- We quantify both the direct and the portfolio re-balancing impact,
emphasizing the role of investor heterogeneity.
- By purchasing large quantities of assets, the central banks aim to affect asset prices
throughout the economy.
- In this paper, we set out to deal with these challenges and to quantify both the direct
and indirect effects of large-scale asset purchases.
- In this paper, we set out to deal with these challenges and to quantify both the direct
and indirect effects of large-scale asset purchases.
- In the first part of the paper, we assess the importance of investor heterogeneity for
the direct effects of central bank purchases.
- Large-scale asset purchases are isomorphic to
negative changes in supply, as they effectively reduce the supply of bonds available to private
investors in the market.
- We weigh the estimates by the portfolio composition (number of
securities held) of the respective asset types for the average fund.
- ECB Working Paper Series No 2938
5
price of one security depends on the slope of the demand curve for that asset.
- However, in our setting, supply changes are not
necessarily randomly allocated, as purchases could be correlated with asset characteristics.
- To address this challenge, we isolate a random component of purchases and, with that,
estimate the price elasticity of each type of investor.
- For each investor type, we regress
the security holdings on that security?s yield, instrumenting the yield with ECB purchases.
- Given the estimated investor elasticities, we construct a security-level measure of the
price elasticity of the investor base.
- To address this concern, we exploit quasi-exogenous cross-sectional variation in
investor base composition across securities, measuring the share held by each investor type
prior to the announcement and implementation of asset purchases.
- With this measure, we can finally test how investor composition affects the direct impact of
central bank purchases.
- In
this part of the analysis, we focus on mutual funds, one investor type we estimated to be
elastic.
- This insight can help assess the calibration and targeting of asset purchase programs by central banks based on their objective
function.
- This paper relates to several strands of literature on unconventional monetary policy, intermediary and demand-system asset pricing.
- First, we relate to the literature on large-scale asset purchases in the Euro Area, studying announcement effects of purchases (Andrade et al.
- We contribute to
this literature by estimating spillover effects on assets that are not eligible for central bank
purchases while emphasizing the role of investor heterogeneity and developing a novel identification methodology.
- Our contribution to this literature is to analyze the role of investors?
preferences for the propagation of large scale asset purchases.
- Then the aggregate demand curve for the asset is
X
k
?k x k =
X
?k (?k ? p?k )
k
where ?k is the share of investor k in the market.
- P
By adding market clearing (S = k ?k xk ) we can derive that the price of the asset takes
the following form:
P
?k ?k
S
p = Pk
?P
.
- We denote with Dt = [dt,k , dt,k? , ...dt,K ] the N xK matrix of
demand of each investor type for each security.
- 4
Direct effects and investor heterogeneity
In this section we study the following research question: do effects of asset purchases depend
on investor composition?
- In order to answer this question, we test how purchases affect yields and how these effects
vary as investor composition changes.
- Third, we measure investor composition by exploiting
quasi-random variation in ex-ante investor composition.
- In the next section we will investigate how this
security-level heterogeneity in investor composition matters for the effects of asset purchases.
- 4.2
Demand elasticities of investors
In this section we investigate how different investor types react to Large Scale Asset Purchases
(LSAPs).
- In section 4 we estimated heterogeneity in elasticity across investor types based on how
price-elastic they were to ECB purchases.
- These results are in line with Table 3,
where we estimate heterogeneity in elasticity across investor types.
- To measure investor composition base heterogeneity we construct
ECB Working Paper Series No 2938
38
security-level Weighted elasticity as we did in section 4.
- Weighted Elasticity: Bn,t?1 =
X
??n,k,t?1 ?k
(8)
k
where ??n,k,t?1 is the share of the asset n held by investor k at t-1 and ?k are the elasticities
of investor types estimated in section 4.
- 7
Conclusion
In this paper, we examine the role of investor heterogeneity for both direct and indirect
effects of central bank purchases.
- Our findings show substantial heterogeneity in
exposure to large-scale asset purchases across mutual funds.
