.png)
Dr. Filipa Sá is a distinguished economist, who has undertaken research in a wide range of areas, from applied economics to health economics. After completing a PhD at MIT in 2008, Dr. Sá worked as an economist at the Bank of England for two and a half years during the Global Financial Crisis. Dr. Sá’s research has been published in The Economic Journal, the Journal of the European Economic Association, the Journal of Money, Credit and Banking, among others.
Given that a substantial amount of your research has been on housing, do you think that was the main cause of the 2008 crisis?
My research looks at the role of capital flows in driving up house prices in the run up to the Global Financial Crisis. The Global Financial Crisis came in a context where the macroeconomy was doing well in most countries. This period was known as the Great Moderation. This means that most countries had low and stable inflation and reasonable growth rates. Therefore, I do not think anyone was expecting the Global Financial Crisis to happen because this was seen as a generally benign period, when most economies were doing well. However, there were some vulnerabilities building up, especially in the housing market. Credit from banks was expanding very rapidly to households, some of which maybe could not afford their loans. Thereafter, when prices started decreasing, and banks started tightening credit, many people could not afford their mortgage payments and, as a result, had to default on their mortgages.
Expanding on the scope of her research, Dr Sá added:
In my research, I look at capital flows and the role they could have played in fuelling these housing bubbles or housing booms in many countries. We used data for a panel of OECD countries and identify capital inflows shocks. We do this using an econometric framework called Vector Autoregressive Model (VAR) and then use sign restrictions to see how macroeconomic variables should respond if there is a capital inflow shock.
Read more about VAR here
Using the US as an example, an increase in capital inflows would result in interest rates falling and the dollar would appreciate. This would result in the current account going into deficit. So, we have an open economy model to explain how the macro variables should respond to different shocks: capital inflow shocks, monetary policy shocks, as well as other types of shocks. This research comes in the context of a literature that became prominent in the mid-2000s about global imbalances – economies like China and oil exporting countries were saving and buying assets such as government bonds in the US, because their financial markets were not so developed or liquid. The inflow of money depresses interest rates in the US.
In the run-up to the Global Financial Crisis, interest rates were very low, not only because central banks were actively keeping them low, but also because you had this inflow of foreign capital, which was further depressing rates. All in all, this meant it was cheaper to borrow, and therefore cheaper to buy a house. So, in this paper, we find that countries where you see a larger inflow of capital have a higher increase in house prices. It is, however, important to note that we do not look at the subsequent decline in house prices during the crisis, but we are just trying to show this link between capital inflows and house prices. Research by Mian and Sufi, authors of “House of Debt”, shows that the increase in mortgage credit in the run-up to the crisis was one of the factors that caused the Global Financial Crisis.
We understand that the global macroeconomy was stable and as a result, for many it was difficult to predict such a large-scale shock. If we were to have known the crisis was coming, could we have avoided it and if so, how?
One of the things we do in the paper is look at the role of financial market development, in particular mortgage-backed securitisation, in amplifying the propagation of capital inflow shocks to house prices. We found that in countries where you can borrow against the value of your house for consumption expenses i.e. mortgage equity withdrawal, or in countries where mortgage-backed securities were allowed in a widespread manner, there is a higher effect on house prices from the same shock to capital inflows. Therefore, from a policy perspective, if we could have seen that there was a vulnerability, policymakers could have restricted the use of mortgage-backed securities and mortgage equity withdrawal. They could also have imposed more affordability checks on bank lending, which were implemented after the crisis. Bringing such tools that were used after the crisis into place before it occurred could have mitigated the effect of the shocks. The same applies to macroprudential policy (policies implemented to preserve financial stability and mitigate vulnerabilities and absorb the impact of shocks - read on to learn more). It could have been used before the crisis had we known there was a vulnerability.
Focusing more closely on your work on capital inflows and the US housing boom, can you explain in simple terms what capital inflows are and the measures and tools used to find the impact on the US housing market?
In essence, capital inflows are the movement of money from one country to other countries. When investors invest in a foreign country, they may be buying anything from government bonds to company shares to a factory. In our research, we focus particularly on investment in government bonds because that is what global imbalances were mostly about.
So, you have China and other countries that are saving more than they are investing and their excess saving is then directed to the US and other countries where bonds are seen as a very safe and liquid asset. We then collected data on interest rates, short and long rates, the current account, the exchange rate, and the dependent variables (residential investment, private credit, and house prices) for a panel of OECD countries. We then use theories from an open-economy DSGE model where we can see what effects a capital inflow and monetary policy shock would have. We then estimate a VAR model and use sign restrictions to identify capital inflows shocks. After this, we plot impulse responses to see the impact on all the variables in the model, for capital inflow shocks and other types of shocks. You have time on the horizontal axis and then the variable, for example house prices, on the vertical axis, allowing us to see what they do on impact and what they do over time. This also allowed us to see the dynamics of the effect, or the time taken for shock to have a full impact. We also do this to study the effect of the mortgage market’s structure and mortgage-backed securities. To do this, we introduce interaction terms in the model so the interaction terms will give you the amplification or dampening of the effect depending on whether the country used mortgage-backed securities and mortgage equity withdrawal or not. You can have two sets of impulse responses: the first for countries that do not allow the use of mortgage-backed securities and the second for countries that allow this use. We can see if the difference in these impulse responses is statistically significant through the coefficient on the interaction term. This provides us with confidence bands around the differences in impulse response.
After completing your PhD, you went to work at the Bank of England during the time of the 2008 Global Financial Crisis. Could you tell us about your role here and what you did?
I joined the Bank of England in February 2007, working in the Financial Stability area in the International Finance Division. I was part of the policy team, which looked at the risks to the international financial system and liaised with the IMF. Part of my job was to give our opinion on what the IMF could usefully do to help policymaking in the UK. We were working closely with the Treasury and Bank of England officials who were in Washington and were providing advice on what we thought could be a beneficial role for the IMF. It was interesting because I joined in this Great Moderation period; when the IMF was not lending very much because macroeconomically, most countries, even emerging markets, were relatively stable. Therefore, we were thinking “If the IMF is not doing its traditional lending because there are no crises, then what should they be doing?”. As a result, we looked at international spillovers, building models of capital flows and understanding their role as well as international surveillance. This means that the IMF can account for the links between all different economies and internalise all the possible externalities. Then come late 2007 we have the Global Financial Crisis, which resulted in IMF lending, even to some advanced economies.
The Financial Policy Committee (FPC) is a relatively new committee at the Bank of England, in charge of creating macroprudential policy. Could you briefly explain what macroprudential policy is and some of the tools the FPC can use to fulfil their objectives?
The idea of macroprudential policy is to try and mitigate boom and bust cycles in credit, essentially. The idea is to have counter-cyclical policy i.e. when the economy is doing well, we do not extend too much credit and during times when the economy is not doing well, we could perhaps extend more credit and help the economy. There are various tools and many instruments that can be used. It can be to alter reserve requirements, it can be capital or liquidity requirements, it can even be things like the loan-to-value ratio or affordability ratios, like the loan-to-income ratio. We can also limit cross-border lending by banks. There are many different tools that can be used, but the overall main purpose is to dampen credit cycles. Different countries have used macroprudential policy to different degrees.
Watch this video to learn more about Macroprudential Policy
What actions do you think developed countries such as the UK and US should take to limit future recessions/financial crises?
I think the important thing is to have a good framework for monetary policy. Inflation targeting works well in keeping inflation under control.
Central Bank independence is important. This independence needs to be partnered with transparency and accountability. It is crucial to learn from history. We did it with the Global Financial Crisis and the creation of the FPC. Now with the FPC, the framework is better than it was before, and we are always trying to see where the vulnerabilities may be and understand where the next crisis could be coming from. Whilst we do this, there are many factors we simply cannot predict. We could not predict COVID, we could not predict the war in Ukraine and so much more. Many things in life are unpredictable by nature, but it is still important to address the vulnerabilities that we can see and improve the policy frameworks when they need to be improved. This goes for every country.
But it is also important to remember that not every country is at the same stage of economic and institutional development. That is where the IMF and the World Bank have a very important role to play. Because they see what is going on in different countries and they see what kinds of policy frameworks work well, allowing them to provide advice to countries that may need to improve their frameworks.
You have an extremely distinguished career having obtained your PhD from MIT, to lecturing at Trinity College Cambridge and now at King’s College London. What advice would you give to students hoping to gain admission to top universities, as well as those hoping to undertake a career in academia?
It is important to remember that getting into these top universities is not easy, and it is getting harder. I did my undergraduate studies in Portugal, masters at the LSE and am very lucky and very grateful to have been able to do my PhD at MIT. The experience was extremely enriching. However, it has gotten harder to get into the top universities, but I think the most important thing is to have a true interest in your subject.
With regards to academia, it is also very difficult. I did not pursue that path immediately after my PhD as I went to the Bank of England. The hard part is the motivation. There is no one to tell you “Well done” when you complete a piece of research, and often that research is rejected at journals. But what is most important to me is the intellectual challenge. At the Bank of England, I was doing conjunctural policy analysis, and I enjoyed it, but after a year and a half, I was missing the intellectual challenge and the opportunity to think about research questions in a deep way. For me, being an academic is a vocation.
It is also important not to give up. I have days when the regressions that I run do not make any sense. You also do not get much external reward. But if you do a good job as a teacher, that is rewarding. When your research does get published, that is also rewarding.
Lastly, what are some books, articles, academic pieces of work you would recommend to aspiring economists?
When I was a student, I used to read The Economist and I would cut out the bits that I found inspiring which gave me ideas for what I wanted to research further. I also think The Financial Times is a great platform as well.
In terms of books, I probably would recommend an econometrics textbook. One I would recommend is “Mastering Metrics” by Angrist and Pischke. Angrist was one of my PhD supervisors and won the Nobel Prize in Economics in 2021. His approach to data analysis and identification of causal effects has shaped the way I do research. This book was written with Pischke, who is a professor at the LSE. It is an undergraduate textbook; therefore, it is accessible and has quite a bit of humour. A slightly more advanced book is “Causal Inference: The Mixtape” by Scott Cunningham, which is free online, and is a good resource to learn how to use data and Stata.
Comments