## wp18262

## Source details

**Canonical URL:** [wp18262](https://www.imf.org/-/media/files/publications/wp/2018/wp18262.pdf)

## Other formats

- [Markdown version](/-/media/files/publications/wp/2018/wp18262.pdf.md)
- [Structured JSON version](/-/media/files/publications/wp/2018/wp18262.pdf.json)

---

### Corporate net lending and current account positions
- Definition: CA = Saving − Investment = NetLending of the Total Economy = ∑s NetLends, where s ∈ {HH&NPISH, FC, NFC, GG}.
- Sample: dataset starts from 1995 and uses ESA 2010 consistent with SNA 2008.
- Key empirical facts:
  - The corporate sector became a net lender everywhere in the sample except for France and Korea, with timing from the early 2000s.
  - In Korea, corporations remained net borrowers but dramatically decreased net borrowing over the last 15 years.
  - Increase in corporate excess saving is robust across major economies, regardless of current account position.
  - During large upward spikes in a country’s current account balance, most of the increase is driven by the corporate sector (examples: Germany early 2000s, Korea after the Asian crisis, the US in 2009, Italy in 2012).
  - Households and non-profit institutions (HH&NPISH) did not offset the shift to corporate net lending by increasing net borrowing; household net lending remained largely unchanged.
  - Government behavior matters: in the UK and the US, government borrowing more than compensated private saving surplus, producing persistent current account deficits; in Germany, Japan, Korea, and the Netherlands, government did not offset private saving, producing persistent current account surpluses.
- Implication: understanding drivers of corporate saving and net lending is increasingly important for explaining global imbalances.

### The sources of corporate net lending
- Net lending decomposition: NetLend = GrossSaving − CapitalExpenditure.
- Empirical patterns:
  - Gross saving shows a positive trend in most countries; particularly significant and persistent rises in Germany, Japan, Korea, and the Netherlands.
  - Capital expenditure fell everywhere or showed no trend; fall particularly stark in Germany and UK.
- Summary findings:
  - (i) Firms save more of their profits over time but are not financing greater capital expenditures with those savings.
  - (ii) Shift of corporations to net lenders driven by both a rise in gross saving and a fall in capital expenditure.
  - (iii) Germany, Japan, Korea and the Netherlands—countries with largest and most persistent current account surpluses—show the most significant rise in firms’ gross saving.
- Drivers of gross saving increases:
  - GrossSaving = GrossProfit − Dividends paid = NetIncome + Depreciation − Dividends paid.
  - NetIncome = GOS + PI + NetREFDI + NetTransfers − Net Interest paid − IncTaxes.
  - Increases in gross saving supported by increased profits from domestic operations (GOS) and/or from investments abroad (FDI related retained earnings and net PI), as well as by lower interest payments.
  - Dividend payouts increased but by much less than corporate income, yielding saving rates up to 3-7 percent of GDP higher now compared to 20 years ago.
  - Corporate tax payments declined in almost all advanced economies shown; where taxes increased (Germany, Korea) they did so much less than corporate profits.
- GOS drivers:
  - GOS = GVA − CompEmp − ProdTax.
  - Long-run changes in GOS accompanied by higher share of corporate GVA and lower labor income shares.
- Overall:
  - Rise in corporate net lending driven by combination of rising gross saving and falling capital expenditure.
  - In Japan, Korea, Netherlands, Germany this was mostly due to higher gross profitability; only after the global financial crisis did falling dividends contribute meaningfully.
  - Rising operating surpluses supported by lower interest payments and stagnating tax payments (declining effective tax rates).

### The uses of corporate net lending
- Accounting identity for uses of gross saving:
  - GrossSav = CapInvestment + EquityBuybacks + ∆NetFinancialAssets (∆FinAssets) − ∆DebtLiab.
  - Measurement note: aggregate value of equity buybacks not directly observable because net change in equity liabilities combines buybacks with new equity issuances and entrants.
- Empirical evidence:
  - Constructed Net Financial Asset (NFA) = financial assets − debt (where debt = total liabilities − equity).
  - Panel regression: a 1 percent of GDP higher net lending rate within countries is associated with a 1.6 percent higher net financial asset position of the corporate sector on average (coefficient highly statistically significant).
  - Approximation used: NetLend ≈ GrossSav − CapInvestment ≈ ∆NetFinAssets.
- Key patterns:
  - Firms did not systematically use excess saving to pay down debt. Except for Italy and Japan, debt levels increased in all countries.
  - Firms increased both financial assets and debt, but increased asset positions more than debt.
  - Accumulation of cash (cash ratio: cash in percent of total financial assets) is strongly associated with higher corporate net lending positions; cash accumulation identified as the most salient use of corporate net lending.
- Correlations:
  - Tight within-country correlation between corporate net lending (share of GDP) and cash & liquid assets (share of total financial assets).
  - Cash & liquid assets include short-term securities, mutual fund shares and repo.
  - Strongest correlations observed in Germany and the Netherlands.
- Implication: rise in corporate net lending has been mirrored by rise in cash and liquid assets; in the short run, firms are accumulating excess saving to boost cash holdings.
- Transition: micro data analysis used to identify firm characteristics behind rising cash holdings and net lending flows.

### Firm-level analysis: data and stylized facts
- Data: consolidated financial statements of publicly listed companies from Thomson Reuters Worldscope, covering G7+2 countries between 1999 and 2014.
- Rationale:
  - Public firms encompass the largest and most profitable firms and likely drive aggregate saving dynamics.
  - Micro data captures on average a large share of aggregate profits, saving and investment—more than 30 and up to 75 percent of national gross saving.
- Validation:
  - Aggregate NFC net lending rate from national accounts plotted against median net lending rate (percent of sales) from Worldscope shows strong correlation (except Canada).
  - Correlation coefficient ranges from 0.3 (France) to 0.9 (Germany).
- Stylized facts:
  - Cash and size:
    - Median cash and short-term investment over total assets declines with firm size within each period (smaller firms hold higher share as cash).
    - Cash ratios have increased over time for all size classes in most countries, except Japan (largest firms’ cash decreased somewhat) and Italy (smaller and medium-size firms showed some reduction).
    - Time period referenced: 1995−2014.
  - Corporate net lending and industry:
    - Median share of net lending over total sales by firm-size quantiles has increased over time across the firm-size spectrum.
    - Rise in cash accumulation over 1995−2014 common across industries and particularly pronounced for IT-intensive industries.
  - Composition vs incumbent changes:
    - Decomposition shows increase in average cash ratios primarily driven by rising cash ratios within incumbent firms (intensive margin).
    - US: entry of high-tech/R&D firms drove early rise pre-2000s; since early 2000s incumbent firms’ cash holdings continued to rise.
    - Germany, France, Canada, UK: rising cash-as-share-of-assets within incumbent firms.
    - Japan and Italy: pick-up in incumbent firms’ cash holdings around the global financial crisis.
- Empirical framework and sample:
  - Regression framework follows literature on precautionary, transactional, and agency motives.
  - Sample restriction for main regressions: firms observed for at least 18 years (out of the 20-year sample).
  - Industry cash-flow volatility: standard deviation of a firm’s cash-flow to assets over the past ten years; aggregated to two digit SIC industries.
  - Aggregate uncertainty measured by EPU index (Baker et al. (2016)).

### Regression evidence on cash holdings and motives
- Transactional motive:
  - Cash ratios decrease with firm size (log of total book assets).
  - Positive association between cash ratios and industry-level and aggregate cash-flow volatility.
- Precautionary motive:
  - Positive, statistically significant effects of Tobin’s Q, R&D intensity (R&D spending in percent of sales), and age (IPO year) on cash ratios.
  - Positive effect of EPU on cash ratios.
- Agency motive and profits/dividends:
  - Firms that pay dividends have lower cash ratios.
  - Positive effect of profit on cash ratios (consistent with sticky dividend policies).
- Tax-motive proxy:
  - Share of foreign sales has a strong positive effect on consolidated firm cash holding (ambiguous mechanism: internal financing vs profit-shifting and hoarding to avoid repatriation taxes).
- Working capital substitution:
  - Negative correlation between inventory stocks and cash ratios.
- Controls and robustness:
  - Controlling for identified drivers lowers the estimated coefficient on the linear time trend by about half.
  - Results robust to excluding US firms.
  - Handling of missing R&D: treating missing R&D as zero and restricting to non-missing R&D sample yield similar results; R&D effect smaller in non-missing sample as it captures intensive-margin variation.
- ETR (Effective Tax Rate) results:
  - ETR = Sum of all taxes (domestic and foreign) paid / Pre-tax book income (consolidated).
  - Firms with lower ETR have higher cash ratios (pooled regressions); within-firm effect weaker.
  - R&D and foreign sales effects remain after controlling for ETR.
  - Limiting sample to firms with zero foreign assets leaves R&D effect nearly identical; foreign-sales effect strengthens (exporter channel).

### Drivers of corporate net lending (firm-level regressions)
- Definition: Net Lending = Net Income + Depreciation − Dividend paid − Capital Expenditure.
- Pooled regression findings:
  - Larger firms have higher net lending rates.
  - More profitable firms have higher net lending rates.
  - Firms with foreign exposure have higher net lending rates.
  - Firms with larger working capital have higher net lending rates.
  - Firms with higher market valuation have higher net lending rates.
  - Such firms tend to be less leveraged and pay less dividends.
  - Firms with higher R&D intensity have higher net lending positions.
- Within-firm findings:
  - Contemporaneous R&D coefficient becomes insignificant, implying long-run adjustment to innovation spending does not occur immediately.
- Forward-looking R&D:
  - Controlling for one and two-year forward-looking R&D intensity, firms that expect higher R&D spending in 1-2 years increase retained earnings ahead of time.
  - Effect more precise in non-missing R&D sample.
- Profitability heterogeneity:
  - More profitable firms retain more earnings 1-2 years prior to ramping up R&D spending.
  - For a firm in the upper quartile of profitability, an anticipated increase of 10 percent in R&D spending (as a share of sales) is associated with cumulatively 1 percent higher excess saving in the preceding 2 years.
  - For the average firm, the same anticipated 10 percent increase in R&D spending is associated with 0.4 percent higher excess saving.
- Share repurchases:
  - Proxy for buybacks: year-on-year change in number of shares outstanding (percent) when shares decrease, zero otherwise.
  - Firms hoard cash 1-2 years ahead of buybacks and retain more earnings in the year of the buyback.
  - For buyback volumes corresponding to 2 to 3 percent of shares outstanding, estimates predict a net lending rate increase of roughly one tenth of the average rate.
  - Inclusion of buyback proxy does not overturn R&D results.
- Ability to explain trend:
  - Unconditional average public firm has a 3 percent higher excess saving rate compared to mid 1990s (within-firm and cross-sectional average).
  - Conditional trend after controlling for explanatory variables is flat, suggesting identified factors explain much of the surge in corporate net lending.
- Decomposition (1995–2012) using regression column 5 and sample with non-missing R&D:
  - Unconditional average corporate net lending rate increased by about 3.5 percent of total sales over the 16 year period.
  - More than two-third of the overall change can be explained by the increase in size of the average (publicly listed) firm.
  - A significant 40 percent of the total change can be explained by the average increase in R&D intensity.
  - Decline in average effective tax rates also played a significant role.
  - Almost all underlying trends in explanatory variables likely interrelated.
- Broader links:
  - Positive effect of firm size, negative effect of leverage and ETR connect to documented phenomena: industrial concentration, rising market power, rising corporate profits, lower cost of capital, and increased incentives for tax minimization.

### Discussion of mechanisms and macro implications
- Core empirical relationship: consistent positive relationship between innovation intensity and propensity to retain earnings and accumulate cash by large firms in major advanced economies.
- Proposed mechanisms (from literature):
  - Intangible capital and limited collateralizability reduce external debt capacity, prompting higher cash holding (Falato et al. (2013)).
  - Financing profile and capitalizability of intangible capital affect cash holding (Döttling et al. (2016); Ma et al. (2014)).
  - Globalization and technology jointly raise innovation activity and cash holding, particularly among export-oriented firms (Adler et al. (2017); related literature: Autor et al. (2016), Bloom et al. (2016)).
- Structural forces and complementarities:
  - Technology and globalization likely important but not sole drivers; question whether structural forces jointly raise innovation incentives and liquidity demand.
- Market structure and persistence of profits:
  - Profit shares increased together with industrial concentration; size strongly correlated with profit/net lending rates and cash holding.
  - Open questions: drivers of increasing concentration; whether R&D and new technology enable concentration or policy has consolidated incumbents.
- Distributional and external balance considerations:
  - Trend in rising corporate saving is broad-based, structural, and not reversing.
  - Drivers: increased ability of large, publicly listed firms to extract larger profits, expand in size, limit payouts and taxes.
  - Sustained profitability did not spur higher fixed capital investment; firms accumulated liquid assets instead.
  - Firms with largest increases in cash and saving also saw largest gains in profitability, market valuation and R&D spending; they limited dividend payout in favor of buybacks, reduced leverage and reduced effective tax rates.
  - Household sector did not offset increased corporate saving; corporate saving translated into persistent current account surpluses in countries including Germany, Netherlands, Korea and Japan.
  - With capital income concentrated more than labor income, higher corporate saving did not generate proportionately higher household consumption; rise in corporate net lending may have contributed to larger current account surpluses in countries with rising wealth inequality and declining labor income share.
- Policy implications suggested:
  - Ensure competition and prevent inefficient concentration on the corporate side (antitrust, product market regulation).
  - Help alleviate income and wealth inequality on the household side (corporate income taxes, estate and wealth taxes).
  - Note governance and tax management: weak corporate governance correlated with high cash holdings in Japan (Aoyagi and Ganelli (2014)).
- Unresolved research priorities:
  - Identify causal channels linking R&D/intangible capital to cash accumulation: financing constraints, precautionary motives, tax management, strategic payout behavior.
  - Determine role of market structure: whether R&D-driven change inherently raises concentration or policy amplified incumbents.
  - Explore distributional transmission: how rising corporate saving affects household income shares, wealth concentration, and aggregate demand.

*Source: wp18262*

### 3.1  Corporate net lending and current account positions

### 3.1  Corporate net lending and current account positions

### Corporate net lending and sectoral contributions to CA
- Definition: CA = Saving − Investment = NetLending of the Total Economy = ∑s NetLends, where s ∈ {HH&NPISH, FC, NFC, GG}.
- Sample coverage note: dataset starts from 1995 and uses ESA 2010 consistent with SNA 2008.
- Key empirical facts:
  - The corporate sector became a net lender everywhere in the sample except for France and Korea, with timing from the early 2000s.
  - In Korea, corporations remained net borrowers but dramatically decreased net borrowing over the last 15 years.
  - Increase in corporate excess saving is a robust feature across major economies, regardless of their current account position.
  - During large upward spikes in a country’s current account balance, most of the increase is driven by the corporate sector (examples: Germany early 2000s, Korea after the Asian crisis, the US in 2009, Italy in 2012).
  - Households and non-profit institutions (HH&NPISH) did not offset the shift to corporate net lending by increasing net borrowing; household net lending remained largely unchanged.
  - Government behavior matters: in the UK and the US, government borrowing more than compensated private saving surplus, producing persistent current account deficits; in Germany, Japan, Korea, and the Netherlands, government did not offset private saving, producing persistent current account surpluses.
- Implication: understanding drivers of corporate saving and net lending is increasingly important for explaining global imbalances.

### _Source: wp18262 - 3.1  Corporate net lending and current account positions_

### 3.2  The sources of corporate net lending

### Aggregate decomposition of NFC net lending
- Net lending decomposition: NetLend = GrossSaving − CapitalExpenditure.
- Empirical patterns:
  - Gross saving shows a positive trend in most countries; particularly significant and persistent rises in Germany, Japan, Korea, and the Netherlands.
  - Capital expenditure fell everywhere or showed no trend; fall particularly stark in Germany and UK.
- Summary findings:
  - (i) Firms save more of their profits over time but are not financing greater capital expenditures with those savings.
  - (ii) Shift of corporations to net lenders driven by both a rise in gross saving and a fall in capital expenditure.
  - (iii) Germany, Japan, Korea and the Netherlands—countries with largest and most persistent current account surpluses—show the most significant rise in firms’ gross saving.

### Drivers of gross saving increases
- GrossSaving = GrossProfit − Dividends paid = NetIncome + Depreciation − Dividends paid.
- NetIncome = GOS + PI + NetREFDI + NetTransfers − Net Interest paid − IncTaxes.
- Decomposition results (Figure 4 referenced):
  - In countries with increased gross saving over the last 20 years, increases were supported by increased profits from domestic operations (GOS) and/or from investments abroad (FDI related retained earnings and net PI), as well as by lower interest payments.
  - Dividend payouts increased but by much less than corporate income, yielding saving rates up to 3-7 percent of GDP higher now compared to 20 years ago.
  - Corporate tax payments declined in almost all advanced economies shown; where taxes increased (Germany, Korea) they did so much less than corporate profits.
- Country-specific notes:
  - Japan, Korea, Germany, Netherlands show persistent positive trend in firms’ profitability.
  - Distributed income as a share of profits fell particularly post-GFC in Germany.
  - Profitability rose in the US except during the global financial crisis.
  - Japan and Korea in mid-1990s to early 2000s: higher corporate net income was largely driven by declining interest payments due to deleveraging after banking crises.

### GOS drivers: GOS = GVA − CompEmp − ProdTax
- Long-run changes in GOS accompanied by:
  - Higher share of corporate GVA (could reflect higher productivity growth or structural transformation).
  - Lower labor income shares (consistent with documented trends attributable to technology, globalization, or changes in bargaining power).
- Overall summary:
  - Rise in corporate net lending driven by combination of rising gross saving and falling capital expenditure.
  - In Japan, Korea, Netherlands, Germany this was mostly due to higher gross profitability; only after the global financial crisis did falling dividends contribute meaningfully.
  - Rising operating surpluses supported by lower interest payments and stagnating tax payments (declining effective tax rates).

### _Source: wp18262 - 3.1  Corporate net lending and current account positions_

### 3.3  The uses of corporate net lending

### Accounting identity for uses of gross saving
- GrossSav = CapInvestment + EquityBuybacks + ∆NetFinancialAssets (∆FinAssets) − ∆DebtLiab.
- Measurement note: aggregate value of equity buybacks not directly observable because net change in equity liabilities combines buybacks with new equity issuances and entrants.

### Empirical evidence on balance-sheet uses
- Method: construct Net Financial Asset (NFA) = financial assets − debt (where debt = total liabilities − equity).
- Panel regression result: a 1 percent of GDP higher net lending rate within countries is associated with a 1.6 percent higher net financial asset position of the corporate sector on average (coefficient highly statistically significant).
- Approximation used: NetLend ≈ GrossSav − CapInvestment ≈ ∆NetFinAssets.
- Key patterns (Figure 7 referenced):
  - Firms did not systematically use excess saving to pay down debt. Except for Italy and Japan, debt levels increased in all countries.
  - Firms increased both financial assets and debt, but increased asset positions more than debt.
- Regression evidence on uses (Table 1 referenced):
  - Accumulation of cash (cash ratio: cash in percent of total financial assets) is strongly associated with higher corporate net lending positions in pooled and within-country regressions.
  - Cash accumulation identified as the most salient use of corporate net lending.

### Correlation between net lending and cash/ liquid assets
- Tight within-country correlation between corporate net lending (share of GDP) and cash & liquid assets (share of total financial assets) (Figure 9 referenced).
- Cash & liquid assets measure includes short-term securities, mutual fund shares and repo.
- Strongest correlations observed in Germany and the Netherlands.
- Implication: rise in corporate net lending has been mirrored by rise in cash and liquid assets; at least in the short run, firms are accumulating excess saving to boost cash holdings.

### Transition to micro analysis
- Aggregate evidence suggests cash accumulation is a major use of excess saving; micro data analysis is used next to identify firm characteristics behind rising cash holdings and net lending flows.

### _Source: wp18262 - 3.1  Corporate net lending and current account positions_

### 4  Firm-level analysis

### Data and relevance
- Firm-level dataset: consolidated financial statements of publicly listed companies from Thomson Reuters Worldscope, covering G7+2 countries between 1999 and 2014.
- Rationale:
  - Public firms encompass the largest and most profitable firms and likely drive aggregate saving dynamics.
  - Aggregate profits and investment are concentrated among large firms, making the dataset suitable to study corporate saving behavior.
- Coverage statistics (Table 2 referenced):
  - Micro data captures on average a large share of aggregate profits, saving and investment—more than 30 and up to 75 percent of national gross saving.
- Validation:
  - Aggregate NFC net lending rate from national accounts plotted against median net lending rate (percent of sales) from Worldscope shows strong correlation (except Canada).
  - Correlation coefficient ranges from 0.3 (France) to 0.9 (Germany).
- Implication: studying saving and cash management of publicly listed firms offers microeconomic determinants relevant for aggregate net lending trends.

### _Source: wp18262 - 3.1  Corporate net lending and current account positions_

### 4.1  Stylized facts

### 4.1 Stylized facts

### Cash holdings and firm size
- Median holding of cash and short-term investment over total assets plotted by quantiles of firm size shows a mostly downward sloping cash-size schedule within each period, consistent with smaller firms holding a higher share of assets as cash.
- Cash ratios have increased over time for all size classes in most countries, except:
  - Japan: cash holdings of the largest firms decreased somewhat (from very high levels).
  - Italy: smaller and medium-size firms showed some reduction in cash stocks.
- Time period referenced: 1995−2014.

### Corporate net lending and industry patterns
- Median share of net lending over total sales by firm-size quantiles has increased over time across the firm-size spectrum, mirroring the trend in cash holdings.
- The rise in cash accumulation over 1995−2014 is common across industries and is particularly pronounced for IT-intensive industries.

### Composition versus incumbent-firm changes (extensive vs intensive margins)
- Cash-ratio changes can be driven by:
  - Extensive margin: entry and exit (composition effects).
  - Intensive margin: evolution of cash holding among incumbent firms.
- Decomposition formula used (variables in consolidated accounting terms; incumbents superscript I):
  ∆CH_t / A_t = (A^I_t / (A^I_t + A^entr_t) CH^I_t / A^I_t − A^I_{t−1} / (A^I_{t−1}+A^exit_{t−1}) CH^I_{t−1} / A^I_{t−1}) [intensive margin]
  + (A^entr_t / (A^I_t + A^entr_t) CH^entr_t / A^entr_t − A^I_{t−1} / (A^I_{t−1}+A^exit_{t−1}) CH^exit_{t−1} / A^exit_{t−1}) [extensive margin]
- Findings from G-7 (Worldscope coverage sufficient):
  - Increase in average cash ratios in the last decade is primarily driven by rising cash ratios within incumbent firms (intensive margin).
  - US: prior to the 2000s, rise in average cash holding was driven by high-tech/R&D firms entering the market; since early 2000s this entry effect flattened and reversed around mid-2000s, while incumbent firms’ cash holdings continued to rise.
  - Germany, France, Canada, UK: cash-as-share-of-assets rising within incumbent firms.
  - Japan and Italy: pick-up in incumbent firms’ cash holdings around the global financial crisis; in Italy this coincided with a substantial increase in corporate net lending reversing a current account deficit into a surplus.
- Given the prominent role of the intensive margin, the subsequent analysis focuses on drivers of cash holding among and within incumbent firms.

### Empirical framework and sample
- Regression framework follows literature on precautionary, transactional, and agency motives; panel includes firm-level and aggregate-level data across industrial countries.
- Sample restriction for main regressions: firms observed for at least 18 years (out of the 20-year sample).
- Industry cash-flow volatility computed: for each firm-year the standard deviation of a firm’s cash-flow to assets over the past ten years; computation limited to firms with at least three running 10-year observations; firm-level observations averaged over two digit SIC industries.
- Aggregate uncertainty measured by the policy uncertainty (EPU) index from Baker et al. (2016).

### Main regression findings (pooled and within-firm)
- Transaction motive evidence:
  - Cash ratios decrease with firm size (log of total book assets).
  - Positive association between cash ratios and industry-level and aggregate cash-flow volatility.
- Precautionary motive evidence:
  - Positive, statistically significant effects of Tobin’s Q (market-to-book), R&D intensity (R&D spending in percent of sales), and age (IPO year) on cash ratios.
  - Positive effect of EPU on cash ratios.
- Agency motive and profits/dividends:
  - Firms that pay dividends have lower cash ratios.
  - Positive effect of profit on cash ratios, consistent with sticky dividend policies and avoidance of negative signaling.
- Tax-motive proxy:
  - Share of foreign sales has a strong positive effect on consolidated firm cash holding.
  - Noted ambiguity of mechanism: could reflect exporter/multinational need for internal financing or profit-shifting and hoarding to avoid repatriation taxes.
  - WSJ example cited: “S&P 500 companies including Pfizer, Apple Inc. and Microsoft Corp. indefinitely put away a record $2.3 trillion of cumulative foreign earnings by the end of 2015.”
- Working capital substitution:
  - Negative correlation between inventory stocks and cash ratios, consistent with freed-up resources from lower inventories contributing to higher cash ratios.
- Controlling for identified drivers lowers the estimated coefficient on the linear time trend by about half (comparison of regression columns reported in source).

### Net lending regressions
- Replacing dependent variable with net lending ratio (in percent of sales) and keeping explanatory variables shows:
  - Controlling for contemporaneous profits, higher foreign sales exposure and R&D intensity are also associated with higher excess saving, indicating part of their positive effect on cash may operate through corporate saving/retained earnings.

### Within-firm (fixed effects) and first-difference results
- Signs, significance, and magnitudes largely consistent with pooled regressions, indicating part of pooled correlations reflect within-firm changes.
- Notable differences:
  - Effect of higher profits on cash ratio within firms is considerably larger (0.12 to 0.2 in within regressions versus 0.05-0.06 in pooled), supporting dividend stickiness explanation.
  - The positive effect of foreign sales exposure appears driven by cross-sectional variation; within-firm coefficient not statistically significant—supporting tax motive interpretation where variation is mainly across firms.
  - Cash-flow volatility effect is much weaker in first-difference estimation, suggesting industry-specific volatility operates over time rather than contemporaneously.
- Results robust when excluding US firms (non-US sample) — most coefficients not significantly different across subsamples.

### Robustness: R&D missing-data handling
- Main regressions treated missing R&D spending as zero (convention in literature).
- Re-estimation using only firm-year observations with non-missing R&D reduces sample by roughly a third but yields remarkably stable results.
- Estimated positive effect of R&D is quantitatively smaller in the non-missing R&D sample (still strongly significant) because it captures intensive-margin variation in innovation spending rather than extensive-margin entry into being an innovator.

### Tax motive, effective tax rate (ETR), and disentangling channels
- Effective tax rate defined as ETR = Sum of all taxes (domestic and foreign) paid / Pre-tax book income (consolidated).
- Identification assumption: conditional on firm characteristics, ETR variation (particularly between multinational and domestic firms) reflects profit shifting and tax-minimization strategies.
- Pooled and within regressions controlling for ETR (non-missing R&D sample):
  - Firms with lower ETR have higher cash ratios (column 1 of Table 6), consistent with tax motive of cash hoarding.
  - Caveat: consolidated cash ratio dependent variable means this finding may also reflect that lower taxes mechanically translate to higher consolidated cash if tax savings are retained temporarily.
  - Positive effects of R&D spending and foreign sales exposure remain unchanged after controlling for ETR, suggesting these correlations are not primarily driven by tax-minimization strategies.
  - Within-firm effect of lower ETR on cash holding is weaker (column 2), as ETR variation occurs more across firms than within a given firm.
  - Conditional on ETR, increases in a firm’s R&D spending still associate with higher cash holding; increased foreign sales exposure is not systematically associated with higher cash within firms.
- Additional robustness: limiting sample to firms with zero foreign assets (no wholly owned foreign subsidiaries) to reduce channels for profit shifting:
  - Despite sample reduction, coefficient estimates similar to baseline.
  - ETR coefficient only weakly statistically significant as expected.
  - R&D effect nearly identical to baseline; foreign-sales effect strengthened (likely reflecting exporters rather than multinationals).
- Conclusion: while the tax motive matters for some firms (particularly multinationals), evidence indicates other forces (precautionary, transactional, agency, and operational changes such as inventory reductions) also substantially shaped the demand for corporate cash across major advanced economies.

*Source: wp18262 - 4.1  Stylized facts (IMF Working Paper content provided).*

### 4.3  Drivers of corporate net lending

### 4.3  Drivers of corporate net lending

### Definition and empirical approach
- Net Lending = Net Income + Depreciation − Dividend paid − Capital Expenditure
- Strategy: replace dependent variable in cash-holding regressions with firm-level net lending (excess saving/change in retained earnings) and estimate pooled and within-firm regressions to test whether the same drivers of cash holding explain net lending.

### Key empirical findings
- Pooled regression (Table 7, column 1) results:
  - Larger firms have higher net lending rates.
  - More profitable firms have higher net lending rates.
  - Firms with foreign exposure have higher net lending rates.
  - Firms with larger amounts of working capital have higher net lending rates.
  - Firms with higher market valuation have higher net lending rates.
  - Such firms tend to be less leveraged and pay less dividends.
  - Firms with higherR&Dintensity have higher net lending positions.
- Within-firm regression (Table 7, column 2):
  - ContemporaneousR&Dcoefficient becomes insignificant, implying long-run adjustment to innovation spending does not occur immediately.
- Forward-looking R&D effects (Table 7, columns 3-6):
  - Controlling for one and two-year forward-lookingR&Dspending intensity, firms that expect higher R&Dspending in 1-2 years increase retained earnings ahead of time.
  - Effect is more precisely estimated when sample is restricted to firms with non-missingR&Dspending.
- Profitability heterogeneity (Table 7, columns 5-6):
  - Restricting regressions to firms with profit rates above the median and above the 75th quintile shows more profitable firms retain more earnings 1-2 years prior to ramping up R&Dspending.
  - For a firm in the upper quartile of the profitability distribution, an anticipated increase of 10 percent in R&Dspending (as a share of sales) is associated with cumulatively 1 percent higher excess saving (also relative to sales) in the preceding 2 years.
  - For the average firm, the same anticipated 10 percent increase in R&Dspending is associated with 0.4 percent higher excess saving.

### Share repurchases (buybacks) and timing
- Constructed proxy for share repurchases: year-on-year change in number of shares outstanding (in percent) in years that see a decrease in public shares, and zero otherwise.
- Regression results controlling for buyback proxy (Table 8):
  - Firms tend to hoard cash 1-2 years ahead of a buyback event.
  - Firms retain more earnings (increase net lending) in the year of the buyback.
  - For a common volume of buyback (corresponding to 2 to 3 percent of shares outstanding), estimates predict a net lending rate increase of roughly one tenth of the average rate.
  - Inclusion of buyback proxy does not overturn prior results; coefficients onR&Dintensity are largely unchanged.

### Ability of model to explain broad-based trend in rising corporate net lending
- Unconditional firm-level trend (Figure 14, left panel):
  - The average public firm in major advanced economies now has a 3 percent higher excess saving rate compared to the mid 1990’s (within-firm and cross-sectional average).
- Conditional trend after controlling for explanatory variables (full model from Table 7):
  - Remaining time trend is flat, suggesting factors driving higher demand for cash explain much of the surge in corporate net lending.
- Profitability stratification (Figure 14, right panel):
  - More profitable firms saw their retained earnings increase more steeply.
  - Up until the mid 2000’s, rising spending onR&Daccounted for much of the rising propensity for corporate excess saving of the most profitable firms (difference in residual time trend between models with and withoutR&D).
- Since the mid 2000s:
  - Combination of other trends driving profitability and asset expansion appear to contribute most to higher corporate saving and net lending (residual time trends flat and statistically identical between models with and withoutR&D in the right panel of Figure 14).

### Decomposition of long-run change in net lending (1995-2012)
- Method: use regression result in column 5 of Table 8, apply estimated coefficients to long-run change in each explanatory variable over sample period (1995-2012), limiting decomposition to firms with non-missingR&Dspending at the start and end of the sample.
- Aggregate change:
  - Unconditional average corporate net lending rate increased by about 3.5 percent of total sales over the 16 year period.
- Contributions to the 3.5 percent increase:
  - More than two-third of the overall change can be explained by the increase in size of the average (publicly listed) firm.
  - A significant 40 percent of the total change can be explained by the average increase in R&Dintensity.
  - Decline in average effective tax rates also played a significant role in boosting corporate net lending positions.
- Empirical interpretation:
  - Expanding firm size and R&Dspending of the average public firm can account for most of the increase in excess saving over the period.
  - Almost all underlying trends in explanatory variables are likely interrelated.

### Broader interpretation and link to other documented trends
- The positive effect of firm size, and the negative effect of leverage and ETR indicate related phenomena documented in other literature have contributed to increasing corporate saving/net lending:
  - Industrial concentration and rising market power of large firms (Autor et al. (2017), De Loecker and Eeckhout (2017)).
  - Associated rise in corporate profits (Barkai, 2016).
  - Lower cost of capital (Chen et al. (2017), Dao et al. (2017)).
  - Increased incentives and capacity for tax minimization (Zucman, 2014).

*Source: IMF Working Paper — wp18262, section 4.3*

### 4.4  Discussion of mechanisms

### 4.4 Discussion of mechanisms

### Main empirical relationship
- The paper establishes a "consistent positive relationship between the intensity of innovation activity and the propensity to retain earnings and accumulate cash by large firms in major advanced economies."
- The analysis "does not directly point to the channel through which R&D affects cash and saving."

### Proposed mechanisms from the literature
- Intangible capital and limited collateralizability:
  - Falato et al. (2013): "the lack of collateralizability of intangible capital reduces external debt capacity of financially constrained firms, prompting higher cash holding to finance future investment opportunities."
- Financing profile and capitalizability of intangible capital:
  - Döttling et al. (2016) and Ma et al. (2014) discuss how the nature of intangible capital can affect cash holding.
- Globalization and technology as joint drivers:
  - Adler et al. (2017): "globalization, by boosting incentives to engage in risk-enhancing innovation, can jointly raise innovation activity and cash holding particularly among export-oriented firms."
  - Emerging literature on globalization’s impact on innovation cited: Autor et al. (2016), Bloom et al. (2016).

### Structural/market forces and complementarities
- The paper raises the question whether "structural forces" drive both:
  - the incentive to innovate (R&D intensity and intangible capital accumulation), and
  - the demand for liquidity (cash accumulation, retained earnings).
- Technology and globalization are likely important, but "likely are not the only drivers for the corporate saving trends documented in this paper."

### Market structure, concentration, and persistence of profits
- Observation: "If technology and globalization increase corporate profits and retained earnings through benign forces such as trade and intangible capital, then why are these profits not contested by new entrants but instead sustained over such a prolonged period of time?"
- Evidence: "profit shares have increased together with industrial concentration (see Autor et al. (2017), Barkai (2016)), consistent with our result that size is strongly correlated with profit/net lending rates and cash holding."
- Open questions posed:
  - "What are the drivers of increasing concentration?"
  - "Are they related to the innovation channel and if so, how?"
  - "Has R&D and the nature of new technology enabled stronger concentration, or has policy action consolidating incumbent advantage led to inefficient concentration?"
- The authors stress that "Answering these questions will be crucial to understand drivers of the rise in concentration and corporate saving" and their role in "current accounts and external imbalances."

### Connection to broader macro outcomes and inequality
- Aggregate implications emphasized in the conclusion (Section 5 reproduced here):
  - The trend in rising corporate saving is "broad-based" and "appears to be of structural nature and shows no sign of reversing."
  - Drivers identified at the firm level: "increased ability of large, publicly listed firms across advanced economies to extract larger profits and expand in size over time while limiting payouts to shareholders and taxes."
  - Resulting behavior: "sustained gains in profitability have not spurred higher investment in new fixed capital, leaving firms instead with growing stocks of liquid assets on their balance sheets."
- Firm characteristics associated with rising cash and excess saving:
  - "firms that register the strongest increase in both cash and saving also saw the largest gains in profitability, market valuation and R&D spending."
  - These firms "limited dividend payout in favor of share buybacks, reduced leverage and managed to reduce their effective tax rates."
- Distributional and external balance considerations:
  - Household sector did not offset increased corporate saving: the "veil" that separates corporate saving from household consumption "seems to have thickened over time in many advanced economies," translating corporate saving into persistent current account surpluses in countries including Germany, Netherlands, Korea and Japan.
  - "With capital income being much more concentrated among the population than labor income, reinforcing rising wealth inequality worldwide (see Zucman (2014)), it should not be surprising that higher corporate saving and net worth did not give rise to proportionately higher aggregate household consumption."
  - Hence, "the rise in corporate net lending may have contributed to larger current account surpluses in countries where wealth inequality is relatively high, and where the corporate profit share has risen at the expense of a declining labor income share (see Behringer and van Treeck (2018))."

### Policy implications and recommendations highlighted
- To address persistent global imbalances and underlying drivers the paper suggests policies that:
  - "ensure competition and prevent inefficient concentration on the corporate side (antitrust, product market regulation),"
  - and "help alleviate income and wealth inequality on the household side (corporate income taxes, estate and wealth taxes)."
- The authors note governance and tax management strategies as likely contributors and mention that "weak corporate governance has been found to be correlated with high cash holdings in Japan" (Aoyagi and Ganelli (2014)).

### Unresolved issues and research priorities
- Identify and test causal channels linking R&D/intangible capital to cash accumulation:
  - Distinguish between financing constraints (collateralizability), precautionary motives, tax management, and strategic payout behavior (dividends vs. buybacks).
- Determine the role of market structure:
  - Whether R&D-driven technological change inherently raises concentration, or whether policy and regulatory frameworks have amplified incumbent advantage.
- Explore distributional transmission:
  - How rising corporate saving translates into macro outcomes via household income shares, wealth concentration, and aggregate demand.

*Source: wp18262 - 4.4  Discussion of mechanisms (excerpts from the supplied PDF).*

### References

### References

### Cited works and themes
- Extensive bibliography covering corporate cash holdings, corporate saving, taxation, corporate governance, competition, innovation, market power, income distribution, and global imbalances.
- Representative entries (authors, year, and key focus as listed):
  - Adler, K., Ahn, J., and Dao, M. C. (2017). Corporate cash holding and innovation in the era of globalization. IMF Working Paper, forthcoming.
  - Aoyagi, C. and Ganelli, G. (2014). Unstash the cash! corporate governance reform in Japan. IMF Working Paper, 14/140.
  - Armenter, R. and Hnatkovska, V. (2017). Taxes and capital structure: Understanding firms savings. Journal of Monetary Economics, 87:13–33.
  - Autor, D., Dorn, D., Hanson, G. H., Pisano, G., Shu, P., et al. (2016). Foreign competition and domestic innovation: Evidence from US patents. NBER Working Paper No. 22879.
  - Azar, J., Kagy, J.-F., and Schmalz, M. C. (2016). Can changes in the cost of carry explain the dynamics of corporate cash holdings? Review of Financial Studies, 29(8):2194–2240.
  - Bacchetta, P. and Benhima, K. (2015). The demand for liquid assets, corporate saving, and international capital flows. Journal of the European Economic Association, 13(6):1101–1135.
  - Bates, T. W., Kahle, K. M., and Stulz, R. M. (2009). Why do us firms hold so much more cash than they used to? The journal of finance, 64(5):1985–2021.
  - Chen, P., Karabarbounis, L., and Neiman, B. (2017). The global rise of corporate saving. Journal of Monetary Economics.
  - Dao, M. C., Das, M., Koczan, Z., and Lian, W. (2017). Why is labor receiving a smaller share of global income? Theory and empirical evidence. IMF Working Paper No. 17/169.
  - Dittmar, A., Mahrt-Smith, J., and Servaes, H. (2003). International corporate governance and corporate cash holdings. Journal of Financial and Quantitative analysis, 38(1):111–133.
  - Falato, A., Kadyrzhanova, D., and Sim, J. (2013). Rising intangible capital, shrinking debt capacity, and the US corporate savings glut. FEDS No. 2013-67.
  - Foley, C. F., Hartzell, J. C., Titman, S., and Twite, G. (2007). Why do firms hold so much cash? a tax-based explanation. Journal of Financial Economics, 86(3):579–607.
  - Gruber, J. and Kamin, S. B. (2016). The corporate saving glut and falloff of investment spending in oecd economies. IMF Economic Review, 64(4):777–799.
  - Graham, J. R. and Leary, M. T. (2016). The evolution of corporate cash.
  - Johansson, Å., Skeie, Ø. B., Sorbe, S., and Menon, C. (2017). Tax planning by multinational firms. OECD Economics Department Working Paper No. 1355.
  - Poterba, J. M., Hall, R. E., and Hubbard, R. G. (1987). Tax policy and corporate saving. Brookings Papers on Economic Activity, 1987(2):455–515.
  - Zucman, G. (2014). Taxing across borders: Tracking personal wealth and corporate profits. The Journal of Economic Perspectives, 28(4):121–148.

### Figures and tables referenced
- Figure 16: Decomposing change in gross saving rates into sources - year by year.
  - Countries shown: Germany, Japan, Netherlands, Korea.
  - Year axis includes: 1995 1997 1999 2001 2003 2005 2007 2009 2011 2013 2015.
  - Legend items visible in the figure excerpt: NFC_OS_MI_gdpRE, PropIDiv, InterestSS, TransfTax, GS.
  - Source: OECD, National Sources and IMF Staff calculations.
- Table 9: Count of Unique Firm ID by Sector in Each Country.
  - Sectors listed: Agriculture, Mining, Construction, Manufacturing, Transp., Communic., Elect., Wholesale, Retail, Services.
  - Country columns: Canada, France, Germany, Japan, Italy, Korea, Netherlands, UK, USA.
  - Selected counts as provided (preserve exactly):
    - Agriculture: Canada 16, France 13, Germany 6, Japan 1, Italy 11, Korea 1, Netherlands 9, UK 23, USA 37.
    - Mining: Canada 745, France 19, Germany 15, Japan 6, Italy 17, Korea 9, Netherlands 7, UK 166, USA 501.
    - Manufacturing: Canada 482, France 566, Germany 574, Japan 208, Italy 2079, Korea 116, Netherlands 1481, UK 853, USA 3250.
    - Services: Canada 299, France 373, Germany 279, Japan 52, Italy 963, Korea 73, Netherlands 244, UK 850, USA 2198.

### Data sources, variables, and construction notes (Appendices B and C)
- Aggregate income-account sources by country:
  - Canada: Statistics Canada, Current and Capital Account by Sector (multiple series IDs cited in source list).
  - Japan: National Accounts of Japan - Income and Outlay Accounts classified by Institutional Sectors (Japanese Cabinet Office).
  - South Korea: National Account - Production, Income and Capital Accounts - Income Accounts by Institutional Sectors (Bank of Korea).
  - USA: Financial Accounts of the USA (Federal Reserve).
  - France, Germany, Italy, Netherlands, UK: Eurostat, Quarterly Sector Accounts; Non-financial transactions.
- OECD Detailed National Accounts (SNA 2008: Non-financial accounts by sectors, annual (Edition 2015)) used for detailed decomposition of Gross saving (Profits and Distributed Income) and Profit components: Gross Operating Surplus, Interest, Rent, Taxes, Other Transfers. Data expressed in national currency current prices.
- Flow of Funds / Financial stocks sources:
  - Statistics Canada: National Balance Sheet Accounts by Sector.
  - Eurostat: Financial flows and stocks for listed European countries.
  - Federal Reserve: Balance sheet data on nonfinancial corporations (table B103); data downloaded from Federal Reserve datadownload portal.
  - Bank of Korea: ecosystem link.
  - Bank of Japan: stat-search BOJ variable nme_a000_en (selection 11$1).
- Variable construction and component definitions:
  - Liquid Assets (by country-specific components):
    - Canada: Currency and deposits; Short-term debt securities; Investment fund shares.
    - Eurostat countries: Currency and deposits (F2); Short-term debt securities (F31); Investment fund shares (F52).
    - Japan: Currency and deposits (Currency; Transferable deposits; Time and saving deposits; Certificates of deposits; Foreign currency deposits); Short-term debt (Treasury discount bills; Central government securities and FILP bonds; Local government securities; Public corporation securities; Bank debentures; Industrial securities; External securities issued by residents; Commercial paper; Investment trust beneficiary certificates; Trust beneficiary rights; Structured-financing instruments); Short-term loans (Repurchase agreements and securities lending transactions (liquid sub-category of loans); Call loans and bills (liquid sub-category of loans)).
    - Korea: Currency and deposits; Short-term debt (Government bonds; Financial debentures; Commercial papers).
    - USA: Currency and Deposits; Short-term debt: ”Nonfinancial corporate business; debt securities; asset”; Liquid Shares (Money market fund shares; Repo).
  - Accounts Receivables:
    - Eurostat countries: Accounts receivables (F8): Trade credits and advances and other accounts receivables.
    - Japan: Trade credits and foreign trade credits.
    - Korea: Trade credits.
    - USA: Trade receivables.
  - Financial Debt:
    - Constructed by subtracting the value of equity liabilities from the value of Total Financial Liabilities.
- General definitions:
  - Financial flows represent the difference between opening and closing financial balance sheets over the year; financial stocks are the accumulation of financial flows and are the focus of the balance-sheet data used.

*Source: wp18262 - References (PDF).*

---


_Source: https://www.imf.org/-/media/files/publications/wp/2018/wp18262.pdf_