The Growth and Investment Puzzle

Why it matters and where it comes from

Press conference of the OECD's Interim Economic
The Growth and Investment Puzzle : Why it matters and where it comes from - Editorial Staff


Dr. Catherine L. Mann has been OECD Chief Economist, G20 Finance Deputy, and Head of the Economics Department since October 2014. Dr. Mann is responsible for advancing the Strategic Orientations of the OECD, including the OECD’s Economic Outlook, country-based economic surveys and the Going for Growth report. She was most recently a Professor of Global Finance at the International Business School at Brandeis University and Director of the Rosenberg Institute of International Finance.

Why is that?

A couple of things. One is we’ve had an increase in income inequality in the United States. That means lower-income consumers have not gotten the benefits of growth. The import intensity of those consumers is quite high and they are not in a position to consume much more. The second thing is that if we go back and look at the period of time when we had tremendous increases in consumer goods imports, it was when we had a housing boom. People had to furnish two houses or furnish bigger houses and they’re not doing that anymore. On the other hand, people do have more money in their pockets, even lower-income consumers to some extent. This is because of lower gas prices, more benefits and more employment: even if your wages have not gone up, at least you have a job.

So there is a capacity to consume but where is it going?

It seems to be going much more for the types of things that you had completely given up: you had to give up going out to restaurants. You gave up your vacations and stayed home, remember the term “staycations”? So now with that additional disposable income, people are spending it on services: food away from home, restaurants, and entertainment. 

Looking through the lens of trade, that is a collection of both short-term and medium-term trends that I think are very important for the global economy. If we think about five or ten years from now the most important thing is will China become a closed economy? Will it become like the United States? The United States is basically a closed economy, if we look at (X+M)/GDP. If you take Europe as a whole they are not really a closed economy and China has not been one, but the question is will it be?

What effect would that have?

For probably two or three decades the United States always talked about the need for other countries to pursue domestic demand-led growth. You don’t need to depend on a global cycle with that type of growth and it would be the source of more stable development. Well that was intercepted by China and the fragmentation of the global value chain. If China becomes more like the US, more like a closed economy, then the source of the global value chain-generated trade and growth is going to go away. And so we will be back once again to talking about the need for domestic demand led growth, but China will have transitioned to a closed economy. So then you will have some countries with no demand-source of a trade-oriented pathway to GDP growth and I see that as something that policymakers should really be thinking about as they pursue their agendas.

What time span is this going to happen for China?

That is entirely dependent on whether or not they can navigate towards domestic production and domestic consumption: specifically, domestic production of intermediates and domestic consumption as opposed to exports for GDP growth. Going back to the mid-90s, there were two major times when the Chinese said that they were going to rebalance for consumption based growth, away from commodity-intensive investment and the export led-growth model. Both times they tried and they have not yet succeeded and frankly, right now when I look at the stimulus program that they have under way it’s the same old: it’s credit heavy, so credit extensions to heavy industry, to real estate, to commodity intensive production. Those are not the foundation for consumption based growth. They could have spent money for more education, for more of a social safety net, for health insurance, all which would be ways you could reduce the household savings rate. Right now there’s always precautionary savings for those purposes, so what you need to get consumption based growth going is to have a type of support that brings down the precautionary household savings rates. The way to do that is to have insurance programs, whether it be education, health, or old age. So even though it could be the approach of the fiscal program, it does not appear to be where the focus is. 

The OECD has linked credit overexpansion and the current structure of the financial market to slowing growth. How have these led to a slowdown and what should the policy response be? 

On the OECD website is a section called Finance, Growth and Inequality. The issue here is that there is a U-shape to financial depth, which can be measured by bank credit, bond, and equity capitalizations. At very low ratios of financial depth expanding all three types of credit are the foundation for faster growth. A lot of emerging markets have very thin capital markets and they have very little bank credit, so for those economies the argument to restrict financial markets is wrong. You need more financial depth in order to do the things that credit does in a system. However, in a number of the OECD economies including the U.S. and Europe (but not for example Korea, Mexico or Israel), we want to decompose different kinds of finance -- bank credit depth, bond credit depth and equity market capitalization. For bank credit depth what we find is that when bank credit as a share of GDP exceeds something in the neighborhood of 80% there is a negative correlation between increases in bank credit depth and growth rates. So this is the research finding about over-expansion of credit. 

How does this happen?

We first want to make a distinction of who is getting the credit. We can look at bank credit overall and then ask is it going to households? Is it going to firms? And what we find is that this over expansion negative correlation is much more extensive for household credit than for credit that is extended to firms. 

So why is that? Think about what you are supposed to do when you get credit: you are supposed to get a loan and then acquire an investment that will earn at a rate of return that is sufficient to repay the obligation. That is how it is supposed to work. And now consider that the only way to repay a housing loan is through housing price appreciation. It’s not that the house is going to change, the only reason your house gets a rate of return is because prices go up. But if prices do not go up enough or if they go down then all of the sudden you have the negative correlation. The house price is not enough to repay the obligation and then you have a bankruptcy and then you can’t spend: that is the channel through which we get this particularly strong relationship between too much credit and negative implications for growth. 

The story for firm credit is not as strong because at least some of the firm credit is going into investments that are going to repay the obligations. However, we know that firms borrow in order to buy back stock, they borrow to do mergers and acquisitions, and they do all sorts of borrowing that actually doesn’t involve creating an asset that is going to earn a rate of return over the lifetime and repay the obligation. So there are a number of things that firms do with credit that don’t have good rates of return and that’s also where the negative from credit overexpansion comes from.

Are there other factors influencing growth?

For equity market cap we find that there is throughout the sample a positive correlation between equity market depth and growth. Perhaps this results because there’s a closer link between equity valuations and wealth or the marginal propensity to consume wealth is large enough so that you get the positive growth, but I also think that a lot of it is driven by the fact that equity market depth in Europe is very low. When we look at the whole sample, there’s very low equity market depth in parts of the OECD like Korea, and Mexico as well. We know from other research on productivity that stock market capitalization and availability of risk capital is positively associated with productivity growth, because availability of risk capital allows new firms with new ideas to start up and therefore push productivity. It’s the new firms that really push the productivity frontier out.

Looking at capital markets in Europe and limits on bank credit for some countries that have excessive credit to GDP, I think what you’re seeing there is happening in the context of limiting bank expansion. The new financial stability boards and total loss absorbing capital (TLAC) or Dodd-Frank in the United States are mechanisms to effectively limit the ability of banks to extend excessive credit. Then you have loan to value ratios and other similar mechanisms, which are macroprudential measures to limit the extent to which credit goes into housing.

You have spoken before about the lack of investment compounding slowing growth in OECD countries. How do you see this as a problem and what policies do you think are needed to make it more attractive again?

We’ve been doing a lot of work on investment, so let’s start off with some of the facts: investment in the Euro area is 15% below in real terms where it was in 2007. In the United States it’s only 5% more and at this time in a standard business cycle, if we compare it to past decades, investment would be 20-30% higher. In normal cycles at this point in the OECD we’d be 20% higher. In fact, in the OECD as a whole we’re only 5% higher. It’s very disparate because commodity producers like Australia and Canada are about 10% above, then you’ve got the Euro area that is 15% below  in real terms and Japan’s 5% below. So it’s very disparate, but the bottom line is that it’s sluggish.

Why is this a problem?

Investment is a fundamental component of demand through job creation, so it has that short term element to it. But, I’m at least as worried about the lack of investment from the standpoint of potential output. Potential output (YP) is a function of innovation A, the capital stock K, and labor L.  (YP)=AF(K,L).  If you have no investment you have significant deterioration in capital stock and that’s part of the reason that we see a very significant slowdown in potential growth. It’s also the case that you have this labor market interaction as well, because if you don’t have capital you don’t have good labor productivity. The third element is of course that investment is a key component of A, of the Total Factor Productivity (TFP).

How do you get an increase of TFP?

Well you have to invest in order to take advantage of the innovations that are available. Investment is really a core of potential output growth and the slowdown in investment is a core element of a slowdown in potential output growth. And again, why do we care about that? For me, potential output is a measure of the economy’s capacity to make good on promises. It’s the wherewithal with which you make good on your promises to young people that they’re going to have a better life, to older people that you will pay their pensions and to investors that you will  repay bonds.

So when you see a declining rate of growth of potential output then you know that someone’s not going to get their promises kept. And right now, in Europe I’d say it’s mostly young people. When we look at the unemployment rates of young people it’s dramatically high. And that’s a problem.

What can be done?

One piece of the puzzle is the importance of the neighborhood effect. In our traditional investment accelerator model, we have investment as a function of rate of return on capital (or interest rates) and then GDP growth or consumption. If your demand is high and your cost of capital is low then you’ll have a lot of investment. We know that the interest rate part isn’t very sensitive right now, since everybody’s rates are basically zero. However, we do know that in Europe even though interest rates are really low from the central bank side the actual spread is pretty high. So credit constraints still play a role in affecting investment in Europe, but think about Europe as a whole. You realize that the investment accelerator is not about French demand for French investment, it’s about Europe’s demand for Europe’s investment. There’s really a sense of collective, but if you’ve got one country growing and their next door neighbor is not then all of a sudden firms don’t get the signals to invest. It’s signaling from a collective group.

But is a firm looking at the whole group or just at their individual picture?

Firms are looking at everything, at the world or the local neighborhood. So if the neighborhood is not growing strongly then it doesn’t really matter what’s happening at home. For example, Germany is a country with strong demand and low unemployment but no investment. Why? Because the neighborhood’s not growing. The neighborhood is what is important.

What does this mean for policy?

There’s a way to get the neighborhood to grow and that is having a collective public investment program. One country doing it by themselves is not going to make the difference because there’s too much leakage across borders. But if you can do it collectively then you gain.


 You might argue that governments can’t spend any more money, that the stability and growth pact constrains them. But actually, no. We have a simulation in our Economic Outlook that says half a percentage point of GDP public investment program undertaken collectively by all of the OECD economies yields a .7 increase in OECD growth, .5 increase in the Euro area, and .4 globally even though not everybody is undertaking the policy action. You get GDP growth and because of that growth, debt to GDP falls. And there are two things in the stability and growth pact- there’s the deficit target and also the debt target. What the simulation is basically arguing is that the debt target is the one that you ought to care about, since debt is what you have to pay back. You pay debt back with GDP, so that’s what you need to care about. Giving some leeway on the fiscal target in the SGP wins you the benefits of a reduced debt to GDP ratio.


Now what are the assumptions? First, it’s collective. Number two, it has to be in high value, high multiplier projects. The third assumption is critical and also an essential ingredient of the Juncker plan, which has a lot of similarities to what is the simulation. The key ingredient is that there has to be a focus on high externality projects where you get a lot of spillovers, like network industries, telecoms, energy, and transport. And again, it has to be undertaken in an environment where there have been structural policies being undertaken collectively. For collective investment within a region you have to have collective improvements in structural policies—in regulations. The simulation is a call to action on the part of policymakers. It’s not just about Europe, but I’m focusing on Europe because they have the worst investment performance. So it’s a call to action for European policymakers to think collectively, to act to support the Single Market through regulatory harmonization, and to spend.

If the collective structural policy is a key assumption what do you think is the correct timeline for this call to action?


But at the national level or collective?

It’s got to be collective. It’s so important to emphasize the collective nature of this. It’s why I come back to the Single Market. Revive the notion of the Single Market. You don’t hear about that as much anymore, but then you see what’s happening to the Schengen zone with the re-entrenchment of borders. Policymakers need to act collectively. They’re still individual countries, but the benefits that would be obtained from this collective spending and collective approach to regulatory strategies are tremendously important.


We have some excellent research that looks exactly at regulatory harmonization of European network industries and it finds that the gains to investment from cross-border investment could go up by 25%. We’re not talking about regulations going to the lowest common denominator, it’s regulatory harmonization. What stands in the way are an awful lot of things where harmonization can get you gains. For example, I cross national borders and get a message about different rates on my phone, there’s the lack of an energy grid, there are different transportation standards, and so on.  We’re not talking about deregulating markets down to no regulation, it’s the disparity and the dispersion in regulations that are particularly dramatic and negative for investment within Europe.

Finally, inequality. It’s increasingly being talked about, so what are your thoughts on the issue?

There are many different directions to go with this topic. First, we have to recognize that some of the rise in income inequality, in particular since the financial crisis, has to do with the fact that there hasn’t been a significant enough recovery to bring people back to work. One cause of widening income inequality is simply that growth has been too slow to bring people back into the labor force and back into employment. Another piece of the puzzle is tax and transfer systems, which differ between countries. We can calculate for all countries what the 90-10 income inequality between the 90th percentile and 10th percentile is for all market incomes. This is what the market delivers in terms of wages, and then we can look at how taxes and transfer systems reduce that inequality. There’s a time series of what these look like and what we observe is that the tax and transfer systems have done less to mitigate market income inequality over time.


What we don’t know is have we chosen to not deploy taxes and transfer systems as much, since it’s different depending on the country and therefore it’s a research problem? I know in the U.S. the answer to that is yes- we’ve reduced capital gains taxes and reduced the top brackets of income taxes. So the answer for the U.S. is we have not deployed tax and transfer systems to mitigate market income inequality as much as in earlier times, but we don’t know the answer for other countries. It is also perhaps the case that the tax and transfer systems that countries had in place 10 years ago don’t work as well now. Is it because the bottom decile is more disadvantaged that before so you need different kinds of tax and transfer programs? Is it because of skill-based technological change? Is it that the tax and transfer systems we had before really don’t work as well? Is it that populations are heterogeneous in other ways where the tax and transfer systems we used before are less effective? These are different reasons for tax and transfer systems to not mitigate market inequality, but not deploying is very different from not working as well. They are observationally equivalent in the data, which is a research problem that I think is very serious.


Going back to finance, the third piece of the puzzle is extreme inequality in the financial sector. For example if you’re a janitor in a financial firm you earn 15% more than the average janitor in other sectors, but if you’re a CEO you earn 40% more compared to somebody who’s exactly the same in terms of characteristics. This is from individual level data—a huge data set --looking at Europe. When the financial sector expands as a share of activity in the economy and it has this characteristic of being really unequal in terms of delivering salaries then you have that adding to inequality.


Then we have a whole set of other reasons that have to do with mismatching of skills. People are captured in low-productivity firms and although they’ve got more skills than are needed, they’re not going to get paid for those skills. Why don’t they leave and get a better job? That goes back to not having enough demand in the system to warrant leaving a job just because it’s boring. You’re going to stick with the job, leading to another source of inequality.