Financial stagnation

Enrico Perotti 18 December 2018




We interpret the current phase of secular stagnation in terms of long-term trends in funding supply and credit demand.

Secular stagnation, a notion from the 1930s recently recalled by Summers (2014), can be defined as a prolonged period of low or no growth when savings exceed investment. Weak economic growth since the crisis has been explained by lack of demand, driven by the need to absorb credit and confidence losses (Lo and Rogoff 2015). However, its persistence suggests a structural supply explanation. Eichengreen (2015) and Justiniano et al. (2015) review the evidence and concludes that technological explanations appear more plausible than temporary demand factors.

Financial stagnation occurs when an excess supply of savings is not reabsorbed by market adjustment, as interest rates fail to adjust the imbalance. Drawing on new conceptual and empirical insights, we explain this process as the combination of a technologically driven drop in loan demand by firms, combined with an inelastic supply of savings. The goal is a long-term perspective on the evolution of financial intermediation that may help to guide public policy in the medium to long term, necessary for sustained stability.

Credit market trends

Over the last two decades, the credit market has experienced a stunning reversal of roles. The productive sector has progressively reduced its leveraged, with net lender flows to the household and government sector in almost all major developed countries (see Figure 1a for the US and Figure 1b for major industrialised economies). The fall in firm net leverage reflects both lower investment spending and rising cash holdings. The trend was only briefly interrupted in the crisis, and has strongly resumed since.

Figure 1 Net financial position of nonfinancial corporate sector

a) US

b) G7

Source: Kamin and Gruber (2015)

The combination of steadily falling rates and a drop in both tangible investment and corporate net borrowing is a puzzle. The steady fall in real rates should have propped up investment and depressed savings, yet their ratio kept falling.

Understanding a global trend across countries requires a broader interpretation. Labour economists have long detected a technologically driven shift in productivity. Increasing automation has reduced the role of physical labour, while boosting the productivity of skilled human capital. The result has been a major shift in wage inequality, with falling labour income and a rising skill premium (Acemoglu and Autor 2011). We argue that a similar trend is altering loan demand and the allocation of credit. Our view is that a technological shift favouring skills and intangible capital has a direct impact on corporate demand for funding (Doettling and Perotti 2017).

Excess savings has been compounded by net capital flows from emerging markets, and redirected to fund existing assets, specifically real estate. Mortgage credit in all OECD countries has expanded vastly since 1980 (Jorda et al. 2016) and shows no signs of abating even after the crisis. Real estate finance has in fact benefited from policy efforts since 2008 to revitalise the economy.

The next sections consider credit demand and credit supply trends, and conclude with implications for long-term policy.

A technologic driver of credit demand

The combination of a drop in both the price and quantity of corporate funding at a time of abundant credit strongly points to a structural decline in firm demand. In recent years falling corporate leverage and rising corporate cash holdings (Bates et al. 2009) have led the nonfinancial corporate sector in OECD countries (with the exception of France and Italy) into financial surplus. Firms increasingly supply net funding to the public and household sector. While the trend is driven by firms with more intangible capital, falling net leverage affects almost all sectors.

Figure 2 US Corporate leverage and cash reserves 

Source: Doettling et al. (2018)

The reality of firms as net lenders overturns received wisdom on the role of financial intermediation in macroeconomic models. Can technological change account for such a striking evolution?

Firms in all sectors have been undergoing a major technological shift to intangible capital (Corrado and Hulten 2010), largely reflecting advances in IT technology. Intangible capital is created by human capital investment. The ratio is steadily rising since 1980 (see Figure 3).1 A rise in the ratio of intangible capital points to a rise in its relative productivity, a redistributive effect, combined with an absolute drop for physical assets (such as traditional CAPEX and physical labour).2  A shift of productivity growth from physical to intangible capital reflects a greater ability to incorporate and elaborate advanced knowledge, just as it shifted productivity from physical to skilled labour (Autor and Acemoglu 2011).

As physical investment has been weak, recent GDP growth in developed countries may be largely attributed to rising productivity of intangible capital.

Figure 3 Evolution of the intangible capital ratio since 1980

Physical investment has significantly lagged the level predicted by measures of growth opportunities (such as Tobin's Q) and profitability (Gutierrez and Philippon 2017, Alexander and Eberly 2016, Crouzet and Eberly 2018).3 Since 1996 firms and sectors raising more external funding have lower Q measures than the average firm (Lee et al. 2016). In other words, firms with the highest market valuation invest less. A driver for low investment in the US is a rising concentration of market power (Gutierrez and Philippon 2017). However, the trend is also visible in Europe, where market concentration has not risen much.

Technological change offers an explanation for both lower investment and corporate funding needs. A shift in the composition of productive assets leads to lower spending on physical investment which is easily funded, while intangible assets are poor collateral to investors. This leads to a more prudent financial policy. Firms with high R&D spending and cash flow volatility hold more cash and have lower net leverage holdings (Almeida and Campello 2007, Bates et al. 2009). However, firms with more intangibles also have a structurally lower need for investment funding. Their lower external finance reflects reduced credit supply but also much lower loan demand.

Doettling and Perotti (2018) formalise a broad explanation. Physical capital requires external financing as it can be acquired by firms. In contrast, the creation of intangible capital relies on talented human capital. Intangible capital is developed through the creative long-term effort of key innovative employees (or independent entrepreneurs). Firms cannot acquire human capital (Hart and Moore, 1994), they hire it and seek to retain it, since employees are free to leave. Thus, ensuring the commitment of human talent requires deferred compensation. As a result, much intangible creation is self-financed as the bulk of its reward is deferred.

Indeed, high intangible firms (HINT) across sectors show lower investment and thus higher free cash flow (Doettling et al. 2017) as Figures 4 and 5 show. High intangible firms also pay out at least as much as firms with more tangible assets. Overall they do not appear more constrained. On the other hand, their lower ex ante costs are counterbalanced by much larger delayed compensation.

Figure 4 Tangible, intangible investment and free cash flow in high and low intangibles firms 

Source: Compustat data.

Figure 5 Investment coverage from cash holdings, HINT and LINT firms 

Source: Doettling et al. (2017)

Since employees are free to leave, optimal deferred compensation must match external options, implying a risky long-term reward. For moral hazard reasons this claim cannot be hedged. The rise in firm cash holdings may be thus seen as a form of insurance to unvested long-term claims, reducing the cost of compensation (Doettling et al. 2018).

This trend further contributes to rising safe asset demand, even from financially unconstrained firms. HINT firms also prefer repurchases to dividend, thus protecting and insuring unvested promises to human capital.

Overall, rather than raising funding, HINT firms are net lenders to the rest of the economy.

Investable value and capital share

As intangible capital becomes more productive, firms must share more value with creative employees (or purchase it from innovators), consistent with anecdotal evidence. Rising innovation rents lead to a rising capital share for innovative entrepreneurs and highly skilled employees (Smith et al. 2017, Eisfeldt and Papanicolaou 2013).

Figure 6 shows how high intangible firms grant to their employees a significantly larger amount of deferred equity, more than a quarter of their overall financing.

Figure 6 Sources of cash inflows and deferred equity 

Source: Doettling et al. (2018)

Innovative firms raise modest funding ahead of their IPO, and even once listed they remain largely self-financed. In fact, recent IPO listings have been used to allow insiders to cash in their vested shares rather than raise much investment funding. Equity markets have seen net equity withdrawals for years (Lazonick 2015). Even the number of listed securities is in rapid decline (Stulz 2018). Stock market capitalisation has fallen as a fraction of GDP, even though stock valuations as price-earning ratios are at historical highs.4

Some intangible value is investable, as it is incorporated in listed equity prices. Yet investors not contributing human capital are diluted by a high share issuance rate to human capital and managing insiders (including private equity).5

Next to a shrinking set of investment opportunities, savers face growing disruption risk as innovating entrants wipe out business models. Garleanu and Panageas (2017) argue that innovation-driven risk is undiversifiable, so disruption boosts demand for safe assets.

A declining supply of investables contributes to the historical decline in real interest rates since the 1980s in a context of inelastic savings (Doettling and Perotti 2017). Structurally lower rates boost the valuation of existing long-term assets, including stocks and housing. This process steadily increases house prices (Knoll et al. 2017),6 with as direct effect a rapid growth in real estate credit (Figure 5). The trend is visible in all OECD countries, with mortgage credit now dwarfing productive lending (Jorda et al. 2016). Such a generalised trend calls for a broad explanation, beyond compounding national factors such as political choices or the country scale of capital inflows.

Figure 7 Commercial and Mortgage Credit in OECD countries 

Source: Jorda et al. (2016)

The next section reviews evidence and new insights on the limited elasticity of supply to interest rates, a key ingredient of financial stagnation.

Inelastic savings and safety demand

A sustained phase of financial stagnation calls for novel thinking, as a situation of excess supply should be self-correcting. The evidence suggests the savings rate is not reduced by low rates, and may even rise (Canzoneri et al. 2007). A positive response may reflect an income effect in the presence of fixed retirement goals.

The puzzle of a credit supply inelastic to interest rates can be understood in the light of new evidence on a structural, segmented demand for safe assets7 (Gorton et al. 2012, Krishnamurthy and Vissing-Jorgensen 2012, 2105). Ahnert and Perotti show how a structural safety preferences induces investors to maintain large holdings of safe financial assets even as their rewards drops. Shocks to bank funding supply driven by factors such as public debt or wealth shifts can thus drive credit volume independently of productivity (see Figure 7).

A segmented demand for safe asset is further suggested by their pricing discontinuity (Krishnamurthy and Vissing-Jorgensen 2012, Diamond 2017). A possible implication is a reduced ability to contain credit expansions by interest rates alone.

Figure 8 Response of bank credit to changes in public supply of safe assets

Source: Kryshnamurthy and Vissing-Jorgensen (2015)

Implications for financial stability and policy

Whether financial stagnation compromises financial stability depends on its persistence. Yet even a prolonged phase of falling interest rates is not leading to a reabsorption of the surplus. An unresponsive supply of bank funding independently of their lending prospects is an alarming prospect, since the private provision of safety is inherently fragile. Only short-term financial debt allows for (near absolute) safety (Brunnermeier and Oehmke 2012, Ahnert and Perotti 2018). Figure 8 shows evidence that bank credit and maturity mismatch rises as public safe debt falls. Maturity shortening and withdrawals ensure a safe outcome for risk intolerant savers, but at the cost of fragility.

Excess savings have been absorbed by real estate financing via an adjustment in the value of the housing stock rather than its supply, as housing in desirable locations is limited.

Mortgage demand is also boosted in general equilibrium by the redistributive effects of technological change. As unskilled labour income does not rise in line with GDP (Acemoglu et al. 2011), over time mortgage debt demand rises. A higher ratio of mortgage debt to income induces rising default risk over time, even under constant income risk (Doettling and Perotti 2017). Additional factors contributing to excess savings are demographic change (higher longevity) and capital flows seeking safe assets. Both factors boost the stock of savings and bank funding, though they could not account for weak investment at a time of falling rates.

Ultimately, a sustained phase of financial stagnation creates large risk. Evidence from historical expansions shows how growth in credit in excess of productive opportunities leads to instable lending booms (Gorton and Ordonez 2017). Credit supply-driven booms stimulate local demand rather than supporting productivity, and amplify the business cycle (Krishnamurthy and Vissing 2015, Krishnamurthy and Muir 2016, Mian et al 2017).

The analysis has implications for macroprudential policy, as countercyclical intervention requires a proper measure of excess credit relative to trend. Structural changes question the wisdom of extrapolating credit trends, as this fails to separate real credit demand and supply effects. A technological interpretation for financial stagnation also raises questions on the ability of monetary authorities to stimulate investment by lower rates (or to contain credit by higher rates8). If investment is unresponsive to rates, low rates are likely to only boost consumption and asset valuation, so that the productive response does not justify increased leverage or more fiscal spending.


The combination of a falling demand for credit against an inelastic demand for safety explains several observed trends on rates, credit allocation and investment. Understanding the structural, non-conjunctural causes of the current phase of financial stagnation is critical to inform a long-term policy perspective on economic and financial stability goals.

Financial stagnation implies that abundant funding is not invested but stored in existing assets. As firm demand for funding declines thanks to technological change, an inelastic demand for safe assets implies a steady funding for banks even as their lending opportunities decline. A result is higher valuation for long term assets, a rising mortgage share in bank lending, higher household leverage and more maturity transformation.

Financial stagnation has structural implications. It implies less effective monetary policy (as financing is not scarce, and investment less rate-sensitive), a changing allocation of credit and rising maturity transformation. For policymakers, the key long-term challenge is to move from a focus on cyclical conditions to a sustainable approach focused on long term stability, in a critical phase of declining central bank balance sheets.


Acemoglu, D and D Autor (2011), “Skills, tasks and technologies: Implications for employment and earnings”, Handbook of labor economics 4: 1043-1171.

Alexander, L, and J Eberly (2018), “Investment hollowing out”, IMF Economic Review 66, 5-30.

Almeida, H and M Campello (2007), “Financial constraints, asset tangibility, and corporate investment”, Review of Financial Studies 20, 1429-1460.

Autor, D (2014), “Skills, education, and the rise of earnings inequality among the other 99 percent", Science 344, 843-851.

Ahnert, T and E Perotti (2018), “Seeking Safety”, University of Amsterdam mimeo.

Bates, T W, K M Kahle, and R M Stulz (2009), “Why do U.S. firms hold so much more cash than they used to?”, Journal of Finance 64, 1985-2021.

Berk, J, R Stanton, and J Zechner (2010), “Human capital, bankruptcy and capital structure”, Journal of Finance 65, 891-926.

Borio, C (2014), “The financial cycle and macroeconomics: What have we learnt?”, Journal of Banking & Finance 45, 182—98.

Caballero, R, E Farhi and P-O Gourinchas (2017), “The Safe Assets Shortage Conundrum”, Journal of Economic Perspectives (Summer 2017): 29-46

Canzoneri, M, R Cumby and B Diba, (2007), "Euler equations and money market interest rates: A challenge for monetary policy models" Journal of Monetary Economics 54(7), 1863- 1881.

Corrado, C A, and C R Hulten (2010), “How do you measure a technological revolution?”, American Economic Review 100, 99-104.

Crouzet, N, and J Eberly (2018), “Understanding weak capital investment: The role of market concentration and intangibles”.

Dell'Ariccia, G, D Kadyrzhanova, C Minoiu and L Ratnovski (2017), “Bank Lending in the Knowledge Economy”, IMF working paper 17/234, November

Diamond, W (2017), “Safety Transformation and the Structure of the Financial System”, Mimeo, Harvard University

Doettling, R, T Ladika, and E Perotti (2017), “The (self-)funding of intangibles”, CEPR Discussion paper no 12618

Doettling, R, T Ladika, and E Perotti (2018), “The Creation of Intangible Capital”, University of Amsterdam mimeo

Doettling, R and E Perotti (2017), “Secular Trends and Technological Progress”, CEPR Discussion paper no 12519, December

Drechsler, I, P Schnabl and A Savov (2017), “The Deposits Channel of Monetary Policy”, Quarterly Journal of Economics 132 (4), 1819-1876

Eichengreen, B (2015), “Secular Stagnation: The Long View”, American Economic Review, 105 (5):66-70

Eisfeldt, A L, and D Papanikolaou (2013), “Organization capital and the cross-section of expected returns”, The Journal of Finance 68, 1365-1406.

Falato, A, D Kadyrzhanova, J Sim, and R Steri (2018), “Rising intangible capital, shrinking debt capacity, and the US corporate savings glut”, Finance and Economics Discussion Series Working Paper 2013-67.

Garlenau, N and S Panageas (2017), “Finance in a time of disruptive growth”, U Berkeley mimeo.

Gorton, G, S Lewellen, and A Metrick (2012), “The Safe-Asset Share”, American Economic Review P&P 102 (3), 101-06.

Gorton, G and G Ordonez (2016), “Good booms, bad booms”, NBER WP 22008

Gruber, J and S Kamin (2015), “The Corporate Saving Glut in the Aftermath of the Global Financial Crisis”, International Finance Discussion Papers 1150, Federal Reserve Board

Hart, O and J Moore (1994), “A Theory of Debt Based on the Inalienability of Human Capital", The Quarterly Journal of Economics 109(4):841-79.

Jorda, O, M Schularick, and A M Taylor (2016), “The great mortgaging: housing finance, crises and business cycles”, Economic Policy, 31(85):107-152.

Justiniano, A, G E Primiceri, and A Tambalotti (2015), “Credit supply and the housing boom”, National Bureau of Economic Research.

Krishnamurthy, A and A Vissing-Jorgensen (2012), “The Aggregate Demand for Treasury Debt”, Journal of Political Economy 120 (2), 233–67.

Krishnamurthy, A and A Vissing-Jorgensen (2015), "The impact of treasury supply on financial sector lending and stability", Journal of Financial Economics 118(3), 571–600.

Krishnamurthy, A and T Muir (2017), "How credit cycles across a financial crisis", Yale U mimeo

Lazonick, W (2015), “Stock buybacks: From retain-and-reinvest to downsize-and-distribute”, Center for Effective Public Management, Brookings Institution, April, 15.

Lee, D W, H-H Shin, and R Stulz (2016), “Why does capital no longer flow more to the industries with the best growth opportunities?”, Ohio State University mimeo

Oyer, P (2004), “Why do firms use incentives that have no incentive effects?”, The Journal of Finance 59, 1619-1650.

Peters, R H, and L A Taylor (2016), “Intangible capital and the investment-q relation,” Journal of Financial Economics.

Philippon, T, and G Gutierrez (2016), “Investment-less growth: An empirical investigation”, New York University working paper.

Philippon, T and G Gutierrez (2017), “Declining competition and investment in the US”, New York University working paper.

Rampini, A (2018), “Financing durable assets”, American Economic Review

Richter, B, M Schularick and P Wachtel (2017) “When to Lean Against the Wind ?”, CEPR DP 12188

Smith, M, D Yagan, O Zidar and E Zwick (2017), “Capitalists in the twenty-first century”, mimeo Chicago Booth

Stulz, R (2018), “The Shrinking Universe of Public Firms: Facts, Causes, and Consequences”, NBER reporter.

Stein, J (2012), “Monetary Policy as Financial-Stability Regulation.” Quarterly Journal of Economics 127 (1): 57-95

Summers, L H (2014), “US economic prospects: Secular stagnation, hysteresis, and the zero lower bound”, Business Economics, 49(2):65-73.


1 The intangible capital measure capitalizes investment into R&D, goodwill and organizational capital (Peters 2017), while the BEA data measures intellectual property rights.

2 A drop in the productivity of physical factors at the technological frontier reflects also relocation of physical production. Diffusion of traditional technology has contributed to rising physical investment and labour wages in emerging countries.

3 Including intangibles improves the fit of Tobin’s Q investment models (Luke Taylor 2016).

4 This interpretation subsumes other causes for secular stagnation, such as a fall in the productivity- adjusted cost of physical capital driven by higher intangible value incorporated in production.

5 This fact may help to explain the findings of Lee et al. (2016), since HINT firms typically have high market to book valuations.

6 Cross section evidence suggests that as US firms increase their intangible assets, local banks shift to more mortgage lending (Dell’Ariccia et al. 2017) while local house prices rise more.

7 Default-free claims include public debt and short-term financial debt, mainly provided by banks.

8 Banks lose some funding when rates rise, as they do not fully adjust deposit rates (Drechsler et al. 2017). However, reallocation of savings to competitively priced market claims (e.g. money market funds) may simply shift bank funding to less stable sources.



Topics:  Financial markets

Tags:  financial stagnation, savings, loan demand, financial intermediation

Professor of International Finance, Amsterdam Business School; EEA Fellow; Research Fellow, CEPR

CEPR Policy Research