net international investment position, OTHERt, (for example, due to changes in reporting methods and data revisions).10 In addition, as noted above, part of the change in stock not PI captured by the current account is due to retained earnings on portfolio investments, REt (equal to a proportional change in the net stock of portfolio investment everything else equal). However, these capital gains are in nature very diﬀerent to windfall capital gains. Thus teasing out the part that is due to retained earnings on portfolio investment can potentially improve our understanding of external valuation adjustments.

PI SFAt = KGt + OTHERt + REt (6)

2.2 Reapportioning Method

In order to have a consistent treatment across types of firms and investors, we impute retained earnings to portfolio investors and reapportion the resulting income to the country of residence of the investor.11 As portfolio investment is present of both sides of the balance sheet, we need to take into account both credit and debit flows. Let’s take country A for instance. We calculate the imputed retained earnings of attributed A A to foreign portfolio investors, it, as the share of portfolio equity liabilities (P EQL )inthetotal outstanding stock of domestic equities liabilities (EQA), multiplied by the total retained earnings firms (measured by gross savings of corporates, that is after-tax profits less dividends to shareholders, GSA).

A A P EQLit 1 A it = A GSit (7) EQit 1 ⇥ The underlying assumption (A): GS is approximately distributed across DI and PI according to the ownership shares (consistent with Chen et al. 2017). We can test this by using the DI share and comparing it to the measured retained earnings on FDI (debit).12 On the debit side of country A’s current account, we reapportioning these imputed earnings to the country of residents of the investors using country weights derived from portfolio equity assets and liabilities holdings from the IMF’s CPIS database.

10See Curcuru et al. (2008) for further discussion on the importance of the net other statistical term. 11We can also impute the remaining proﬁts to resident investors (GOV and HH). 12We use gross saving as the cost of depreciation of foreign-owned capital should be borne by the foreign shareholder. However, savings net of depreciation are used by BOP to measure reinvested earnings. We compute the net measure to test assumption A.

7 A A PEQLB !B = (8) P EQLA For allocating retained earnings from country A to B, we use the share of foreign liabilities to A country B relative to all foreign liabilities of country A, !B. We can then reallocate the retained A A earnings ( ) according to their share in the total portfolio equity liabilities holdings (!B). Now turning to the credit side of country A’s current account, we want to reallocate back to country A its share of retained earnings in all other countries. Taking the example of its share in country B’s retained earnings on portfolio investment, we multiply the share of portfolio equity B assets of country A in B (alternatively the liabilities of country B to A, !A ) by the imputed retained earnings of country B, B. The results for 2015 are Table 2. This reallocation will change our measure of gross national income, the current account and the saving of the countries in question. Given that our sample of countries does not cover the world, there is a potential downward bias of the credit side of the correction for portfolio retained earnings. Indeed, we rely on the available data concerning the host country of the portfolio investment asset. This, in turn, might overestimate the negative impact on the current account, gross national income, and other measures. However, Table 1 shows that we cover a large share of the partner countries in the country’s portfolio. Most of the missing partner countries tax havens, like the Cayman Islands or Bermuda for instance. We show that allocating foreign assets of the Cayman Islands to its partners does not change signiﬁcantly the results, see Figure 7 for values for 2015 for instance.

3 Stylized Facts

[To be updated]

Stylized fact 1: Steady increase in corporate earnings (distributed and retained) in past few • decades

– Figure 1 shows the general increase in our sample of countries. The median proﬁt to GDP increased from 20 percent to over 26 percent of GDP since 1995. This has led to an increase in both distributed (a doubling of the dividends) and undistributed proﬁts (gross saving).

– On the stock side, the market value of domestic equities outstanding (both listed and non-listed) has also expanded rapidly. Figure 4 shows a median increase of 100 percent of GDP to 150 percent of GDP.

Stylized fact 2: Increase in foreign ownership and foreign claims on these proﬁts •

8 – Figure 2 shows that the median share of foreign ownership of domestic equities has increased from under 20 percent to over 30 percent since 1995. Foreign investors can be broken down into two types of investors depending on their ownership share. Direct Investment has increased from under 10 percent of GDP to just under 20 percent (with the 75th percentile reaching over 40 percent of GDP). Portfolio investment increased from under 5 percent to under 10 percent of GDP.

– In terms of flows, this translates into an increase in the share of domestic profit flowing abroad. Figure 3 shows the median share more than doubled since 1995 to over 10 percent (i.e. around 3 percent of GDP).

4 Adjusted Measures of the Current Account and Gross National Income

[To be updated]

Net Retained Earnings on Portfolio Equity •

– The median share of gross savings in 2015 imputed to portfolio investors is 10 percent, 17 percent for direct investors, see Figure 513

– The net correction (retained earnings on portfolio assets minus liabilities ) is shown in Figure 6 for the period 2011-2015 and Figure A1 and Figure A2 for 2008-2011 and 2005-2007 respectively

– Larger corrections are linked to large net portfolio equity positions but also gross saving, see Table 3

– We can adjust for foreign assets in ﬁnancial havens, the results do not change much (Figure 7)

– We can test our assumption by comparing our measure of imputed retained earnings for direct investor s and the measured retained earnings. See Figure A5 for the United Kingdom

Corrected Current Account and Gross National Income • – Table 4 shows the corrected current account for the 2012-2015 period

– Table 5 shows the growth rate of GNI can change signiﬁcantly when correcting for retained earnings on portfolio investment 13Figure A6 shows there is heterogeneity across countries

9 External Valuation Adjustments • – Figure 8 for 2012-2015 shows that our net correction can account for a large share of the stock-ﬂow adjustment term for countries like Norway and Sweden, but can go in opposite directions in many countries

– For averages between 2008-2011 see Figure A3 and for averages between 2005-2007 see Figure A4

5 Concluding Remarks

This paper shows that allocating retained earnings to portfolio investors can lead to a signiﬁcant redistribution of income between highly globalized economies. The correction accentuates and decreases global imbalances depending on the country’s net portfolio equity position and corporate gross saving. Even though the correction can be relatively low in some OECD countries, the challenge can still be signiﬁcant, given current accounts are on average relatively balanced.

10 References

Avdjiev, S., Everett, M., Lane, P. R., and Shin, H. S. Tracking the international footprints of global ﬁrms. BIS Quarterly Review,March2018.

Cesaroni, T., Bonis, R. D., and Infante, L. Firms’ ﬁnancial surpluses in advanced economies: the role of net foreign direct investments. Questioni di Economia e Finanza (Occasional Papers) 411, Bank of Italy, Economic Research and International Relations Area, November 2017.

Chen, P., Karabarbounis, L., and Neiman, B. The global rise of corporate saving. Journal of Monetary Economics, 89(C):1–19, 2017.

Curcuru, S. E., Thomas, C. P., and Warnock, F. E. Current account sustainability and relative re- liability. International Finance Discussion Papers 947, Board of Governors of the Federal Reserve System, 2008.

Dalgaard, E., Eﬀ, C., and Thomsen, A. Reinvested earnings in the national accounts. Review of Income and Wealth,46(4):401–19,2000.

Eggelte, J., Hillebrand, R., Kooiman, T., and Schotten, G. Getting to the bottom of the Dutch savings surplus. DNB Occasional Studies,Vol.12-6,2014.

Fischer, A. M., Groeger, H., Sauré, P. U., and Yesin, P. Current account adjustment and retained earnings. Technical report, 2018.

Gourinchas, P.-O. and Rey, H. From World Banker to World Venture Capitalist: U.S. External Adjustment and the Exorbitant Privilege. In G7 Current Account Imbalances: Sustainability and Adjustment, NBER Chapters, pages 11–66. National Bureau of Economic Research, Inc, 2007.

Gruber, J. W. and Kamin, S. B. The Corporate Saving Glut and Falloﬀ of Investment Spending in OECD Economies. IMF Economic Review, 64(4):777–799, November 2016.

Guvenen, F., Jr., R. J. M., Rassier, D. G., and Ruhl, K. J. Oﬀshore Proﬁt Shifting and Domestic Productivity Measurement. BEA Working Papers 0139, Bureau of Economic Analysis, March 2017.

IMF. Balance of payments and international investment position manual (bpm6). Technical report, 2009.

IMF. External sector report. Technical report, 2018.

11 Lane, P. R. Risk Exposures in International and Sectoral Balance Sheet Data. World Economics, 16(4):55–76, October 2015.

Lane, P. R. The treatment of global ﬁrms in national accounts. Economic Letters 01/EL/17, Central Bank of Ireland, 2017.

Obstfeld, M. Does the Current Account Still Matter? NBER Working Papers 17877, National Bureau of Economic Research, Inc, March 2012.

Poterba, J. M. Tax Policy and Corporate Saving. Brookings Papers on Economic Activity,18(2): 455–516, 1987.

UN. System of national accounts 2008. Technical report, 2008.

12 Table 1: Foreign Asset Coverage of Sample in 2015

Country Coverage United States 79.2* United Kingdom 79.9 Austria 95.5 Belgium 97.8 Denmark 91.4 France 95.2 Germany 97.3 Italy 99 Luxembourg 86.8 Netherlands 88.9 Norway 88.5 Sweden 93.5 Switzerland 91.8 Canada 91 Japan 91.5** Finland 91.5 Greece 99.7 Iceland 97 Ireland 88.9 Malta 1.39 Portugal 83.4 Spain 96.4 Turkey 93.8 Mexico 98.3 Cyprus 54.1*** Korea, Republic of 86.5 Bulgaria 99.5 China 72.6 Czech Republic 97.1 Slovak Republic 99.8 Estonia 97.8 Latvia 96.6 Hungary 95.8 Lithuania 99.2 Slovenia 93 Poland 70.8 Romania 96.6

Note: Share of foreign assets in partner countries present in the sample over total foreign assets. To be updated with China. * 65.7 without Cayman Islands 14%, Bermuda 3%. **55.9 without Cayman Islands 36 percent, ***Russian for 42%.

13 Table 2: Net Correction in 2015 Positive Net Correction A L Net Norway 7.35 1.52 5.83 Canada 2.68 .7 1.98 Cyprus 1.75 .3 1.45 Sweden 2.95 1.53 1.41 Netherlands 4.1 2.81 1.29 Estonia .54 .25 .29 United States 1.39 1.16 .23 Belgium 1.23 1.09 .13 Lithuania .19 .11 .08 Bulgaria .13 .07 .07 Slovak Republic .15 .1 .05 Slovenia .37 .32 .04 France 1.18 1.15 .03 Negative Net Correction A L Net Ireland 13.86 19.4 -5.54 Japan .69 3.86 -3.17 Korea,Republicof .42 3.07 -2.65 Switzerland 2.47 4.39 -1.92 Greece .05 1.8 -1.74 Spain .44 1.98 -1.54 Poland .07 1.4 -1.32 Germany .7 1.8 -1.1 Finland 2.23 3.05 -.83 Portugal .48 1.05 -.57 Italy .59 1.1 -.51 Hungary .2 .53 -.33 Latvia .12 .44 -.32 Denmark 2.75 3.07 -.32 Romania .02 .3 -.28 United Kingdom 2.47 2.72 -.25 Czech Republic .28 .52 -.24 Austria .87 .97 -.1

Note: In percent of GDP. A is the correction on the income on portfolio equity on the credit side, L the correction on the debit side, and NET the net eﬀect. No ﬁnancial haven adjustment.

14 Table 3: Portfolio Equity Position and Gross Saving (2013-2015)

Positive Correction NPEQ GS Net Norway 116.5 17.8 5.1 Netherlands 20.8 20.0 1.9 Canada 29.6 12.5 1.8 Sweden 25.7 15.0 1.4 Cyprus 5.7 12.1 1.0 Belgium 34.2 16.2 0.3 United States 2.5 12.1 0.3 Estonia 9.2 16.0 0.2 Denmark 12.1 20.7 0.1 France -5.8 10.4 0.1 Slovak Republic 4.0 18.2 0.1 Slovenia 5.8 16.2 0.0 Bulgaria 1.9 23.0 0.0 Lithuania 4.7 19.6 0.0 Negative Correction NPEQ GS Net Japan -5.5 23.4 -2.9 Korea,Republicof -16.7 20.5 -2.9 Ireland -568.5 25.7 -2.6 Switzerland -69.8 14.3 -2.1 Spain -8.3 18.7 -1.6 Greece -2.6 13.8 -1.3 Poland -4.7 18.2 -1.2 Germany 4.4 13.5 -1.1 Finland 21.7 14.4 -0.8 Portugal -1.8 13.1 -0.6 Italy 24.6 10.8 -0.5 Hungary -2.7 17.6 -0.3 Romania -1.3 11.8 -0.3 Czech Republic 2.6 16.5 -0.3 Latvia 4.3 20.9 -0.2 United Kingdom -0.2 10.6 -0.2 Austria 10.3 16.1 -0.1

In percent of GDP. In percent of GDP. Average values from 2012 to 2015. NPEQ is the net portfolio equity position, GS is gross saving of the corporate sector and Net is the net correction of retained earnings on portfolio equity.

15 Table 4: CA Correction (2012-2015)

Positive Correction CA CA* Japan 1.66 -1.27 -2.93 Korea,Republicof 6.63 3.75 -2.88 Ireland 6.67 4.07 -2.60 Switzerland 10.60 8.51 -2.12 Spain 1.24 -0.31 -1.56 Greece -1.32 -2.64 -1.32 Poland -1.37 -2.59 -1.22 Germany 7.77 6.63 -1.14 Finland -1.20 -1.95 -0.76 Portugal 0.59 -0.03 -0.61 Italy 1.53 1.04 -0.49 Hungary 2.90 2.57 -0.33 Romania -1.01 -1.30 -0.29 Czech Republic 0.00 -0.27 -0.26 Latvia -1.68 -1.86 -0.19 United Kingdom -5.53 -5.71 -0.18 Negative Correction CA CA* Norway 10.00 15.10 5.09 Netherlands 9.29 11.20 1.92 Canada -3.03 -1.23 1.79 Sweden 4.97 6.36 1.39 Cyprus -3.65 -2.69 0.96 Belgium -0.47 -0.18 0.29 United States -2.22 -1.96 0.26 Estonia 0.90 1.13 0.23 Denmark 8.73 8.87 0.14 France -0.91 -0.77 0.14

In percent of GDP. Average values from 2012 to 2015. We show corrections larger than 0.1 percent of GDP. Countries with smaller corrections include the Slovak Republic, Slovenia, Bulgaria, Lithuania, and Austria.

16 Table 5: GNI Correction in 2014 Positive Correction GNI GNI* Cyprus -.801 -2.18 -1.37 Switzerland 2.31 1.3 -1 Japan -5.49 -6.07 -.575 Ireland 7.61 7.1 -.508 Hungary 1.85 1.66 -.19 Norway -1.89 -2.07 -.18 Korea,Republicof 7.56 7.39 -.168 Estonia 4.17 4.03 -.144 Canada -2.37 -2.5 -.126 Negative Correction GNI GNI* Greece -1.84 -1.25 .586 Denmark 2.4 2.85 .443 United States 4.48 4.68 .197 Finland 1.36 1.53 .168 Sweden -1.01 -.856 .152 France 1.35 1.48 .136 Slovenia 4.12 4.23 .117 Spain 1.11 1.23 .115 Iceland 10 10.1 .105

Note: Growth rates of Gross National Income (GNI) and the corrected Gross National Income (GNI*) and the diﬀerence between the two. We only report countries where the diﬀerence is larger than 0.1 percentage points.

17 Figure 1: Earnings of NFC: Retained and Distributed 30 25 20 15 1995 2000 2005 2010 2015