The decline in homeownership rates has generated significant concern among policymakers and the media. A particular concern is the role of macroprudential policies – such as loan-to-value (LTV) constraints and debt-service-to-income limits – in limiting access to housing.
While these policies appear successful in limiting credit growth and reducing house price growth, many worry that they may inadvertently exclude first-time buyers and low-income households from homeownership.
Calls to relax macroprudential rules
These concerns have led many countries to loosen LTV constraints or other macroprudential measures. At the end of 2025, the Danish government proposed introducing 40-year mortgages to help first-time buyers, Donald Trump posted about introducing 50-year mortgages in the United States, and Sweden relaxed LTV limits on April 1.
In a recent paper, I examine the impact of macroprudential policies on homeownership rates in aggregate and for specific groups. The results are clear: these policies have little impact on homeownership rates.
The actual impact of macroprudential policies
There is limited evidence on the actual effect of macroprudential policies on homeownership rates. Studies on the impact of macroprudential policies have almost exclusively focused on house prices and mortgage credit growth, rather than homeownership rates. A recent meta-analysis compiled data from 58 empirical studies, and 6,000 estimates does not mention homeownership once.
In my study, I use country-level panel data from 28 European countries (2003–2017) and a narrative approach to identify macroprudential policy shocks. I find very limited evidence that macroprudential policies reduce homeownership rates: neither in aggregate, nor for low-income households, nor for mortgagors, and not even up to five years after implementation.
The null results are precisely estimated and hold across multiple specifications. The data rule out declines larger than roughly 0.23 percentage points in the homeownership rate following a macroprudential tightening. For large, well-documented policy shocks where previous literature has found substantial effects on credit and house prices, the 95percent confidence intervals rule out negative effects larger than 78 basis points.
Ownership rates remain consistent even during times of crisis
Underlying this result is a well-known but often overlooked fact: homeownership rates vary little over time, in contrast to what is often implied in policy discussions. It is striking that the homeownership rate is such a slow-moving variable, even during periods of significant economic change. In many countries in Central Europe, households were able to purchase their properties at significant discounts after the fall of the Soviet Union, and the homeownership rate remains very high today. In Norway, the homeownership rate is approximately 80 percent, whereas in Germany, it is below 50 percent. The rate in all these countries has remained approximately constant over the last 15 years.
The null results hold across multiple specifications and are precisely estimated enough to rule out economically large effects. This stands in sharp contrast to the well-documented effects of macroprudential policies on other housing market variables.
The results are also important for theoretical modeling of housing markets. Leading macroeconomic models predict substantial effects of macroprudential policies on homeownership rates. For example, Garriga et al. (2020) predict a 2.8 percentage point decline in homeownership from tighter LTV limits. Kaplan et al. (2020) predict a 3percent increase from relaxing LTV constraints (implying a similar magnitude decline from tightening). In contrast, the empirical estimates are close to zero—an order of magnitude smaller than model predictions.
The case of Sweden and other large shocks
Sweden provides a compelling example. In 2016, Sweden introduced an amortization requirement, a policy that previous research has shown had significant effects on mortgage credit and house prices. Yet, as shown in the figure below, Sweden’s homeownership rate remained remarkably stable, declining only slightly from around 67percent to 64percent over the entire period, with no discernible break following the 2016 policy change.
We can go even further. Examining several other instances of large macroprudential shocks studied in the academic literature reveals that the homeownership rate remains remarkably stable across all of them. Examining policies implemented in Norway, Ireland, Switzerland, Finland, Sweden, the UK, the Netherlands, and Portugal, we do not se. Ie any large movements in the aggregate homeownership rate, even though these policies have been found to influence credit and housing market activity.
The Big Picture
The findings have several important implications for how we think about housing policy.
Concerns that macroprudential policies systematically exclude certain households from ownership appear to be overstated. But there is a catch: based on previous literature, macroprudential policy tightening lowers house prices and credit, which may lead to wealth transfers from existing homeowners. This means that while these policies do not prevent people from becoming homeowners, they do affect the housing market in other ways.
If policymakers want to increase homeownership rates, relaxing macroprudential policies may be missing the mark. My findings suggest that macroprudential demand-side interventions, such as lowering downpayment requirements or providing subsidized savings accounts, may not effectively increase homeownership rates among young and first-time buyers. Instead, they may primarily lead to wealth transfers to existing homeowners through higher house prices, rather than expanding homeownership access. This highlights the importance of considering supply-side factors when designing policies to address housing market access concerns.
Aggregate dynamics of homeownership
Finally, some might argue that aggregate data hides important dynamics at the borrower level. However, homeownership rates are fundamentally an equilibrium object determined by supply and demand interactions in the housing market. Randomized controlled trials or micro-level studies cannot capture general equilibrium effects, and thus, miss market-level responses such as price adjustments, supply responses, or changes in the rental market that determine aggregate ownership rates.
While transaction volumes may change in response to macroprudential policies, they may not affect the aggregate homeownership rate, as existing homeowners may choose to simply remain in their homes. For the homeownership rate to decline, existing homeowners must exit the ownership market by selling to landlords who rent them out, or they must themselves convert their property to rental units. This represents a different mechanism than changes in transaction activity.
A final caveat: These findings should not be interpreted as an endorsement of macroprudential policies more broadly. While the evidence suggests that such policies do not systematically reduce homeownership rates in aggregate, they may still have other unintended consequences or welfare effects. For instance, macroprudential policies could lead to a shift in credit activity to less-regulated sectors such as shadow banking, alternative lenders, or cross-border lending, or there could be changes in the moving patterns within local housing markets. Alternatively, households could reduce their liquid savings to comply with macroprudential policies, with ambiguous effects on financial stability.
For more details, see the full paper: “Macroprudential policies and homeownership rates”
Claes Bäckman is Postdoctoral Researcher at the Leibniz Institute for Financial Research SAFE.
Blog entries represent the authors‘ personal opinion and do not necessarily reflect the views of the Leibniz Institute for Financial Research SAFE or its staff.
