Quarterly Economic Commentary Winter 2024

McQuinn, Kieran; O'Toole, Conor; O'Shea, Dónal Research Report Quarterly Economic Commentary Winter 2024 ESRI Forecasting Series Provided in Cooperation with: The Economic and Social Research Institute (ESRI), Dublin Suggested Citation: McQuinn, Kieran; O'Toole, Conor; O'Shea, Dónal (2024) : Quarterly Economic Commentary Winter 2024, ESRI Forecasting Series, The Economic and Social Research Institute (ESRI), Dublin, https://doi.org/10.26504/qec2024win This Version is available at: https://hdl.handle.net/10419/316961 Standard-Nutzungsbedingungen: Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden. Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen. Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. Terms of use: Documents in EconStor may be saved and copied for your personal and scholarly purposes. You are not to copy documents for public or commercial purposes, to exhibit the documents publicly, to make them publicly available on the internet, or to distribute or otherwise use the documents in public. If the documents have been made available under an Open Content Licence (especially Creative Commons Licences), you may exercise further usage rights as specified in the indicated licence. https://creativecommons.org/licenses/by/4.0/ ESRI Forecasting Series December 2024 Quarterly Economic Commentary Winter 2024 KIERAN MCQUINN, CONOR O’TOOLE AND DÓNAL O'SHEA QUARTERLY ECONOMIC COMMENTARY Kieran McQuinn Conor O’Toole Dónal O’Shea Winter 2024 The forecasts in this Commentary are based on data available by 5 December 2024. Draft completed on 9 December 2024. Available to download from www.esri.ie https://doi.org/10.26504/qec2024win © 2024 The Economic and Social Research Institute Whitaker Square, Sir John Rogerson’s Quay, Dublin 2 This Open Access work is licensed under a Creative Commons Attribution 4.0 International License, which permits unrestricted use, distribution and reproduction in any medium, provided the original work is properly credited. ABOUT THE ESRI The Economic and Social Research Institute (ESRI) advances evidence-based policymaking that supports economic sustainability and social progress in Ireland. ESRI researchers apply the highest standards of academic excellence to challenges facing policymakers, focusing on ten areas of critical importance to 21st century Ireland. The Institute was founded in 1960 by a group of senior civil servants led by Dr T.K. Whitaker, who identified the need for independent and in-depth research analysis. Since then, the Institute has remained committed to independent research and its work is free of any expressed ideology or political position. The Institute publishes all research reaching the appropriate academic standard, irrespective of its findings or who funds the research. The ESRI is a company limited by guarantee, answerable to its members and governed by a Council, comprising up to 14 representatives drawn from a crosssection of ESRI members from academia, civil services, state agencies, businesses and civil society. Funding for the ESRI comes from research programmes supported by government departments and agencies, public bodies, competitive research programmes, membership fees and an annual grant-in-aid from the Department of Public Expenditure, NDP Delivery and Reform. Further information is available at www.esri.ie. THE AUTHORS Kieran McQuinn is a Research Professor at the Economic and Social Research Institute (ESRI) and an Adjunct Professor at Trinity College Dublin (TCD). Conor O’Toole is an Associate Research Professor at the ESRI and an Adjunct Professor at TCD. Dónal O’Shea is a Research Assistant at the ESRI. Research Notes are short papers on focused research issues. They are subject to refereeing prior to publication. Special Articles are published in the Quarterly Economic Commentary in order to foster high-quality debate on various aspects of the Irish economy and Irish economic policy. They are subject to refereeing prior to publication. The Quarterly Economic Commentary has been accepted for publication by the Institute, which does not itself take institutional policy positions. It has been peer reviewed by ESRI research colleagues prior to publication. The authors are solely responsible for the content and the views expressed. i| Table of contents TABLE OF CONTENTS SUMMARY TABLE .................................................................................................................................... II THE IRISH ECONOMY – FORECAST OVERVIEW ....................................................................................... 3 DOMESTIC AND INTERNATIONAL OUTLOOK .......................................................................................... 4 GENERAL ASSESSMENT ......................................................................................................................... 49 RESEARCH NOTE Estimating Ireland’s labour share………………………………………………………………………………………..53 Dónal O’Shea SPECIAL ARTICLE Distributional impact of tax and welfare policies: Budget 2025…………………………………………..71 Karina Doorley, Shane Dunne, Claire Keane, Simona Sándorová, Agathe Simon SPECIAL ARTICLE Assessing expectations of European house prices…………………..…………………………………………..93 Akhilesh Kumar Verma and Kieran McQuinn Quarterly Economic Commentary – Winter 2024 | ii SUMMARY TABLE 2022 2023 2024 2025 Output (real annual growth %) Private consumer expenditure 10.9 4.9 2.6 3.2 Public net current expenditure 3.0 4.2 4.3 3.8 Investment 4.1 2.8 -23.4 19.9 Modified investment 10.4 -4.3 4.1 6.8 Exports 13.6 -6.0 9.8 4.3 Imports 16.4 1.3 8.3 5.7 Gross domestic product (GDP) 8.7 -5.7 -1.1 4.5 Modified domestic demand 9.1 2.7 3.2 4.1 Prices (annual growth %) Consumer Price Index (CPI) 7.8% 6.3% 2.1% 1.0% Labour market Employment levels (’000) 2,639 2,705 2,760 2,811 Unemployment levels (’000) 122 121 125 125 Unemployment rate (as % of labour force) 4.4% 4.3% 4.3% 4.2% Public finance General government balance (€bn) 8.6 7.5 23.4 9.4 General government balance (% of GDP) 1.7 1.5 4.6 1.8 3 | Quarterly economic commentary – Winter 2024 The Irish economy – Forecast overview • As we approach the end of the year, the Irish domestic economy looks set to register strong growth in 2024. We expect modified domestic demand (MDD) to increase by 3.2 per cent in 2024 before growing to 4.1 per cent in 2025 driven by real income growth and higher housing investment. • Overall gross domestic product (GDP) is expected to decline on the back of large intellectual property investment outflows and increased imports. • Overall, the robust performance means that the labour market has seen a continued increase in employment, with unemployment near to its historically low rate. A research note to the Commentary by O’Shea (2024) looks at trends in the Irish labour share over the past 25 years. • Exchequer receipts have also performed strongly in 2024. Even in the absence of the ‘Apple payment’, we believe the adjusted general government balance (GGB) will be around 2 per cent of GDP this year. • However, for 2025, there are notable downside risks, in particular with the potential for the new US administration to implement the economic policies outlined during the presidential campaign. A box by Fitzgerald highlights the ongoing importance of US investment in the Irish economy. • The prospect of a global trade war, given the Trump administration’s proposals on tariffs, the impact of taxation policy on intellectual property location, allied to the possible targeting of the pharmaceutical sector based in Ireland, could have particular implications for both activity levels in the domestic economy and for the Exchequer receipts. • If these impacts materialise more quickly than expected, particularly those on the public finances, some aspects of planned future expenditure levels outlined in Budget 2025 may have to be revised in the new year. • While Budget 2025 had a welcome significant commitment to investment in the Irish economy going forward, there were other elements in the Budget which could have been more targeted, as suggested by Doorley et al. (2024). • With house prices experiencing a resurgence in the present year, the Commentary devotes a significant amount of attention to house price dynamics. A box by Egan and McQuinn assesses the sustainability of recent movements while a special article by Kumar Verma and McQuinn (2024) examines the potential influences on house price expectations. Domestic and International Outlook | 10 FIGURE 7 COMPONENTS OF BUILDING AND CONSTRUCTION INVESTMENT – YEAR-ON-YEAR GROWTH – CONSTANT PRICES Source: Central Statistics Office. Overall, we expect investment to decline by 23.4 per cent in 2024, mainly because of the drop in intellectual property activity outlined above. We expect this sharp decline to reverse in 2025, with a growth rate of 19.9 per cent for the year. 3 For the more stable modified investment, we expect growth of 4.1 per cent in 2024 but a recovery to 6.8 per cent growth in 2025. This reflects the increase in construction investment for dwellings in 2024 and the anticipated lower interest rate environment. Given the expected changes in both consumption and modified investment, we now believe that modified domestic demand (MDD), the preferred indicator of domestic economic activity, will grow by a robust 3.2 per cent in 2024 and at an enhanced rate of 4.1 per cent in 2025. Multinational exports recover but major increase in risk due to likely Trump policy direction Notwithstanding the weakness seen in multinational exports over the past 18 months, there is emerging evidence that a recovery is underway in key sectors, such as pharmaceuticals, which have begun to grow strongly again. Figure 8 contrasts the key sectors of pharmaceuticals and computer services with the underlying economy. Goods exports in pharmaceuticals grew by 26.3 per cent in Q3 2024. Goods exports declined in both categories in 2023; however, nonpharmaceutical goods exports have recovered more slowly. For services, computer 3 This rebound makes the technical assumption that the level of intellectual property investment in Q2 2025 will return to a more normalised level. This leads to an increase in this category of investment in 2025, with knock on implications for exports and imports. -30% -20% -10% 0% 10% 20% 30% 40% 50% 60% Q1 2022 Q2 2022 Q3 2022 Q4 2022 Q1 2023 Q2 2023 Q3 2023 Q4 2023 Q1 2024 Q2 2024 Q3 2024 Dwellings Other building and construction Improvements 11 | Quarterly economic commentary – Winter 2024 services exports have been growing faster than underlying services exports since the beginning of 2023. The final series presented in Figure 8 is labelled ‘internationalisation’ and consists of contract manufacturing and goods for processing. The volatility of this series has driven the volatility of the headline export figure in recent years. FIGURE 8 COMPONENTS OF EXPORTS – VALUE – YEAR-ON-YEAR GROWTH Source: Central Statistics Office. Figure 9 presents goods exports divided into chemicals and other goods. Chemicals consistently account for over 60 per cent of goods exports. Notably, the growth rate of chemicals exports in Q3 2024 (26.3 per cent) far exceeded the growth rate of exports of other goods (11 per cent). FIGURE 9 GOODS EXPORTS BY COMPONENT – VALUE – SHARE OF TOTAL AND YEAR-ON-YEAR GROWTH Source: Central Statistics Office. Note: ‘Other goods’ is calculated by taking chemicals from the total value of goods exports. Despite the recent rebound in multinational exports, there has been a major increase in the downside risks for the Irish multinational sector since the US -60.0% -40.0% -20.0% 0.0% 20.0% 40.0% 60.0% Q1 2022 Q2 2022 Q3 2022 Q4 2022 Q1 2023 Q2 2023 Q3 2023 Q4 2023 Q1 2024 Q2 2024 Q3 2024 Goods - Pharma Goods - Non-pharma Internationalisation Computer services Other services -50% 0% 50% 0% 50% 100% Q1 2022 Q2 2022 Q3 2022 Q4 2022 Q1 2023 Q2 2023 Q3 2023 Q4 2023 Q1 2024 Q2 2024 Q3 2024 Other goods (excluding chemicals) (share) (LHS) Chemicals (share) (LHS) Chemicals (growth y-on-y) (RHS) Other goods (excluding chemicals) (growth y-on-y) (RHS) Domestic and International Outlook | 12 presidential election. The incoming Trump administration has signalled its intention to make a significant departure in terms of US trade policy, with recent announcements of tariffs on imports from Canada, China and Mexico. There has also been a notable focus on the Irish trade surplus and the role of US multinational profits in the Irish economy. It is likely that the impact of any protectionist trade stance in the US would be multifaceted for Ireland. In general, increased tariffs are likely to lower trade bilaterally, but there are likely to be second round effects more generally if world trade is disrupted. As a small, and extremely open, economy Ireland has historically been a major beneficiary of globalisation. Changes in policy direction that lead to greater trade fragmentation are likely to disproportionately impact the Irish traded sector. The implications of these potential disruptions for Ireland are evident. A large proportion of our trade is exported directly to the US. Figure 10 presents the direct export shares of Irish merchandise trade to the US. In 2024 (Q1–Q3) approximately 31.2 per cent of our merchandise exports went to the US. For pharmaceutical products, this share was over 38 per cent for the same period. Given this reliance on the US as a destination market, and the importance of pharma to our exports, any disruption in these trade flows would have notable economic consequences, as well as an impact on the public finances through lower corporation taxation receipts. These data do not include service exports or the internationalisation flows such as contract manufacturing that are undertaken by US companies. FIGURE 10 US SHARE OF IRISH MERCHANDISE EXPORTS Source: Central Statistics Office and ESRI calculations. Another indication of the exposure of the Irish economy to significant changes in international trading conditions can be found in the merchandise trade balance for Ireland with the US. It is presented in Figure 11 as a percentage of Irish GDP. It is 0% 5% 10% 15% 20% 25% 30% 35% 40% 45% 2019 2020 2021 2022 2023 2024 US share of total US share of non-pharma US share of pharma 13 | Quarterly economic commentary – Winter 2024 clear Ireland runs a very large merchandise trade surplus with the US, which has fluctuated anywhere between 6 and 9 per cent of Irish GDP. FIGURE 11 MERCHANDISE TRADE BALANCE WITH US, % OF IRISH GDP Source: Central Statistics Office and ESRI calculations. In terms of the overall foreign direct investment (FDI) flows, Box A below considers further these impacts, and the reliance of the Irish FDI sector on US-owned companies. The impact of the new administration on FDI is also highly uncertain but comes with considerable downside risks. In the shorter term, the increased uncertainty, and desire by the Trump administration to return manufacturing activity to the US, could disrupt or slow new multinational investments into Ireland by US-owned companies. In the medium term, many US multinationals, in particular in the pharmaceutical industry, have made large, sunk-cost investments in Ireland in plant and machinery. These investment choices were driven by multiple factors including taxation, access to EU markets and other structural factors. These investments are also likely to have long payback periods beyond the four years of the incoming Trump administration. BOX A FOREIGN DIRECT INVESTMENT IN IRELAND 1. Introduction The Irish economy is unusual among EU countries in terms of the very large role played by foreign multinationals in national output and income. This role has continued to grow in recent years, contributing to the overall success of the economy in exiting from the global financial crisis and returning to steady rapid growth. 0.0% 2.0% 4.0% 6.0% 8.0% 10.0% 12.0% Q1 2015 Q3 2015 Q1 2016 Q3 2016 Q1 2017 Q3 2017 Q1 2018 Q3 2018 Q1 2019 Q3 2019 Q1 2020 Q3 2020 Q1 2021 Q3 2021 Q1 2022 Q3 2022 Q1 2023 Q3 2023 Q1 2024 Domestic and International Outlook | 14 The contribution to net national product (NNP) from foreign multinationals is accounted for by their wage bill and the corporation tax they pay.4,5 For domestic firms, the contribution is equal to their wage bill and their profits, before tax. Table 1 shows the share of NNP accounted for by foreign and domestic business for 2013 and 2023. TABLE 1 SHARE OF NNP BY SECTOR AND OWNERSHIP, % 2013 2023 Output (real annual growth %) Total Foreign Domestic Total Foreign Domestic Agriculture 1 0 1 1 0 1 Manufacturing 12 5 6 12 9 3 Electricity, gas, and water 2 0 2 1 0 1 Construction 3 0 3 5 1 5 Distribution, transport and restaurants 22 5 17 18 5 13 Information and communication 6 3 3 9 7 2 Financial services 10 4 6 8 4 3 Real estate activities 7 0 7 10 0 9 Professional services 10 2 7 14 5 9 Public admin, education and health 24 0 24 19 1 19 Arts, entertainment etc. 3 0 3 2 0 2 Total 100 21 79 100 32 68 Source: Central Statistics Office institutional sector accounts. Note: Figures may not sum to 100 due to rounding. In 2013 the foreign sector of the economy accounted for around 21 per cent of NNP, but by 2023 it accounted for 32 per cent, a very large increase in share. This growth in share was accounted for by a major expansion in the value added to the Irish economy from foreign-owned firms operating in the manufacturing, IT and professional services sectors. 2. Comparison with EU27 The CSO data do not allow a breakdown of the contribution to NNP by country of ownership of foreign multinational enterprises (MNEs). However, Eurostat data show FDI across the EU for 2021, including the number of enterprises, value added, numbers employed, the wage bill and the gross operating surplus. The foreign investors are disaggregated by country of origin of the investor, as well as by country where the investment is located. 4 The issue of country of ownership is not always clear. As companies get taken over, the nationality of ownership may change. In addition, a company’s shareholders can reside in different countries, which can also be different to the country in which the company’s head office is actually registered. 5 The standard measure now used to measure national income in Ireland today is modified gross national income (GNI*). However, this measure includes some depreciation on the capital stock, capital that is used up and has to be replaced to maintain output (and income) at its current level. Thus a better measure of national economic welfare is net national income (NNI) – excluding depreciation. However, this measure is not always available for other economies, especially when measuring output at constant prices, so GDP and GNI are preferred for international comparisons. The difference between NNI and net national product (NNP), used here, is indirect taxes and subsidies. 15 | Quarterly economic commentary – Winter 2024 The data on corporation tax paid by foreign firms are not available for other EU countries, allowing a comparison of the contribution of foreign firms to national income between Ireland and other EU countries, as in Table 1. Value added data for Ireland are distorted by the relocation of substantial intellectual property assets by US firms. So it is not useful to use value added data when comparing Ireland with the EU. Instead we focus on employment and wages, as the data allow a comparison between Ireland and the EU of the numbers employed and the wage bill paid by foreign multinationals. 2.1 Employment For 2021, Table 2 shows the share of total employment in Ireland and the EU27 that is accounted for by companies originating from a range of different countries. Data for China and Hong Kong are not available for the EU27. In 2021, US firms accounted for 7.8 per cent of total employment in Ireland, compared to only 1.8 per cent in the EU as a whole. Other EU27 firms and UK firms each accounted for around 5 per cent of employment in Ireland. However, UK firms accounted for only 1 per cent of employment in the EU as a whole. The EU shows a very different pattern to that for Ireland, with foreign multinationals accounting for a much smaller share of total employment – 12 per cent for the EU27 compared to 24 per cent for Ireland. For the EU27, investment from other countries within the EU accounts for a higher share of domestic employment than in Ireland, at 7 per cent. This is more than offset by the much smaller share of investment from outside the EU. For Ireland such firms (those outside the EU) account for 19 per cent of total employment compared to only 5 per cent for the EU27. TABLE 2 SHARE OF TOTAL EMPLOYMENT ACCOUNTED FOR BY FOREIGN FIRMS, 2021, % Share of employment( %) Ireland EU27 Intra-EU27 (from 2020) 4.9 6.8 Norway 0.0 0.1 Switzerland 0.7 0.8 UK 5.4 0.9 Türkiye 0.0 0.0 Extra-EU27 (from 2020) 18.7 5.0 Canada 0.7 0.1 United States 7.8 1.8 China except Hong Kong 0.1 Hong Kong 0.1 Japan 0.5 0.4 Australia 0.2 0.0 Domestic country 76.4 88.2 All FDI 23.6 11.8 Economy 100.0 100.0 Domestic and International Outlook | 16 TABLE 3 SHARE OF EMPLOYMENT IN THE EU27 LOCATED IN IRELAND FOR MULTINATIONALS Ireland’s share of EU27 (%) Employment Wages Intra-EU27 (from 2020) 0.9 1.3 Norway 0.5 0.6 Switzerland 1.1 1.7 UK 7.8 6.8 Extra-EU27 (from 2020) 4.7 5.3 Canada 7.4 6.4 United States 5.5 6.7 Japan 1.7 2.3 Australia 6.9 6.3 All FDI 2.5 3.2 Economy 1.3 1.7 Table 3 shows that, while Ireland accounted for 1.3 per cent of total employment in the EU in 2021, it accounted for 2.5 per cent of employment by foreign multinationals. While Ireland’s share of intra-EU employment was lower than average, this was more than offset by the higher shares of employment by UK and US firms. 2.2 Wages The data in Table 3 show the wage bill of foreign multinationals in each country. With higher living standards in Ireland, and hence higher pay rates, the wage bill in Ireland of MNEs accounted for 1.7 per cent of the EU wage bill, higher than the employment share. The share of the wage bill paid in Ireland by firms from other EU countries was higher than the employment shares, as employment in Ireland, accounted for by such firms, was relatively highly paid. This is especially true for US and Japanese firms operating in Ireland. By contrast, UK firms tend to employ people at lower pay rates in Ireland than elsewhere in the EU. TABLE 4 AVERAGE EARNINGS IN FOREIGN OWNED ENTERPRISES, €, THOUSANDS Ireland EU27 Ireland/EU27 Intra-EU27 (from 2020) 60.0 42.5 1.41 Norway 61.0 50.4 1.21 Switzerland 79.2 51.5 1.54 UK 46.7 52.9 0.88 Türkiye 75.4 23.8 3.17 Extra-EU27 (from 2020) 62.3 55.2 1.13 Canada 47.5 54.6 0.87 United States 76.5 63.0 1.21 China except Hong Kong 57.9 Hong Kong 58.1 Japan 73.7 55.9 1.32 Australia 50.3 55.2 0.91 Domestic 44.1 34.9 1.27 All FDI 61.9 47.8 1.29 Economy 48.3 36.4 1.33 17 | Quarterly economic commentary – Winter 2024 Table 4 shows that average earnings in Ireland were a third higher than for the EU. The table shows average earnings per employee for Ireland and the EU, broken down by firms’ country of ownership. Foreign MNEs pay more than domestic firms: in Ireland foreign firms paid €62,000 a year compared to €44,000 for the rest of the economy, while in the EU foreign firms paid €48,000 compared to €35,000 for the rest of the economy. US firms paid 21 per cent more in Ireland than the EU average for such firms. Because of their high pay rates, and their big share of employment, these firms explain why pay rates in foreign MNEs in Ireland were 29 per cent higher than for the EU. Pay rates for UK-owned firms in Ireland were under 90 per cent of what UK firms pay elsewhere in the EU, reflecting the fact that UK firms in Ireland are concentrated in lower paid sectors, with a smaller share of skilled workers. In the rest of the EU, UK firms employ a greater share of highly educated employees. 3. Sectoral detail Table 5 shows Ireland’s share of foreign multinational employment and wages in the EU for a range of different sectors where there are detailed data. As can be seen from the table, 2.5 per cent of all employees in foreign MNEs in the EU were located in Ireland. For US firms, the share was even higher at 5.5 per cent. The share of the wage bill paid in Ireland by foreign firms was very high: 5.5 per cent of the wages paid by such firms across the EU. Table 5 also shows that a large share of foreign firms’ employment in the EU in the IT, financial and professional services sectors occurred in Ireland. The IT sector in Ireland accounted for over 11 per cent of all employment in that sector in the EU by US firms. In addition, 14 per cent of the wage bill of US firms in the EU IT sector was paid in Ireland. TABLE 5 IRELAND’S SHARE OF EU TOTAL, % Employment Wage bill Ireland’s share of EU27 (%) All FDI US FDI All FDI US FDI Economy 2.5 5.5 5.5 6.7 Manufacturing 1.6 5.5 2.2 Manufacture of food products etc. 2.8 3.5 Wholesale and retail trade etc. 2.3 2.5 2.4 2.4 Information and communication 4.5 11.1 6.6 Computer programming etc. 4.7 11.4 7.1 14.2 Financial and insurance activities 6.5 20.2 6.8 18.0 Professional, scientific etc. 3.1 5.1 3.6 Administrative and support service 2.5 3.2 Domestic and International Outlook | 18 Separate data are not available for the pharmaceutical sector, which is an important sector in the Irish economy. It is also dominated by foreign, especially US, MNEs. They are included in the total for the manufacturing sector. 4. Corporation tax While the Revenue Commissioners give details of the corporation tax paid by all foreign MNEs, they don’t break it down by nationality of investor. However, Eurostat gives details of the profits (gross operating surplus) of multinationals by nationality. Assuming that the tax paid is proportional to their profits, Table 6 shows the distribution of corporation tax payments by nationality. TABLE 6 ESTIMATED CORPORATION TAX BY NATIONALITY OF INVESTOR, 2021, € MILLIONS €, Million Intra-EU 700 Norway 32 Switzerland 75 UK 603 Canada 109 United States 10,312 China except Hong Kong 97 Hong Kong 26 Japan 86 Australia 12 Other 234 All foreign multinationals 12,300 Irish firms 3,024 Total 15,324 This shows that the vast bulk of corporation tax paid by foreign firms is paid by US MNEs. This contrasts with the fact that US MNEs, despite paying high wages, only accounted for 40 per cent of wages paid by all foreign MNEs in Ireland in 2021. 5. Conclusions The Eurostat data show that US MNEs play a major role in the Irish economy, much greater than in the EU as a whole. Table 7 summarises the contribution to Irish net national product (NNP) of US firms, through corporation tax and wages. It also shows the contribution of all foreign multinationals. TABLE 7 CONTRIBUTION TO IRISH NNP 2021, PERCENTAGE POINTS US Total Corporation tax 6.0 7.1 Wage bill 8.4 21.4 Total 14.3 28.5 19 | Quarterly economic commentary – Winter 2024 Any changes in the coming years due to US or EU legislation, which affected the role of US firms in Ireland, could, as a result, have a big impact on the economy. This Box was prepared by John Fitzgerald. Global headwinds increase with US policy change While it is noted above that the Irish economy is at risk from a number of channels given a shift in US trade and fiscal policy under the forthcoming Trump administration, these headwinds extend to the global economy as well. As noted above, Ireland’s economy is small and highly globalised, which means that any reduction in global growth will naturally have an knock-on impact on Irish exports. In its most recent economic World Economic Outlook, the International Monetary Fund (IMF) attempted to quantify the impact of various policy changes and uncertainties on global growth. While their headline growth rates had pointed towards an upgrade of the outlook relative to April 2024 forecasts on the back of an improved US economy, it provided scenarios to assess the downside risk of a policy pivot towards protectionism. The IMF baseline forecast and scenarios are presented in Figure 12. Their first scenario (A) tested for the impact of a reduction in trade following a 10 per cent tariff imposed on trade flows among the EU, US and China, and a 10 per cent tariff by the US on trade from the rest of the world. This would lead to a 0.1 per cent reduction in world economic output in 2025 and 2026, and a larger impact in 2027. A second scenario (B) explores the impact on investment of higher trade uncertainty in the US and the eurozone. Thirdly, they apply a further scenario with lower migration flows, which again lowers economic output. Finally, they layer on the tighter financial conditions that are likely to occur if A, B and C materialise. The impacts of these cumulative effects are notable, with growth decreasing from 3.2 per cent in 2025 and 2026 to 2.5 and 2.1 per cent respectively. Given the sensitivity of Irish activity to world growth, it is likely that these impacts would pass fully through to the Irish economy, lowering the growth rate in the traded sectors substantially. Indeed, the scenario that the IMF deploy is a 10 per cent tariff, which may be on the benign side given recent announcements from the Trump administration. However, it must be acknowledged that considerable uncertainty surrounds the extent to which the policies proposed by the Trump campaign during the US presidential election will accord with the actual policy choices of the incoming administration there. Domestic and International Outlook | 26 model is estimated over the period 1981–2024 and the actual prices and modelbased estimates are compared in Figure 18 below.12 FIGURE 18 ACTUAL AND FITTED VALUES FROM THE HOUSE PRICE EQUATION: 2006–2024 Source: Quarterly Economic Commentary estimates. From the above figure, it is clear that there appeared to be significant undervaluation in the Irish market in the period up to 2018. This was a fall-out from the significant reduction in prices that occurred after the global financial crisis. House prices increased sharply and persistently during this period. From 2018 through 2022, the housing market was in equilibrium, with actual prices and those suggested by the model being practically the same. However, a divergence has emerged over the past 18 months, with actual house prices now somewhat larger than those suggested by the model. In addition to the model-based estimate, we also examine Irish house prices relative to trend by applying a Hodrick–Prescott (HP) filter. While not without its limitations,13 the HP filter is a commonly used tool to establish the trend of a variable over time, and is used extensively as part of central banks’ countercyclical capital buffer. The choice of smoothing parameter, λ, is set at 400,000 in line with work such as Drehmann et al. (2010). Both the model and filter-based estimates of house price overvaluation are plotted in Figure 19. Based on the most recent data from Q2 2024, the figure Irish house prices are over-valued by somewhere in the region of 8 to 10 per cent. 11 See https://www.rte.ie/news/business/2024/1016/1475774-house-prices-analysis/ for example. 12 Note all monetary variables are deflated by the CPI. 13 As pointed out by Alessi and Detken (2014), the HP filter is sensitive to the choice of the smoothing parameter and also suffers from endpoint bias. 150,000 200,000 250,000 300,000 350,000 400,000 450,000 2006Q1 2008Q1 2010Q1 2012Q1 2014Q1 2016Q1 2018Q1 2020Q1 2022Q1 2024Q1 House prices Actual Model based 27 | Quarterly economic commentary – Winter 2024 FIGURE 19 ESTIMATES OF DISEQUILIBRIUM IN THE IRISH RESIDENTIAL MARKET, 2006–2024 Source: Quarterly Economic Commentary estimates. As the above analysis clearly shows some sign of house price overvaluation, it is prudent to monitor other vulnerabilities related to the Irish residential real estate market. To do so, we follow an approach by Bengtsson et al. (2017), who analyse a set of indicators related to three dimensions of real estate sector vulnerabilities – valuation, household indebtedness and the credit cycle. Table 8 provides an overview of these indicators in the Irish context. TABLE 8 INDICATORS RELATED TO RESIDENTIAL REAL ESTATE VULNERABILITY INDICATORS DESCRIPTION SOURCE VALUATION Price-to-income (PTI) Nominal house price/disposable income OECD – Analytical house price indicators Price-to-rent (PTR) Nominal house price/nominal rent OECD – Analytical house price indicators Overvaluation (OVERVAL) % Deviation of actual house prices from model based equilibrium CSO, Central Bank of Ireland and authors’ calculations HOUSEHOLD INDEBTEDNESS Household debt-to-disposable income (DTI) Ratio of household debt to disposable income Central Bank of Ireland and authors’ calculations Debt Service Ratio of Households (DSR) Disposable income to mortgage repayments Central Bank of Ireland and authors’ calculations Households debt to total assets (DTA) Ratio of household debt to household total financial assets Central Bank of Ireland and authors’ calculations CREDIT CYCLE Credit for house purchases (CREDHP) Credit to domestic households for house purchases to GNI* Central Bank of Ireland, CSO and authors’ calculations Lending spreads (SPREAD) Difference between lending rates for house purchases and money market rates Central Bank of Ireland, Refinitiv and authors’ calculations Loan-to-deposit ratio (LTD) Ratio of banks total loans to total deposits Central Bank of Ireland and authors’ calculations -40% -30% -20% -10% 0% 10% 20% 30% 40% 50% 2006Q1 2008Q1 2010Q1 2012Q1 2014Q1 2016Q1 2018Q1 2020Q1 2022Q1 2024Q1 % Model Based Filter Based Domestic and International Outlook | 28 Figures 20 and 21 present the set of indicators related to Irish real estate vulnerability in two separate ways. Figure 20 examines the vulnerabilities at three selected points in time – Q4 2008, Q4 2016 and the most recent data available for 2024 – in the form of a radar chart. Figure 21 looks at the evolution over the entire 2003 to 2024 period in the form of a heat map.14 The data have been normalised, so that a value of 1 represents the highest level of vulnerability and a value of 0 the lowest. Figure 20 highlights the significant level of vulnerability in Q4 2008 across all six indicators related to valuation, household indebtedness and the credit cycle. Figure 20 also shows that by Q4 2016, the vulnerabilities had reduced significantly, with banks reducing their lending exposures, overvaluation in the housing sector dissipating across the three different indicators, and risks related to indebtedness receding. In the most recent period, a number of indicators appear to be showing signs of vulnerability. This includes the overvaluation of property prices relative to fundamentals, as described above. The other two indicators in this category, price-to-income (PTI) and price-to-rent (PTR), are also above the levels seen in Q4 2016, but still well below those of Q4 2008. In addition, there would also appear to be a significant increase in vulnerabilities across single indicators in the household indebtedness and credit cycle categories, namely in the DSR and the indicator related to lending spreads (SPREAD). The elevated level of vulnerability is likely driven solely by the higher interest rate environment, however. This is visible in Figure 21, which shows elevated vulnerability beginning in mid-2022. It is important to note that all other indicators across these categories remain low and the vulnerabilities are likely to dissipate as interest rates fall. FIGURE 20 VULNERABILITY IN IRELAND’S RESIDENTIAL REAL ESTATE MARKET – Q4 2008, Q4 2016 AND CURRENT 14 In the heat map, red indicates the highest level of vulnerability while green represents the lowest. 0 0.2 0.4 0.6 0.8 1PTI PTR OVERVAL DTI DTADSR CREDHP SPREAD LTD 2008Q4 2016Q4 Current 29 | Quarterly economic commentary – Winter 2024 FIGURE 21 VULNERABILITY IN IRELAND’S RESIDENTIAL REAL ESTATE MARKET – Q1 2003 – PRESENT PTI PTR OVERVAL DTI DTA DSR CREDHP SPREAD LTD 2005 VALUATION INDEBTEDNESS CREDIT 2003 2004 2023 2024 2017 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2018 2019 2020 2021 2022 Source: Quarterly Economic Commentary estimates. Conclusion The accelerated increase in house prices experienced so far in 2024 has led to concerns in the domestic market about the sustainability of such increases and the prospect of a painful correction such as that witnessed between 2007 and 2012. It is evident that an increasing number of Irish households are facing elevated leveraged positions in terms of the mortgaged debts they are carrying. This renders these households quite vulnerable, particularly to any labour market shock, both in terms of a sudden rise in unemployment and/or a decline in real wages. It also raises question marks around the capability of certain cohorts of the population to engage in homeownership, as both the DSR and house price-to-income ratio are increasing significantly. While credit growth is not as significant a factor as it was in the pre-Celtic Tiger era, there is recent evidence (Egan et al., 2024b) to suggest the growing contribution to recent house prices of changes in the loan-to-income ratio. Consequently, the Central Bank of Ireland must be particularly vigilant and prudent in any review of the mortgage measures in its macroprudential policy framework. Domestic and International Outlook | 30 References Bengtsson, E., M. Grothe and E. Lepers (2017). Home, safe home: cross-country monitoring framework for vulnerabilities in the residential real estate sector, Working Paper Series 2096, European Central Bank. Blanchard O. and W. Watson (1982). ‘Bubbles, rational expectations and financial markets’, in P. Wachtel (ed.) Crises in the economic and financial structure, Lexington Books, Lexington, MA, pp. 295–316. Drehmann, M., C. Borio, L. Gambacorta, G. Jimenez and C. Trucharte (2010). Countercyclical capital buffers: Exploring options, BIS Working Paper No. 317. Egan P., K. McQuinn and C. O’Toole (2024a). ‘How supply and demand affect national house prices: The case of Ireland’, Journal of Housing Economics. Vol. 65, September, 102006. Egan P., K. McQuinn and C. O’Toole (2024b). ‘Credit and house prices in the Irish residential market’, Intereconomics, Vol. 59, No. 5, pp. 293–300. McQuinn, K. (2017). ‘Irish house prices: Deja vu all over again?’, Special article, Quarterly Economic Commentary, winter, Dublin: ESRI. This box was prepared by Paul Egan and Kieran McQuinn. INFLATION The rate of inflation in Ireland has continued to fall, with the most recent Consumer Price Index (CPI) inflation figure for October detailing an annual rate of inflation of 0.7 per cent. The rate of growth of the Harmonised Index of Consumer Prices (HICP) index has fallen even further, to 0.5 per cent in November. HICP is lower because it excludes owner-occupied housing costs, in particular mortgage interest rates. Figure 22 shows the downward trend in both measures of inflation. 31 | Quarterly economic commentary – Winter 2024 FIGURE 22 IRISH CPI AND HICP Source: Central Statistics Office and Eurostat. The rate of inflation in Ireland is lower than the EU average. Figure 23 outlines the gap between the two, which has opened over the course of 2024. Monetary policy in the eurozone will be set based on the average rate, which may result in interest rates being set at a level that is different to the rate of inflation in the Irish economy. However, with recent growth in house prices as outlined in Box B above by McQuinn and Egan (2024), a more expansionary eurozone monetary policy would not necessarily benefit the Irish housing market. It would lower the cost of finance and increase, ceteris paribus, the demand for housing. 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% Jan-23 Feb-23 Mar-23 Apr-23 May-23 Jun-23 Jul-23 Aug-23 Sep-23 Oct-23 Nov-23 Dec-23 Jan-24 Feb-24 Mar-24 Apr-24 May-24 Jun-24 Jul-24 Aug-24 Sep-24 Oct-24 Nov-24 Ireland HICP Ireland CPI Domestic and International Outlook | 32 FIGURE 23 IRISH HICP COMPARED WITH EURO AVERAGE Source: Central Statistics Office and Eurostat. Drivers of CPI inflation in Ireland Inflation has continued to trend downwards through the third quarter of 2024. Figure 24 presents developments in the contributions to CPI inflation by key sectors in 2023 and 2024. We can observe three issues by comparing the first half of 2023, when inflation was high, with the recent low inflation period of 2024. 0% 1% 2% 3% 4% 5% 6% 7% 8% 9% 10% Jan-23 Feb-23 Mar-23 Apr-23 May-23 Jun-23 Jul-23 Aug-23 Sep-23 Oct-23 Nov-23 Dec-23 Jan-24 Feb-24 Mar-24 Apr-24 May-24 Jun-24 Jul-24 Aug-24 Sep-24 Oct-24 Nov-24 Ireland HICP Euro area HICP 33 | Quarterly economic commentary – Winter 2024 FIGURE 24 WEIGHTED CPI DEVELOPMENT Source: Central Statistics Office and authors’ calculation. First, inflation in the ‘restaurants and hotels’ sector has fallen at a much slower rate than other sectors. Note that the chart above presents the contribution of the different sectors to the overall rate. Each sector has a rate of inflation that is weighted and combined into the headline rate plotted by the dashed black line. The actual rate of inflation in the ‘restaurants and hotels’ sector averaged 8 per cent in 2023. While it has fallen to an average of 5 per cent in 2024 to date, it remains the largest contributor to inflation in each month of the year. Second, in early 2023 the ‘housing and energy’ sector of the CPI was responsible for over half of overall inflation. This element is now experiencing negative price growth. The reversal has been central to developments in overall inflation in Ireland, particularly through second round effects. Figure 25 demonstrates that the CPI has been driven by lower energy prices, while housing costs have continued to increase. 15 15 ‘Housing’ consists of rents, mortgage interest, maintenance and repair, and water supply. ‘Energy’ consists of liquid fuels, electricity, solid fuels and gas. 7.7% 8.6% 7.6%7.2%6.6% 6.1%5.9% 6.4%6.4% 5.0% 3.9% 4.6%4.1% 3.4%2.9%2.6%2.6%2.2%2.2% 1.7% 0.7% 0.7% -2% 0% 2% 4% 6% 8% 10% Food and non-alcoholic beverages Alcoholic beverages and tobacco Housing and energy Health Transport Recreation and culture Restaurants and hotels Other CPI Domestic and International Outlook | 34 FIGURE 25 HOUSING AND ENERGY Source: Central Statistics Office and authors’ calculation. Third, trends in price developments by sector have tended to be quite consistent on a month to month basis, with the exception of the ‘transport’ sector, which appears to be more volatile than the others. For example, the ‘recreation and culture’, ‘health’ and ‘alcoholic beverages and tobacco’ sectors have been quite consistent in contributing small positive amounts to overall inflation. ‘Housing and energy’ and ‘food and non-alcoholic beverages’ have declined over time but the decline has been gradual. Figure 26 highlights developments in the rate of inflation in the ‘transport’ sector. The three most heavily weighted sub-components in this sector are purchases of motor cars, air fares and ‘fuels and lubricants for personal transport equipment’, which consists primarily of petrol and diesel. Purchases of motor cars have seen a gradual decline in the rate of price increases but air fares and prices for motor fuels have been quite volatile. -10% -5% 0% 5% 10% 15% 20% 25% 30% Jan-23 Feb-23 Mar-23 Apr-23 May-23 Jun-23 Jul-23 Aug-23 Sep-23 Oct-23 Nov-23 Dec-23 Jan-24 Feb-24 Mar-24 Apr-24 May-24 Jun-24 Jul-24 Aug-24 Sep-24 Oct-24 Housing Energy CPI Housing and energy 35 | Quarterly economic commentary – Winter 2024 FIGURE 26 CPI INFLATION IN THE TRANSPORT SECTOR AND LARGEST SUB-COMPONENTS Source: Central Statistics Office and authors’ calculation. Summary Inflation is trending downwards at a faster pace than previously expected. When combined with expected growth in nominal wages, this means real wages will grow to a greater degree. Falling energy prices are exerting downward pressure on the rate of inflation, while the ‘restaurants and hotels’ sector remains the largest contributor to Irish inflation. Ireland’s inflation is lower than the prevailing European rate of inflation. Overall, we expect CPI inflation in 2024 to average at 2.1 per cent and at 1 per cent in 2025. -30% -20% -10% 0% 10% 20% 30% 40% Jan-23 Feb-23 Mar-23 Apr-23 May-23 Jun-23 Jul-23 Aug-23 Sep-23 Oct-23 Nov-23 Dec-23 Jan-24 Feb-24 Mar-24 Apr-24 May-24 Jun-24 Jul-24 Aug-24 Sep-24 Oct-24 Transport Motor cars Fuels and lubricants for personal transport equipment Passenger transport by air Domestic and International Outlook | 42 FIGURE 33 CORPORATION TAX RECEIPTS BY MONTH (€, THOUSANDS) Source: Department of Finance and authors’ calculations. Note: ‘CJEU Oct Nov’ refers to an estimate of the funds received in October and November resulting from the CJEU judgement. 2024 does not include figures for December. However, the outlook for future corporation tax receipts is particularly uncertain given the likely stance of the incoming US administration. Two policy dimensions could affect the Irish public finances. First, the Trump campaign signalled its intention to reshore to the US profits arising from intellectual property that is located in Ireland. If this were done, it could have a significant impact on future corporation tax receipts. Second, aggressive US trade policy could affect decision making in large multinationals. Ireland’s corporation tax receipts are heavily dependent on a small number of firms. Figure 34 presents corporation tax receipts by sector and highlights the exposure to ICT manufacturing and pharma manufacturing. In these sectors, there may be a higher risk of relocation because firms with complex, globalised manufacturing processes could be liable for transatlantic tariffs more than once for a given product. -5,000,000 0 5,000,000 10,000,000 15,000,000 20,000,000 25,000,000 30,000,000 35,000,000 40,000,000 2020 2021 2022 2023 2024 January February March April May June July August September October November CJEU estimate Oct Nov December 43 | Quarterly economic commentary – Winter 2024 FIGURE 34 CORPORATION TAX BY SECTOR (€, MILLIONS) Source: Revenue Commissioner Corporate Tax Analysis, 2022–2024. Headline and adjusted surpluses The funds resulting from the CJEU judgement also impact on the general government balance (GGB). Table 9 presents our forecast for 2024 and 2025 for the headline figures, and an adjusted balance for 2024 excluding the one-off receipt of these funds. 19 In the case of the adjusted balance we subtract the €14bn from government revenues. This means that the adjusted GGB balance in 2024 would have been €9,418m or 1.9 per cent of GDP, whereas the actual, headline figure is €23,418m or 4.6 per cent of GDP. 19 Note: This adjusted surplus makes no comment on the windfall nature of recent corporation tax receipts. It simply adjusts for the funds received resulting from the CJEU judgement. 0 5,000 10,000 15,000 20,000 25,000 30,000 2021 2022 2023 Chemical and pharma manufacturing ICT manufacturing Other manufacturing Information and communication Financial and insurance Wholesale and retail trade Administrative and support services Other Domestic and International Outlook | 44 TABLE 9 HEADLINE AND ADJUSTED GENERAL GOVERNMENT BALANCE (€, MILLIONS) 2024 2024 adjusted 2025 Revenue 148,753 134,753 140,653 Taxes 116,793 113,150 106,083 Social contributions 22,745 22,745 25,475 Investment income 2,005 2,005 1,965 Other 7,210 7,210 7,130 Expenditure 125,335 125,335 131,290 General government balance €, million 23,418 9,418 9,363 % of GDP 4.6% 1.9% 1.8% Contributions to investment funds 4,050 4,050 6,080 Source: Authors’ calculations. Note: We assume that all of the CJEU funds will be received in 2024. We assume in line with Department of Finance projections that there will be contributions to investment funds of €4.05bn in 2024 and €6.08bn in 2025, consisting of €4.08bn to the Future Ireland Fund and €2bn to the Infrastructure, Climate and Nature Fund. Expenditure Budget 2025 outlines increases in current and capital spending in the year ahead. A Special Article published with the Commentary discusses the distributional effect of the tax, welfare and expenditure decisions taken in the Budget (Doorley et al., 2024). Capital expenditure is currently growing faster than current expenditure, likely reflecting a catch-up period following low investment in the years following the global financial crisis. As a result, capital expenditure is increasing as a share of overall expenditure (Figure 35). 45 | Quarterly economic commentary – Winter 2024 FIGURE 35 GROSS VOTED CURRENT AND CAPITAL EXPENDITURE (€, MILLIONS) Source: Department of Public Expenditure, NDP Delivery and Reform Databank. Increased capital expenditure raises the question of where the expenditure will be directed. Figure 36 compares the areas into which budgeted capital expenditure for 2025 will be directed with the equivalent figures for recent years. The lefthand panel highlights the increase in capital spending on housing. The righthand panel shows the percentage of capital spending allocated to each area. The share of total capital spending that is spent on housing and the environment will increase in 2025 relative to the average over the last four years. 0 20,000 40,000 60,000 80,000 100,000 120,000 2022 2023 2024 2025 Current Capital Domestic and International Outlook | 46 FIGURE 36 GROSS VOTED CAPITAL EXPENDITURE BY DEPARTMENT (€, MILLIONS) Source: Department of Public Expenditure, NDP Delivery and Reform Databank. Investment funds and pro-cyclical capital expenditure Figure 37 presents a long-term perspective on the ratio of capital expenditure to total expenditure. The relative prioritisation of capital expenditure has largely depended on the strength of the economy. FIGURE 37 RATIO OF CAPITAL EXPENDITURE TO TOTAL GROSS VOTED EXPENDITURE Source: Department of Public Expenditure, NDP Delivery and Reform Databank and authors’ calculations. Previous editions of the Commentary have welcomed the establishment of two new state investment funds in the hope that the existence of such funds during future downturns will lead to public spending, in particular capital spending, being less pro-cyclical in future. In addition, the funds should mitigate against excessive public spending in light of the windfall nature of recent corporation tax receipts. 0 2,000 4,000 6,000 8,000 10,000 12,000 14,000 16,000 2021 2022 2023 2024 2025 Education Health Transport Housing Environment Other 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 2021-2024 Average 2025 Education Health Transport Housing Environment Other 0 0.02 0.04 0.06 0.08 0.1 0.12 0.14 0.16 0.18 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 47 | Quarterly economic commentary – Winter 2024 However, the extent to which the investment funds can meet their stated goals will largely be determined by the eventual size of the contributions to the funds. Figure 38 presents the intended contributions to the funds. 20 Figure 38 highlights the gap between the initial contributions made in 2023 and 2024, and the total contributions envisioned over the lifetime of the funds. The amount currently in the funds is less than the budgeted capital expenditure for 2025 alone. The long-term success of the funds will be determined by the commitment of future governments to persist with the initial commitment and the ability of the economy to continue to generate government surpluses. FIGURE 38 EVOLUTION OF INVESTMENT FUNDS (€, MILLIONS) Source: Authors’ calculations. Note: Department of Finance projections used for contributions to the Future Ireland Fund out to 2030. Conservative assumption applied of no further increases to the annual contribution after 2030. Debt to output ratios declining As illustrated in Figure 39, the debt-to-output ratio has decreased when measured against both GDP and GNI*. Both measures of output have been increasing, while gross general government debt is forecast to continue to decrease into 2025. In addition, the GGB has been in surplus in 2024 and will continue to be so in 2025. While the existence of this surplus is attributable to windfall corporation tax receipts, the effect on the debt-to-output ratio is evident. 20 This analysis is focused on contributions to the funds. It excludes return on investments made by the funds, or potential drawdowns of the Infrastructure, Climate and Nature Fund between 2026 and 2030. 0 10,000 20,000 30,000 40,000 50,000 60,000 70,000 80,000 Future Ireland Fund Infrastructure, Climate and Nature Fund Infrastructure, Climate and Nature Fund (already committed) Future Ireland Fund (already committed) Domestic and International Outlook | 48 FIGURE 39 DEBT-TO-OUTPUT RATIO TREND AND FORECAST Source: Central Statistics Office and authors’ calculations. The National Treasury Management Agency (NTMA) has stated that Irish debt has one of the longest weighted average maturities of all European countries, with only modest redemptions expected in the short term. 21 They also point out that Irish borrowing was higher during the low interest rate period of 2014–2021 than it has been in the higher interest rate period of 2022–2024. Summary In summary, while the funds received following the CJEU judgement will distort the overall picture for 2024, the underlying public finances appear to be robust. We expect the robust health of the public finances to continue into 2025. The GBB is forecasted to be 6.5 and 1.1 per cent of GNI* during these years. We expect this to contribute to a reduction in the debt-to-GNI* ratio to 67.5 per cent by the end of 2025. 21 NTMA (2024). ‘Institutional investor presentation: Annual report and financial statements’, October. 0 20 40 60 80 100 120 140 160 180 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 Debt as % GDP Debt as % GNI* 49 | Quarterly economic commentary – Winter 2024 General assessment Current expected outlook – Real income set to increase As we approach the end of the year, the domestic Irish economy looks set to register another strong performance in 2024. While the underlying economy continued to perform robustly throughout the year, as denoted by modified domestic demand (MDD), headline indicators such as GDP indicated negative growth for most of 2024. Overall we expect GDP to decline in 2024 as investment outflows of R&D capital and higher imports outweigh a recovery in exports. In 2024, we expect MDD to grow by 3.2 per cent with GDP decreasing by 1.1 per cent. The latest Nowcast estimate in the Commentary (see Egan and Kren, 2024, for details) 22 shows that MDD is currently growing at 3 per cent. In 2025, mainly because of the anticipated decline in inflation, real incomes are set to grow significantly in the domestic economy. Furthermore, we expect a rebound in residential construction activity. Accordingly, we believe MDD will grow at a slightly elevated rate of 4.1 per cent next year, with GDP increasing by 4.5 per cent. This however is framed against the backdrop of a ‘business as usual’ set of trading relationships in the global economy. Below we discuss the possible implications if the incoming US administration were to adopt some of the policies proposed during the presidential election campaign there. Given the strong pace of growth in the domestic economy, there are now likely to be 55,000 additional workers employed in 2024 compared with 2023. In 2025 we expect employment in the economy to exceed 2.8 million for the first time in the country’s history. We expect the unemployment rate to fall to 4.2 per cent next year. Forthcoming change of administration in the United States As speculated in the previous Commentary, the re-election of Donald Trump as president of the United States brings with it considerable uncertainty, particularly in terms of the macroeconomic implications of some of the trade policies that have been proposed. For example, the possible introduction of trade tariffs by the new administration would likely provoke retaliatory measures from China and other countries, which would have direct consequences for global trade and, consequently, a small open economy such as that of Ireland. A related but separate possible impact of a second Trump presidency on our domestic economy could be vis-à-vis inward FDI in Ireland by US companies. In a box accompanying the Commentary, Fitzgerald highlights the significant role in the Irish economy played by US multinationals in terms of their contributions via 22 See https://www.esri.ie/news/esri-nowcast. General Assessment | 50 corporation tax and wages. Indeed, over the past number of years, the Commentary has stressed the significant contribution to Irish economic growth made by the ICT and pharmaceutical sectors in particular. Any changes to US tax legislation that sees a major reshoring to the US of profits arising from intellectual property that is located in Ireland could have a serious impact on Irish corporation tax revenue. Because of the importance of this revenue for the Irish economy, such a shock could have a very major and lasting impact on the economy and particularly on the public finances. Previous Commentaries have highlighted the vulnerability of corporation tax receipts to a sudden fall in the ‘windfall’ component of this taxation source. The substantial employment in the domestic pharmaceutical sector could be adversely affected if production were shifted to the US as a result of increased tariffs. This is highlighted by the high value of exports of pharmaceuticals to the US. However, if similar tariffs were imposed in the EU on US exports, there could be some increase in production in Ireland by the same firms to supply a non-US market. This could partially, or even fully, offset the effects of reshoring or pharmaceutical production to the US. Indeed, given the long-term nature of many of the investments made by US pharmaceutical firms in Ireland, and multitude of factors that would have informed those investment decisions (such as EU single market access), it is unclear as to how impactful the change in policy direction could be in this sector, in the short and medium term. It is arguable that employment and wages in the IT and professional services sector would be less vulnerable to increased trade tensions between the US and the EU because most of the services of US multinationals in these sectors are provided to countries outside the US. However, they could be exposed to policies that target US-owned companies, for example in Asia, even though the services being provided globally are sourced from US firms headquartered in Ireland. Budget 2025 The paper published with the Commentary by Doorley et al. (2024) outlines the customary analysis of Budget 2025 by the tax, welfare and pensions team in the Institute. Budget 2025 saw a substantial overall total expenditure package of €10.5 billion. The income tax measures implemented include increases in the standard rate band and tax credits, along with a reduction in universal social charge (USC) liabilities. Some of the welfare measures introduced include increases in personal rates of payments for social welfare schemes, with proportionate increases to qualified adult increases. Weekly payments for child dependants rose and a new ‘Newborn Baby Grant’ of €280 was introduced, along with increase in payments received by carers. As well as Budget 2024, Budget 2025 witnessed further temporary 51 | Quarterly economic commentary – Winter 2024 measures aimed at assisting with ongoing cost-of-living pressures. Energy credits, for example, were implemented, although at a lower rate than in 2024. Doorley et al. (2024) conclude that the permanent measures in Budget 2025 are broadly progressive, with households in the bottom quintile of income expected to see a minor increase in equivalised disposable income. When temporary measures are included, the broadly progressive effect of the permanent measures becomes less clear. While households in the bottom decile of income see the largest relative rise in income of 0.5 per cent, the remainder of the bottom half of the income distribution see either no significant change in income or, in the case of the third decile, a reduction in real income of -0.4 per cent. Finally, Doorley et al. (2024) note that, as with the measures in Budget 2024, it is evident that were it not for the temporary measures in place, the at risk of poverty rate of these groups would have risen more substantially in 2024 and 2025. Therefore, careful consideration must be given to how the permanent welfare system can be developed to ensure that when these temporary measures are withdrawn, lower income groups are not particularly affected. Overall, in the Budget the commitment to increased investment and the further deployment of resources to the investment funds established is welcome. However, the Budget did contain measures that were not particularly well targeted and appear to have been overly generous in nature. Given the likely emergence of significant global trade uncertainty in 2025, there is an even greater requirement for the State’s finances to be prudently managed by any new government put in place. It is imperative that the increased expenditure enabled by the relatively buoyant state of the government coffers must be accompanied by a policy of achieving value for money, and the efficient and effective delivery of large infrastructural projects. Also, given the scale of potentially adverse economic implications that may occur due to the incoming US administration, it would be prudent to put some contingency plan in operation. For example, if it becomes apparent that there is going to be a significant impact on both multinational activity and corporation tax receipts in the domestic economy over the coming years, it may be necessary to re-appraise the spending commitments for future years, which have been agreed to in Budget 2025. House price developments The Commentary contains a number of items on house price developments. This is particularly appropriate given the acceleration in house price inflation through 2024. Research Note | 58 FIGURE 2 LABOUR SHARE USING GNI Sources: AMECO database and author’s calculations. Various well-documented attempts have been made to generate satisfactory measures of Irish national income. Modified GNI or GNI* removes depreciation on intellectual property and leased aircraft, as well as net factor income of redomiciled PLCs. These corrections generate a measure that more accurately captures the total income available to fund consumption or investment in Ireland. FIGURE 3 LABOUR SHARE USING GNI* FOR IRELAND AND GNI FOR PEER GROUP Sources: AMECO database, CSO national accounts database and author’s calculations. Figure 3 shows that the Irish labour share is far more stable when GNI* is used as 0.35 0.4 0.45 0.5 0.55 0.6 0.65 0.7 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Finland Netherlands Ireland Denmark Belgium Austria 0.45 0.5 0.55 0.6 0.65 0.7 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Finland Netherlands Ireland Denmark Belgium Austria Research Note | 59 the denominator. There is some limited evidence of a decline over time. On average, the Irish labour share is broadly similar to the labour share in the peer group of European countries when GNI* is used for Ireland and GNI for the peer group, although it is more volatile. Honohan (2021) describes any such comparison between GNI* and GNI as ‘a crude procedure’, but in this case it is far more informative than the GNI comparison presented in Figure 2. The volatility of the Irish labour share compared with the peer group is noteworthy. This comparison also highlights the scale of the effect of the 2008–2012 period on Irish national income and underlines the importance of considering developments in the labour share over a longer period of time to get a more accurate estimate of the concept. Fitzgerald (2020) argues in favour of using a measure of output that is net of depreciation because of the distortionary effects of depreciation on the Irish national accounts. By excluding all depreciation, Ireland is directly comparable with other countries. Figure 4 presents the labour share using net national income. 1 There is an immediately apparent level effect of approximately 10 per cent for most countries, but the dynamics over time are similar to the labour share using GNI*. In particular, there is no sharp decline in the post-2015 period. Schwellnus et al. (2017) argue that using a measure of national income net of depreciation to calculate the labour share may be more appropriate for considering income distribution. This is because it is only income net of capital consumption that is available to compensate workers and capital owners. However, he argues that gross measures of national income should be used to consider structural trends because capital consumption displays counter-cyclical behaviour. So while the labour share presented in Figure 4 is informative when considering income distribution, the comparison above of GNI* with GNI is preferable for considering structural trends. 1 Specifically, Fitzgerald (2020) argues in favour of using net national product at factor prices as a measure of output. The analysis in this Research Note uses net national income at market prices. The two differ because net national income includes indirect taxes and subsidies. Research Note | 60 FIGURE 4 LABOUR SHARE USING NET NATIONAL INCOME Source: AMECO database and author’s calculations. 3. IMPUTING THE LABOUR INCOME OF THE SELF-EMPLOYED The numerator of the labour share is a measure of total compensation paid for labour. National accounts provide an aggregate figure for compensation paid to employees. Total compensation paid for labour as a factor of production consists of this figure and a measure of labour compensation paid to the self-employed. 𝑙𝑎𝑏𝑜𝑢𝑟_𝑖𝑛𝑐𝑜𝑚𝑒 =𝐶𝑂𝑀𝑃𝐸𝑀𝑃′𝐸𝐸+𝐿𝑎𝑏𝑜𝑢𝑟𝐶𝑂𝑀𝑃𝑆𝑒𝑙𝑓𝐸𝑀𝑃 Figure 5 shows that self-employed workers in Ireland have consistently numbered over 300,000, accounting for between 14 and 20 per cent of the workforce. Although the share of total workers who are self-employed is decreasing, the size of the group underlines the importance of accurately imputing their labour income. 0.5 0.55 0.6 0.65 0.7 0.75 0.8 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Finland Netherlands Ireland Denmark Belgium Austria (1) Research Note | 61 FIGURE 5 PROPORTION OF WORKERS WHO ARE EMPLOYEES Source: CSO Labour Force Survey. Self-employed income is recorded as mixed income in the national accounts. Some of this income is attributable to the labour of the self-employed and some to capital they provide. Their labour income should therefore take the following form, where ϑ ∈ (0, 1): 𝐿𝑎𝑏𝑜𝑢𝑟𝐶𝑂𝑀𝑃𝑆𝑒𝑙𝑓𝐸𝑀𝑃 =𝜗∗𝑀𝑖𝑥𝑒𝑑_𝐼𝑛𝑐𝑜𝑚𝑒 In the figures presented in Section 2, we apply a method suggested by Gollin (2002) to impute the labour income of self-employed workers. This method has the advantage that it can be easily applied to all European countries and that it is sensitive to the number of self-employed workers. This correction assumes that total compensation per worker (earnings) is the same for employees and the selfemployed. 𝑙𝑎𝑏𝑜𝑢𝑟_𝑖𝑛𝑐𝑜𝑚𝑒_𝐴=𝐶𝑂𝑀𝑃_𝐸𝑀𝑃′𝐸𝐸+𝐿𝑆𝑒𝑙𝑓𝐸𝑀𝑃 ∗𝐶𝑂𝑀𝑃_𝐸𝑀𝑃′𝐸𝐸 𝐿𝐸𝑀𝑃′𝐸𝐸 Schwellnus et al. (2017) show that while there is no significant effect at the average level, labour shares in individual countries can be sensitive to the method used to impute the wages of the self-employed. Therefore, if we focus on trends in the Irish labour share rather than comparing the level with other countries, it is important to evaluate the appropriateness of the different methods proposed. How should the labour income of the self-employed in Ireland be imputed? 78% 79% 80% 81% 82% 83% 84% 85% 86% 87% 0 500 1,000 1,500 2,000 2,500 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 % Employees No. of workers Employees Self-employed % Employees (2) (3) Research Note | 62 3.1 Equal hourly wage or equal total earnings? Karabarbounis (2024) proposes a number of different methods for imputing the income of the self-employed that can be applied to US data, one of which is easily transferable to European data. This method uses an assumption of equal compensation per hour (hourly wage) between employees and the self-employed. 𝑙𝑎𝑏𝑜𝑢𝑟_𝑖𝑛𝑐𝑜𝑚𝑒_𝐵 =𝐶𝑂𝑀𝑃_𝐸𝑀𝑃′𝐸𝐸+ 𝐻𝑊𝑆𝑒𝑙𝑓𝐸𝑀𝑃 ∗𝐶𝑂𝑀𝑃_𝐸𝑀𝑃′𝐸𝐸 𝐻𝑊𝐸𝑀𝑃′𝐸𝐸 Cho et al. (2017) recommend imputing the income of the self-employed at a sectoral level. This is a sensible recommendation in an Irish context because of the wide sectoral variation in the proportion of workers who are self-employed, as presented in Figure 6. FIGURE 6 SECTORAL COMPOSITION OF EMPLOYMENT (2023) P – Education Q – Human health and social work activities K,L – Financial, insurance and real estate B–E – Industry I – Accommodation and food service activities G – Wholesale and retail trade H – Transport and storage M – Professional, scientific and technical activities R–U – Other NACE activities F – Construction A – Agriculture, forestry and fishing O – Public administration and defence Source: CSO Labour Force Survey. The Organisation for Economic Co-operation and Development (OECD, 2024) applies a sectoral approach, which builds on Method B (4). They assume that compensation per hour (wages) is the same for employees and the self-employed in each sector. This gives rise to the following correction to labour income, where 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% P Q K,l B-E I J G H M F R-U A O Per cent of overall workers Sector code Self-employed Employee (4) Research Note | 63 the average compensation per hour worked for employees in sector i is multiplied by the amount of hours worked by the self-employed in that sector: 𝑙𝑎𝑏𝑜𝑢𝑟_𝑖𝑛𝑐𝑜𝑚𝑒_𝐵𝑠𝑒𝑐𝑡𝑜𝑟𝑎𝑙 = 𝐶𝑂𝑀𝑃_𝐸𝑀𝑃′𝐸𝐸+∑𝐶𝑂𝑀𝑃_𝐸𝑀𝑃′𝐸𝐸𝑖 𝐻𝑊𝐸𝑀𝑃𝐸𝐸𝑖 𝑖∗ 𝐻𝑊𝑆𝑒𝑙𝑓𝐸𝑀𝑃 𝑖 However, the underlying assumption is still one of equal hourly wages. The National Economic and Social Council (NESC, 2020), in an analysis of data from EU Survey on Income and Living Conditions (EU-SILC) and the Household Budget Survey conducted by the Central Statistics Office (CSO), conclude that income for self-employed individuals is 10 per cent lower than income for employees. However, hours worked by the self-employed average 20–30 per cent higher than hours worked for employees. Taken together, this would suggest a substantial gap in hourly wages between employees and the self-employed. Therefore, an assumption of equal earnings rather than equal hourly wages seems more reasonable, albeit it is unlikely to be exactly correct. Caswell (2024) applies the OECD sectoral version of Method B (5) to UK data and concludes that an assumption of equal hourly wages ‘should be avoided unless compelling empirical evidence states otherwise’. He invokes identity (2) above to show that imputed self-employed income should not exceed the value recorded for mixed income in the national accounts, i.e. that ϑ should not exceed 1. We will apply this method as a check on applications of Methods A, B and C to Irish data. We propose an alternative method for imputing the labour income of the selfemployed. This method assumes equal earnings in each sector between employees and the self-employed. Therefore, Method C is equivalent to Method A but applied on a sector-by-sector basis. 𝑙𝑎𝑏𝑜𝑢𝑟_𝑖𝑛𝑐𝑜𝑚𝑒_𝐶 =𝐶𝑂𝑀𝑃+∑ 𝐶𝑂𝑀𝑃𝑖 𝐸𝑀𝑃𝐸𝐸𝑖 𝑖∗𝑆𝑒𝑙𝑓𝐸𝑀𝑃𝑖 Figure 7 presents estimated imputed labour income for the self-employed based on Methods A, B and C. Method A (3) and Method C (6) both assume equal total earnings, with Method C applying the assumption at a sectoral level. In the period before the global financial crisis (GFC), there is a significant difference between Method A and Method C. The two measures converged for a period, before Method C grew quicker in the post-COVID-19 period. Method B (5), which assumes equal hourly wages between the self-employed and employees, is consistently higher than Method A. This reflects the issues outlined (5) (6) Research Note | 64 above, with an assumption of equal hourly wages between employees and the selfemployed. FIGURE 7 IMPUTED TOTAL LABOUR INCOME FOR THE SELF-EMPLOYED BY METHOD (€, MILLION) Sources: CSO national accounts database, Labour Force Survey and author’s calculations. Figure 8 presents the imputed labour incomes for the self-employed compared with the figure for gross mixed income recorded in the national accounts. 2 In the earlier part of the sample, Method 2 imputes a value for the labour income of the self-employed that is a large share of the total gross mixed income recorded in the national accounts, implying a level of 𝜗 close to 1. On the other hand, Methods A and C impute values for the labour income of the self-employed that imply that around two-thirds of gross mixed income is attributed to labour. Method B implies a value of 𝜗 that is close to 1 and an overall labour share in the range of 0.55 to 0.65. This would suggest that the production technologies used by employees and the self-employed are structurally different. Karabarbounis (2024) argues against such an assumption, and in favour of assuming equal factor shares between the two groups. Method C achieves a result that is broadly in line with this assumption. Using Method C, a relatively constant proportion of gross mixed income is allocated to labour (𝜗 = 0.70 on average). 2 The mixed income series is available from 2010 onwards from the CSO’s website: in ‘CSO Institutional Sector Accounts’, under ‘Gross Operating Surplus / Mixed Income for the Household sector’. 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Method A (equal earnings) Method B (equal hourly wage) Method C (equal earnings - sectoral) Research Note | 65 FIGURE 8 IMPUTED LABOUR INCOME FOR THE SELF-EMPLOYED BY METHOD WITH MIXED INCOME (€, MILLION) Sources: CSO national accounts database, Labour Force Survey and author’s calculations. Method C achieves this result without directly imposing an assumption for 𝜗. Directly imposing an assumption for 𝜗 would generate an imputed value for labour compensation of the self-employed that does not take account of the number of self-employed workers. Caswell (2024) describes such an assumption as ‘somewhat naïve’. This further supports the use of Method C, which assumes equal earnings between the two groups at the sectoral level. Figure 9 presents the labour share estimated using the three different methods. Following the discussion in Section 2, we use GNI* as the denominator. The three methods have converged to a certain degree in recent years. As shown in Figure 5, the share of workers who are self-employed has fallen over time, so different methods to impute their labour income will affect the overall labour share less in recent years. 0 5,000 10,000 15,000 20,000 25,000 30,000 35,000 40,000 45,000 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 Mixed income Method A (equal earnings) Method B (equal hourly wage) Method C (equal earnings - sectoral) Research Note | 66 FIGURE 9 LABOUR SHARE BY METHOD USING GNI* Sources: CSO national accounts database, the National Farm Survey, the Labour Force Survey and author’s calculations. The GFC period represents a clear deviation from the trend regardless of the choice of method. This may be indicative of nominal rigidities in the economy. During the recessionary period, output and income fell quite quickly but wages did not adjust at the same speed. This contributed to a higher than usual labour share during this period. This episode underlines the importance of taking a long-term perspective on the labour share. 4. SECTORAL ASSUMPTIONS The new method proposed addresses the differences across sectors in the share of total workers who are self-employed. There is also a substantial difference in average employee earnings across sectors. Figure 10 highlights these differences, which translate into different assumptions in Method C for the earnings of the selfemployed by sector. 0.45 0.5 0.55 0.6 0.65 0.7 Method A (equal earnings) Method B (equal hourly wage) Method C (equal earnings - sectoral) Research Note | 67 FIGURE 10 AVERAGE EMPLOYEE COMPENSATION BY SECTOR Source: CSO Labour Force Survey. ‘Agriculture, forestry and fishing’ is an outlier with respect to the proportion of workers who are self-employed. Figure 6 above shows that it is the only sector where self-employed workers make up the majority of total workers. In addition, it is the largest single NACE sector for self-employment. This suggests that it warrants specific attention. Figure 11 shows that while the share of those in selfemployment who work in ‘agriculture, forestry and fishing’ is declining, the absolute number remains sizeable (70,000 self-employed). 0 10 20 30 40 50 60 70 80 90 100 1998-2007 2008-2017 2018-2023 Average employee compensation (€, k) Agriculture, forestry and fishing All sectors Arts, entertainment and other services Construction Distribution, transport, hotels and restaurants Financial, insurance and real estate activities Industry (excl. construction) Information and communication Professional, admin and support services Public, education and health Special Article | 74 1. INTRODUCTION Budget 2025 sets out an expenditure package of €10.5 billion. This additional expenditure is comprised of a package of once-off measures worth €2.2 billion, a net tax package of €1.4 billion and an expenditure package of €6.9 billion. In the context of declining inflation, rising wages, and record levels of employment in Ireland (McQuinn et al, 2024), fiscal restraint was urged by the Irish Fiscal Advisory Council, among others, amid concerns that further breaches of the National Spending Rule could contribute to growing underlying deficits, increased consumer prices, and fragility in the case of future recessions (Irish Fiscal Advisory Council, 2024). Nevertheless, the Government has exceeded its own 5% spending rule with Budget 2025, citing the need to compensate households for the rising cost of living. In this Special Article, we examine the tax and welfare measures announced in Budget 2025. We begin by outlining and assessing the taxation measures in Section 2, which is followed by an examination of the social welfare measures in Section 3. Section 4 presents our analysis of the distributional impact of the combined measures using SWITCH – the Economic and Social Research Institute (ESRI) tax and benefit microsimulation model – and ITSim, the indirect tax model jointly developed by the ESRI and the Department of Finance. We also estimate the cumulative impact of tax and welfare reforms announced to date by the coalition government over the period 2021 to 2025. Section 5 concludes. 2. TAXATION MEASURES Table A1 in the appendix lists the main taxation measures announced in Budget 2025, alongside the full year cost estimated by the Department of Finance. 2.1 Income tax The income tax standard rate cut-off, the point at which the higher income tax rate of 40% begins to apply, rose by €2,000 for a single adult, from €42,000 to €44,000, and by a proportionate amount for married couples and civil partners. This represents a rise of 4.8%, substantially ahead of forecast price inflation of 1.2% and just over forecast wage growth of 4.2%. 1 This amounts to an effective tax cut, as these credits are worth more to taxpayers in real terms and a lower share of earnings will be exposed to the top 40% rate of tax. 1 See McQuinn et al. (2024) for price inflation forecasts and Department of Finance (2024) for wage inflation forecasts. Special Article | 75 Most income tax credits rose in nominal terms, and the proportional increase of around 7% was substantially above both inflation and wage forecasts for 2025. This again amounts to an effective tax cut as these credits are worth more to taxpayers in real terms. The Universal Social Charge (USC) threshold for moving from the second to the third band rate increased by 6.3% (from €25,760 to €27,382). In addition, the rate for the third income band decreased from 4% to 3%. As a result of these changes, most USC payers will benefit from a reduction in their liability. 2.2 Taxation and housing The Government also announced a range of tax measures aimed at addressing issues relating to housing. The income tax credit for private renters, introduced in Budget 2023, was increased from a maximum of €750 to a maximum of €1,000 per person per year for those eligible and living in unsupported private rental accommodation. The credit will benefit middle income households most as households need to earn enough to incur a tax liability to benefit from the credit. The Mortgage Interest Tax Credit, introduced on a temporary basis in Budget 2024, is being extended by one further year. This tax credit is available for homeowners with an outstanding mortgage balance of between €80,000 and €500,000 at the end of 2022. The relief is available only to holders of tracker and variable rate mortgages, and amounts to 20% of the increased interest paid in 2024 compared to 2022. This relief is capped at €1,250. Like the rental tax credit, this relief will mainly benefit middleand higher-income households, as there are very few households in the lowest two-fifths of the income distribution with tracker or variable rate mortgages (Byrne et al., 2023). The Minister for Finance also announced an increase in the rate of the Vacant Homes Tax, which will increase from five to seven times the basic rate of Local Property Tax for the property. This tax is a welcome supply side measure and among the recommendations of the Commission on Taxation and Welfare (2022). 2.3 Indirect tax There was a well-flagged increase to the carbon tax, which went from €56 per tonne of carbon to €63.50 per tonne. Excise duties on tobacco products also increased, amounting to an extra €1 on a packet of 20 cigarettes. Other indirect tax measures announced include the introduction of excise on e-cigarette products and the restoration of the 13.5% VAT rate on gas and electricity from May 2025. The 20% reduction to public transport fares first introduced on a temporary basis in 2022 was once again extended for the whole of 2025, thus maintaining fares at the current level. Special Article | 76 3. SOCIAL WELFARE MEASURES The Budget also included many changes to social welfare parameters alongside several temporary measures aimed at cushioning household incomes from supplyside driven inflation (Table A1 in the appendix). As part of the permanent package, personal rates of payment for social welfare schemes were increased by €12 per week, with proportionate increases to qualified adult increases. Maternity, Paternity and Parent’s Benefit rates were increased by €15 per week. Weekly payments of Child Support Payment (previously known as Increase for a Qualified Child) increased by €4 for under 12s and by €8 for those aged 12 and over, and the Working Family Payment income limits increased by €60 per week. A new Newborn Baby Grant of €280 in addition to the first month of Child Benefit was introduced for parents with children born on or after 1 December 2024. Further measures were introduced to support carers. The income disregard for Carer’s Allowance was increased to €625 for a single person and €1,250 for couples, alongside a €150 increase in the Carer’s Support Grant. Carer’s Allowance has also been added to the list of qualifying payments for the Fuel Allowance. The higher income threshold for the Fuel Allowance has been extended to those aged 66 and over, and has been increased to €524 for a single person and €1,048 for couples, which means more people will qualify. For most social welfare recipients, these increases are relatively larger than the forecast wage growth of 4.2% in 2025 (Department of Finance, 2024). However, since retirement age payments tend to be higher than working age payments in nominal terms, the undifferentiated €12 rise results in a lower percentage increase for this group. The welfare package in the Budget also included temporary welfare and universal payments to mitigate an inflation-induced strain on household finances. The universal energy credits were renewed this year, as part of the cost-of-living measures, payable in November 2024 and January 2025. However, the amount decreased from €150 to €125 and the number of payments decreased from three to two compared to 2024. This is a significant fiscal outlay, making up around onequarter of the temporary cost of living expenditure. Two double Child Benefit payments will also be made in November and December 2024. Additionally, oneoff lump sum payments for recipients of certain social welfare benefits were announced, with payment occurring during December 2024. Those in receipt of the Working Family Payment, the Disability Allowance, the Carer’s Support Grant, the Blind Pension and the Invalidity Pension will receive a €400 lump sum, while those in receipt of the Living Alone and Fuel Allowance will receive €200 and €300 respectively. A lump sum of €100 was also made to recipients of the Child Support Payment in November 2024. The usual ‘Christmas Bonus’ to recipients of long-term social welfare payments was announced. The temporary reduction in the student Special Article | 77 contribution fee for third level students, announced in Budgets 2023 and 2024, was repeated in Budget 2025. 4. DISTRIBUTIONAL IMPACT ANALYSIS We use SWITCH – the ESRI’s tax benefit microsimulation model – and ITSim – an indirect tax microsimulation tool jointly developed by researchers at the ESRI and the Department of Finance – to assess the combined impact of taxation and welfare policy changes on household income. 2 The range of policy reforms modelled is detailed in the appendix. SWITCH is linked to data from the 2022 Survey on Income and Living Conditions (SILC), the primary source of information on household incomes collected annually by the Central Statistics Office (CSO). The data is reweighted to be representative of the 2022 population (in terms of demographics, employment, income and social welfare) and uprated to reflect price and income growth between 2022 and the year of analysis. The scale, depth and diversity of this survey allows it to provide an overall picture of the impact of the policy changes on Irish households, which cannot be gained from selected example cases. ITSim estimates the indirect taxes (VAT and excise duties, including carbon taxes) paid by Irish households on the basis of their reported expenditure, collected by the CSO’s nationally representative Household Budget Survey (HBS) in 2015–2016. 3 Given the range of temporary and permanent measures announced as part of Budget 2025, we separate base and reform scenarios to estimate the distributional effect of Budget 2025. These are summarised in Panel A of Table 1. Scenario 1 captures permanent policy changes between 2024 and 2025, while Scenario 2 outlines the effect of changes to both permanent and temporary measures between 2024 and 2025. We also set up a baseline and reform scenario (Panel B of Table 1), which presents a more medium-term picture. In Scenario 3 we estimate the effect of permanent policy changes only, between 2020 and 2025, on the distribution of income compared to a scenario in which 2020 policies were pegged to wage growth. In each case, we compare to a scenario in which policy parameters of the direct tax and welfare system are indexed in line with actual and/or forecast wage growth (Table 1). As argued by Bargain and Callan (2010) and Callan et al. (2019), this provides a distributionally neutral benchmark against which to assess policy 2 See Keane et al (2023) for a description and validation of the SWITCH model. 3 Income rates are uprated to 2025 levels using earnings indices. Expenditures are uprated to 2025 levels using price growth indices. Special Article | 78 reforms. For the indirect tax system, we index our baseline scenario in line with price growth, which is a more appropriate indexation factor for expenditure. A new feature of this year’s analysis is the inclusion of temporary measures in the baseline for the scenario that evaluates the distributional impact of both permanent and temporary measures. Government has now extended or repeated certain temporary measures for several consecutive years, and households have begun to depend on measures labelled as temporary. Therefore, the analysis accounts for the distributional impacts of the withdrawal or reduction of temporary measures in Budget 2025. We use SWITCH to calculate households’ social welfare entitlements, tax liabilities and net incomes under each system. ITSim calculates households’ VAT and excise liabilities. TABLE 1 SUMMARY OF BASELINE AND REFORM SCENARIOS A: 2024–2025 B: 2020–2025 Scenario 1 Scenario 2 Scenario 3 Base Reform Base Reform Base Reform Policy 2024 2025 2024 2025 2020 2025 Indexed to 2025 - 2025 - 2025 - Indexation factor direct tax and welfare* 4.2% - 4.2% - 22.9% - Indexation factor indirect tax* 1.2% - 1.2% - 21.4% - Temporary policies included?** No Yes No Figures 1, 3-6 2, 3-6 7 Notes: * We use CSO data on annualised quarterly average weekly earnings and the Department of Finance 2025 forecast for increase in compensation per employee to index the direct tax and welfare system. We use CSO data on CPI growth until 2024 and forecasts from the ESRI’s Quarterly Economic Commentary for CPI growth in 2025 to index the indirect tax system. ** Temporary policy measures introduced in Budget 2024 (2025) to be paid at the end of 2023 (2024) and beginning of 2025, e.g. energy credit, double social welfare payments, additional Fuel Allowance payments. 4.1 The distributional effect of Budget 2025 4 Figure 1 shows the distributional effect of permanent changes to indirect taxes, direct taxes and welfare announced as part of Budget 2025, compared to a wageindexed 2024 policy system. This corresponds to Scenario 1 in Table 1. 4 As mentioned, reference to ‘Budget 2025’ measures includes the planned increase in PRSI and introduction of the Pay-Related Benefit scheme. Special Article | 79 While not part of Budget 2025, it was announced in 2023 that pay-related social insurance (PRSI) rates will rise from October 2024 onwards to help fund the introduction of the Pay-Related Benefit scheme and tackle State Pension funding pressures. 5 Our analysis therefore includes the planned 0.1% PRSI rise for 2025 as well as the move in 2025 to pay-related benefits. 6 As shown in Figure 1, we estimate that households will experience a rise in real income of 0.5% on average in 2025 due to the permanent measures announced in Budget 2025. The permanent measures are broadly progressive, with households in the bottom quintile of income expected to see an increase of around 0.9% of equivalised disposable income, and those in the top quintile to see increases of 0.5%. We also show the effect of combined changes to the permanent and temporary tax and welfare systems in Figure 2, corresponding to Scenario 2 in Table 1. When accounting for temporary measures, the average household is estimated to see an increase of just 0.2% in their equivalised disposable income. Furthermore, the broadly progressive effect of the permanent measures seen in Figure 1 becomes less clear. While households in the bottom decile of income see the largest relative rise in income, of 0.5%, the remainder of the bottom half of the income distribution see either no significant change in income, or, in the case of the third decile, a reduction in real income of 0.4%. This negative effect is driven by the concentration of retirement aged households and households with disabilities in the third decile; these groups are most significantly impacted by the partial withdrawal or nominal freezing of temporary measures. By this measure, Budget 2025 is neither strongly progressive nor regressive: the effect of permanent measures is reasonably progressive, although real income gains are modest across the board. When factoring in temporary measures, the impacts of the budgetary package on real incomes are modest and don’t appear to be progressive. This effect is driven by the partial withdrawal of the energy credits and the nominal freeze to most other temporary measures. 5 All PRSI rates are set to increase by 0.1 percentage point in 2024 and 2025, 0.15 percentage point in 2026 and 2027 and 0.2 percentage point in 2028 (Department of Social Protection, 2023 (press release), https://www.gov.ie/en/press-release/022d7-minister-humphreys-secures-cabinet-approval-for-major-social-welfarereforms/. 6 The Pay-Related Benefit Scheme is set to be introduced in March 2025. Those who qualify will receive 60% of their previous earnings for the first three months of unemployment, dropping to 55% for the next three months and 50% for the three months following that (sixth to ninth month of unemployment). These rates are subject to maximum levels. Special Article | 80 FIGURE 1 DISTRIBUTIONAL IMPACT OF PERMANENT BUDGET 2025 COMPARED TO INDEXED 2024 POLICIES Source: Authors’ calculations using ITSim linked to the 2015–2016 Household Budget Survey uprated to 2025 prices, and SWITCH run on 2022 Survey on Income and Living Conditions data, uprated to 2025 income levels. Notes: Deciles are based on equivalised household income, using CSO national equivalence scales. FIGURE 2 DISTRIBUTIONAL IMPACT OF PERMANENT AND TEMPORARY BUDGET 2025 COMPARED TO INDEXED 2024 POLICIES Source: Authors’ calculations using ITSim linked to the 2015–2016 Household Budget Survey uprated to 2025 prices, and SWITCH run on 2022 Survey on Income and Living Conditions data, uprated to 2025 income levels. Notes: Deciles are based on equivalised household income, using CSO national equivalence scales. 4.2 The effect of Budget 2025 by household type, gender and disability status -0.40 -0.20 0.00 0.20 0.40 0.60 0.80 1.00 1.20 Decile 1 Decile 2 Decile 3 Decile 4 Decile 5 Decile 6 Decile 7 Decile 8 Decile 9 Decile 10 All % change in disposable income Decile of equivalised disposable income Direct tax and welfare Indirect tax Total -0.80 -0.60 -0.40 -0.20 0.00 0.20 0.40 0.60 0.80 1.00 1.20 Decile 1 Decile 2 Decile 3 Decile 4 Decile 5 Decile 6 Decile 7 Decile 8 Decile 9 Decile 10 All % change in disposable income Decile of equivalised disposable income Direct tax and welfare Indirect tax Total Special Article | 81 We further examine the distributional impact of Budget 2025 by household type, gender and disability status. Figure 3 displays the impact of direct tax and welfare measures and indirect tax policies of Budget 2025 by household type. On average, most household types benefit from the permanent measures in Budget 2025, especially working age couples without children and lone parent households as they are gaining from the income tax/USC measures and welfare increases that are above wage inflation. Retirement aged households were least likely to benefit from the permanent measures in Budget 2025, with single retirement aged households experiencing a marginal decline in real disposable income and retirement aged couples seeing an increase of 0.1%. This is due to two factors. The €12 increase in State Pension rates for this group represents a smaller percentage gain than increases to working age welfare payments. Secondly, additional benefits often received by this group – such as the Living Alone Allowance and the Fuel Allowance – were frozen in nominal terms, representing a real decline in value. When considering both the permanent and temporary changes to the tax and welfare system announced in Budget 2025, the distribution of gains and losses is more uneven across household types. On average, retirement aged households experience losses, while working aged households see gains, particularly households with children. The results for households with children are driven by the two double Child Benefit payment, increases in the Child Support Payments and the introduction of the Newborn Baby Bonus. The pattern for retired households reflects the withdrawal of temporary measures, such as the reduction in energy credit – highlighting the importance of these measures for such groups. Special Article | 82 FIGURE 3 DISTRIBUTIONAL IMPACT OF BUDGET 2025 BY HOUSEHOLD TYPE COMPARED TO INDEXED 2024 POLICIES Source: Authors’ calculations using ITSim linked to the 2015–2016 Household Budget Survey uprated to 2025 prices, and SWITCH run on 2022 Survey on Income and Living Conditions data, uprated to 2025 income levels. Figure 4 shows the estimated effects of direct tax and welfare policy changes from Budget 2025 by gender. 7 For this analysis, we assume that income is split evenly between individuals in a couple. Compared to a wage-adjusted budget, our analysis suggests that Budget 2025 measures affected men and women in a broadly similar manner, with women in the bottom quintiles of income seeing larger gains because of both permanent and temporary policies, likely a reflection of benefit receipt patterns and the higher Child Support Payments, and double Child Benefit payment implemented in Budget 2025. 7 It is not possible to estimate the gender impact of indirect tax changes using ITSim as expenditure data are collected at the household level. -1.50 -1.00 -0.50 0.00 0.50 1.00 1.50 2.00 Single workingage without children Lone parent Working-age couple without children Working-age couple with children Single retirement-age (>= 66) Retirement-age couple (at least one >=66) All % change in disposable income Permanent Permanent + temporary Special Article | 83 FIGURE 4 DISTRIBUTIONAL IMPACT OF BUDGET 2025 BY GENDER COMPARED TO A WAGE INDEXED 2024 POLICY Source: Authors’ calculations using SWITCH run on 2022 Survey on Income and Living Conditions data, uprated to 2025 income levels. Notes: Income is assumed to be fully shared between members of a couple. Quintiles are based on equivalised household income, using CSO national equivalence scales. Figure 5 shows the estimated effects of the Budget 2025 direct tax and welfare measures by disability status. 8 , 9 We identify households with disabilities as those in which there is at least one member who self-declares to have a medical condition that limits them in their daily activities. Overall, permanent measures were estimated to have broadly similar impacts on households with and without disabilities. However, households with disabilities gained more at the bottom end of the distribution. When additionally considering temporary measures, on average, households with disabilities experienced smaller gains compared to households without disabilities. This is largely driven by a negative impact on households with disabilities in quintile 2. Households in this quintile, many of whom are of retirement age, are more reliant on temporary measures, such as energy credits, which were reduced in Budget 2025. 8 ITSim does not currently allow the estimation of indirect taxation measures by disability status. 9 The precise definition we employ in the SILC data is to identify as having a disability those who respond positively to the following two questions: Do you have any chronic physical or mental health problem, illness or disability? Are you hampered [limited] in your daily activities by this physical or mental health problem, illness or disability? 0.00 0.20 0.40 0.60 0.80 1.00 1.20 Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 All Permanent Men Women 0.00 0.20 0.40 0.60 0.80 1.00 1.20 Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 All Permanent + temporary Men Women Special Article | 90 APPENDIX TABLE A1 REFORMS MODELLED IN DISTRIBUTIONAL ANALYSIS Taxation Full year cost/yield, €m Modelled Income tax -1,290 €125 increase to Personal, Employee, and Earned Income Tax Credits; €150 increase to Home Carer and Single Person Child Carer Tax Credits; rise in standard rate cut off by €2,000. ✔ €300 increase in the Incapacitated Child Tax Credit and Blind Person’s Tax Credit, €60 increase in the Dependent Relative Tax Credit. Universal Social Charge USC second band increase from €25,760 to €27,382, USC 4% rate reduced to 3% -540 ✔ Housing Help to Buy amendments and extension to 31 Dec 2029 -185 Increase Rent Tax Credit to €1,000 -65 ✔ Tax relief on pre-letting expenses extended to 31 Dec 2027 -2 Increase stamp duty on bulk home purchases from 10% to 15% 11.6 Increase vacant homes tax to 7 times LPT charge 1 Carbon tax +€7.50 per tonne of carbon 157 ✔ Excise duties +€1 on a packet of 20 cigarettes 69.7 ✔ Introduced e-liquid tax of €0.50 per ml 17 VAT VAT reduction on heat pumps to 9% -4 Once-off cost-of-living measures Extend mortgage interest tax relief for one further year -40 Extension of 9% VAT rate for gas and electricity to 3 Apr 25 -110 ✔ Rent Tax Credit Increase for 2024 -65 ✔ Inheritance tax Thresholds for capital acquisitions tax increased -88 Special Article | 91 TABLE A1 (CONTD.) REFORMS MODELLED IN DISTRIBUTIONAL ANALYSIS Welfare Full year cost/yield, €m Modelled General 919.6 +€12 (under 66) welfare payments, proportionate increase for qualified adults ✔ +€12 (over 66) welfare payments, proportionate increases qualified for adults ✔ +€15 parental benefits11 ✔ Child Support Payment +€4 for qualified child <12 years +€8 for qualified child >12 years 78.5 ✔ Working Family Payment +€60 per week to income thresholds 14.8 ✔ Carers Carer’s Allowance becomes qualifying payment for Fuel Allowance 3.7 ✔ Carer’s Support Grant increased from €1,850 to €2,000 25.4 ✔ Carer’s Benefit extended to self-employed 7.3 Increase in income disregard for Carer’s Allowance to €625 for singles (€1,250 for couples) 11.8 ✔ Domiciliary Care Allowance increased by €20 per month 15.9 Child Benefit Newborn Baby Grant of €280 15 ✔ Miscellaneous Means test for recipients of State Pension (noncontributory), Disability Allowance and Blind Pension – amount not considered from sale of home upon moving into care increased from €190,500 to €337,500 0.2 ✔ Free School Books Scheme extended to Leaving Certificate from Sep 2025 51 ✔ Extension of Hot School Meals to all primary schools from Apr 2025 72 ✔ Pay-related Jobseeker’s Benefit from Mar 2025 * Free Travel Scheme Companion pass extended to all people over 70 7 Free transport extended to 5–8-year-olds * Fuel Allowance means test age criteria reduced from 70 to 66; income disregard increased to €524 for a single person and €1,048 for a couple 4.8 ✔ Once-off cost-of-living measures 2x €125 household energy credits 500 ✔ 11 Modelled for Maternity Benefits only. Special Article | 92 TABLE A1 (CONTD.) REFORMS MODELLED IN DISTRIBUTIONAL ANALYSIS Welfare Full year cost/yield, €m Modelled Fuel Allowance €300 lump sum 126 ✔ Child Benefit double month x 2 371 ✔ Social Protection – autumn double week 350 ✔ Living Alone Allowance €200 lump sum 50 ✔ Working Family Payment €400 lump sum 18 ✔ Disability Allowance, Carer's Support Grant, Invalidity Pension €400 lump sum 143 ✔ Blind Pension, Domiciliary Allowance €400 lump sum €100 lump sum for Increase for a Qualified Child recipients 34 ✔ Foster Carer Allowance double payment 2 €1,000 reduction in student contribution fee, 33% reduction in contribution fee for apprentices. €1,000 increase in Post Graduate Tuition fee contribution. 98 ✔ Additional funding for Student Assistance Fund 18 Fee reduction on School Transport Scheme, State Exam Fee waiver, Additional Schools Capitation and other measures 120 Source: Department of Finance’s Budget 2025 expenditure report and Budget 2025 tax policy changes. Notes: Costs are in millions of euros per annum and are mostly full year costs for 2024. Some small schemes are excluded. Asterisk (*) indicates no costing was available. TABLE A2 SIMULATED INCOME INEQUALITY AND AROP RATES IN 2025 WITH AND WITHOUT TEMPORARY MEASURES Inequality/poverty Indexed 2024 permanent Budget 2025 permanent Indexed 2024 Permanent + temporary Budget 2025 Permanent + temporary Gini index 0.272 0.271 0.266 0.266 AROP rate Adult 0.121 0.119 0.117 0.116 Retirement age 0.159 0.178 0.088 0.130 Child 0.156 0.154 0.150 0.145 Disability 0.229 0.238 0.187 0.209 Source: Authors’ calculations using SWITCH run on 2022 Survey on Income and Living Conditions data, uprated to 2025 income levels. Notes: The poverty rate is calculated based on a poverty line equal to 60% of median equivalised disposable income. The CSO equivalence scale is used. Working age defined as aged 18-65 and children as those under age 18. People with disabilities are identified as those who self-report to having an illness or disability that limits them in their daily activities. This Article has been accepted for publication by the Institute, which does not itself take institutional policy positions. Special Articles are subject to refereeing prior to publication. The authors are solely responsible for the content and the views expressed. ESRI SPECIAL ARTICLE ASSESSING EXPECTATIONS OF EUROPEAN HOUSE PRICES Akhilesh Kumar Verma and Kieran McQuinn Available to download from www.esri.ie https://doi.org/10.26504/qec2024win_sa_verma © 2024 The Economic and Social Research Institute Whitaker Square, Sir John Rogerson’s Quay, Dublin 2 This Open Access work is licensed under a Creative Commons Attribution 4.0 International License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution and reproduction in any medium, provided the original work is properly credited. Special Article | 94 THE AUTHORS Akhilesh Kumar Verma is a Postdoctoral Research Fellow at the Economic and Social Research Institute (ESRI). Kieran McQuinn is a Research Professor at the ESRI and an Adjunct Professor at Trinity College Dublin. KEYWORDS House prices; forecasting bias; fundamental variables JEL CODES D84, G12, R21 Special Article | 95 ABSTRACT Using a new database of consumers’ expectations, this paper examines the nature of house price forecasts across a select sample of European Union (EU) member states for the period 2020 to 2024. Across many EU countries, post COVID-19, house price increases have been apparent. Therefore, understanding the dynamics of house price movements is especially important at this time. In particular, we examine the rationality or otherwise of consumers’ house price expectations, and then examine the relationship between the expectations and forecasts of key fundamental determinants of house prices, such as interest rates and income levels. In this way we distinguish our work from most other studies of house price forecasts, which have not examined links between house price forecasts themselves and forecasts of the variables typically assumed to be determining prices. This is particularly relevant as oftentimes house price expectations themselves are influenced by changes in market fundamentals. Special Article | 96 1. INTRODUCTION One of the economic legacies of COVID-19, observed across different countries, is an acceleration in house price inflation. According to the International Monetary Fund’s (IMF) Global House Price Index, 1 of the over 60 countries participating in the survey, three-quarters witnessed increases in prices in 2020 with the trend continuing into 2021. Indeed, house prices increased by over 5 per cent for 23 countries of the 60. On a cross-country basis, house prices have not experienced such a sustained increase since the period preceding the global financial crisis (GFC) of 2007–2008. Some of this increase may have been due to the accumulation in savings evident among households across European countries. Fitzgerald et al. (2021), for example, note that when European consumers were similarly rationed during the Second World War, excess savings were subsequently converted into physical assets in the housing market. Consequently, given the prominent role of the housing market in credit and asset price cycles, as well as the link between housing finance and the 2007–2009 GFC (Brunnermeier, 2009; Duca et al., 2010 and 2011), assessing the sustainability of current house price movements is of acute interest from a policymaker’s perspective. Typically, the housing literature assumes that, in the long run, house prices are determined by movements in key fundamental variables; for example, a standard approach in the literature is to adopt an inverted housing demand function such that the dependent variable is the house price, as opposed to the quantity of houses. Applications can be found in Peek and Wilcox (1991), Muellbauer and Murphy (1997), Meen (1996), Meen (2000), Cameron et al. (2006), Kelly and McQuinn (2014) and Cronin and McQuinn (2021a); in all these, the model generally assumes that house prices are positively related to income levels and negatively related to the cost of capital. 2 The interrelationship between the housing market and the real economy was particularly evident following the GFC in 2007–2008. A robust housing market often signals a strong economy, as increased home sales and rising property values boost consumer wealth, leading to higher consumer spending and investment. This, in turn, stimulates economic growth through increased demand for goods and services, construction, and related industries. Conversely, a downturn in the housing market can have a ripple effect, leading to decreased consumer confidence and spending, reduced construction activity, and potential job losses, thereby slowing economic growth. Additionally, housing prices and availability 1 See https://blogs.imf.org/2021/10/18/housing-prices-continue-to-soar-in-many-countries-around-the-world/. 2 House prices are also generally assumed to be negatively related to the per capita housing stock. Special Article | 97 impact affordability and mobility, influencing labour market dynamics and overall economic productivity. Case et al. (2005), for example, find that increases in housing wealth significantly boost consumer spending, more so than increases in stock market wealth, while Leamer (2007; 2015) argues that housing is a leading indicator of business cycles, with downturns in residential investment often preceding broader economic recessions. The construction sector, which is closely tied to housing investment, generates significant employment and drives demand for materials and services, thus magnifying its impact on the economy. Mian and Sufi (2009) demonstrate how housing market collapses can lead to severe banking crises, as falling home prices reduce collateral values, leading to a credit crunch. They emphasise the role of excessive mortgage lending and financial leverage in exacerbating economic downturns. While increases in house prices can initially arise due to a particular shock or change in a key economic variable, consumer expectations of future price movements can themselves become an important dynamic in the market. As noted by Duca et al. (2021), house price booms are usually: set in motion by shifts in fundamentals (e.g. in interest rates, income and credit standards) whose dynamic effects interact with supply conditions and can be magnified by a tendency for households to form house price expectations that are very different from the rational expectations associated with efficient markets. Therefore, the interaction of house price expectations and the fundamental determinants of house prices is of particular interest – i.e., to what extent are consumers’ expectations of house prices linked to or associated with their expectations for the key determinants of house prices? To date, however, this relationship does not appear to have been examined in much detail. Much of the literature is concerned with examining the rationality or otherwise of house price forecasts, and when this hypothesis is usually rejected, other dynamics underpinning house price expectations are considered, such as backward looking, extrapolative ones. Few, if any, studies examine the degree to which house price expectations are influenced by consumers’ expectations of key underlying fundamental variables, such as income levels and interest rates. Using new survey data on consumer sentiment regarding this issue, which was collated and published by the European Central Bank (ECB) 3 , we now address this issue. Among other variables, the ECB survey publishes consumers’ expectations for house prices, household income and mortgage interest rates. This allows us to compare the forecast errors for house prices with those of the key fundamental 3 For more information on the survey, see https://www.ecb.europa.eu/stats/ecb_surveys/consumer_exp_survey/html/index.en.html. Special Article | 98 determinants of house prices in the short run; namely, income levels and interest rates. Rational expectations theory has significant implications for economic forecasts, as it posits that individuals and firms form their expectations about future economic variables (such as inflation, interest rates and output) based on all available information, including the likely effects of current policies. This means that if economic agents are rational and markets are efficient, forecasting errors should be random rather than systematic, since individuals would already anticipate the predictable effects of policies or trends. Consequently, policymakers may find it difficult to impact the economy through monetary or fiscal policies if these actions are anticipated. It also implies that traditional macroeconomic models that do not account for rational expectations may overestimate the effectiveness of policy interventions, as individuals will adjust their behaviour to offset the anticipated effects. Our results reveal that, in accordance with the previous literature, we can reject the null hypothesis of rationality among European households in terms of house price expectations. This is important in terms of the mechanisms regarding house price expectations that are adopted in different housing, and more broadly macroeconomic, models. In turn this can have implications for conclusions reached about the presence of housing bubbles or periods of irrationality among consumers in terms of their attitudes to house price developments. We also find an important distinction between the role played by actual and expected changes in real interest rates. It appears that households are more influenced by expected changes in real interest rates than in changes in the actual rate. Finally, our results suggest that variations in changes in real income expectations have significant implications for expectations regarding real house price changes. Therefore, it would appear that it is consumers’ expectations concerning the general economy that is of most importance in shaping their beliefs about the future housing market. The rest of the paper is structured as follows. In the Section 2, we review the literature on house price expectations. The data and empirical methodology adopted are then discussed in the Section 3, following which the results of our analysis are presented, in Section 4. Section 5 offers some concluding comments. 2. LITERATURE REVIEW Given the importance of consumers’ price expectations mechanisms, in their review paper Duca et al. (2021) contend that this ‘suggests that regular surveys of house price expectations should have been a high priority before the boom and bust of the mid-2000s, [yet] surveys are sparse and intermittent’. Kuchler et al. (2022) provide a summary of studies of house price expectations. In the case of the Special Article | 99 US residential market, while the Michigan Survey of Consumers has provided information on the housing market since 1960, data on point estimates for house price expectations have only been available since 2007. Case and Shiller (1989), based on their 1988 survey, contended that people seemed to base their expectations on house prices on past house price movements rather than expectations of key market fundamentals. In 2013, in light of the financial crisis, the Federal Reserve Bank of New York launched the monthly Survey of Consumer Expectations (SCE), which every month contains questions on respondents’ expectations of house prices. Fannie Mae’s National Housing Survey (NHS) has surveyed US households since 2010 on expectations about housing markets. In addition, between 2003 and 2012, Case et al. (2012) conducted surveys of recent home buyers in four US counties that experienced significant price appreciation prior to 2008. On a European wide basis, information on house price expectations, income levels, interest rates and credit standards has only become available with the initiation by the ECB of the Consumer Expectations Survey (CES) with data available from 2020. Eurozone member central banks, such as the Bundesbank, the Bank of Spain and the Bank of Italy, conduct country-level surveys, which include questions on the housing market. However, the CES is the only survey available for a number of Eurozone countries. The CES is an online panel survey of consumers, and is carried out on a monthly basis. The microdata for the CES are collected through a survey of a panel of eurozone consumers, which is currently conducted by Ipsos Public Affairs on behalf of the ECB. The countries included since the beginning of the survey are: Belgium, France, Germany, Italy, the Netherlands and Spain. In 2022, the sample was extended to cover five additional countries: Austria, Finland, Greece, Ireland and Portugal. In comparing variations in the house price expectations of the Michigan Surveys of Consumers and actual house price movements, Kuchler et al. (2022) note two stylized facts among the data: house price expectations tend to be more optimistic after recent periods of actual house price appreciation; and the time-series variation in expectations is actually smaller than the time-series variation in the movement of actual prices. One of the earlier assessments of the role played by house price expectations was conducted by Abraham and Hendershott (1996). Here, these authors outline the manner in which expectations can interact with market fundamentals. In the context of an equilibrium correction model, they also discuss the concept of positive ‘bubble-builder’ effects on house prices – from recent rises in house prices – and negative ‘bubble-burster’ effects – from high levels of real house prices relative to fundamentals. The bubble-builder effect arises if many agents base their Special Article | 106 to 1.07 and 1.11, respectively, maintaining their statistical significance at the 1 per cent level. However, a significant contrast emerges when considering the influence of real interest rates. In the regression for actual house price growth, the real interest rate has a negative but statistically insignificant effect. In contrast, for expected house price growth, the expected real interest rate has a significantly negative impact, with coefficients of -0.57 and -0.65 where both are statistically significant. This suggests that while actual house prices may not respond immediately to changes in real interest rates, market expectations of future house prices are more sensitive to anticipated changes in interest rates. 10 TABLE 2 EXPECTED HOUSE PRICE GROWTH AND DETERMINANTS Dependent variable: ∆𝑃𝑖𝑡 𝐸. Variable (1) (2) (3) ∆𝑌𝑖𝑡 𝐸 0.61*** 1.07*** 1.11*** (0.07) (0.18) (0.20) ∆𝑅𝑖𝑡 𝐸 -0.57*** -0.65*** (0.20) (0.23) ∆𝑆𝑖𝑡 𝐸 0.01 (0.01) Country fixed effect Yes yes yes R-squared 0.46 0.49 0.50 No of obs. 108.00 102.00 89.00 Note: The table reports the regression estimates, where the dependent variable is house price growth (∆𝑃𝑖𝑡) 𝐸 and the explanatory variables are household income growth (∆𝑌𝑖𝑡 𝐸), real interest rate (∆𝑅𝑖𝑡 𝐸) and house supply growth. 4.3 Rationality tests We now move to analyse the rationality of house price expectations and that of its key determinants – real income and interest rate. Tables 3, 4 and 5 present the rationality tests for house price, income and interest rate, respectively. For each case, we estimate two models, where column (1) describes the estimation result for a panel setup with all countries combined and column (2) captures crosscountry variation. In other words, the former tests for the rationality of house price, income and interest rates for Europe as a whole, while the latter tests for the rationality for each country in our sample separately. In column (1) of Tables 3, 10 We also conduct a robustness check in this case, using an alternative estimation strategy – dynamic panel GMM – and find similar results (please refer Table A2 in the appendix for the results). Special Article | 107 4 and 5, we test whether the mean estimate of expected house prices, expected interest rates and expected income are significantly different from 1. Furthermore, in column (2) of each table, we test whether the sum of the mean estimate of the variable of interest (expected house prices, expected interest rate and expected income) and the corresponding country level estimate are statistically different from 1. For instance, in Table 3, we reject the rationality of Belgium’s house prices, as the sum of the coefficient of mean estimate (∆𝑃𝑖𝑡 𝐸), 5.22, and country-level estimate BE × ∆𝑃𝑖𝑡 𝐸, -1.75, is statistically different from 1. Table 3 shows that the coefficient on expected house price growth is 2.49 for the panel setup and 5.22 for the crosscountry variation case. The coefficient at the country level in column (2) is such that the combined coefficient of expected house price growth and the respective country’s expected house price growth is different from 1, which confirms the rejection of the rationality hypothesis for house prices. TABLE 3 RATIONALITY TEST: HOUSE PRICE GROWTH Dependent variable: ∆𝑃𝑖𝑡. (Combined panel) (Cross-country variation) Variable (1) (2) ∆𝑃𝑖𝑡 𝐸 2.49*** 5.22** (0.31) (2.37) BE × ∆𝑃𝑖𝑡 𝐸 -1.75 (2.57) DE × ∆𝑃𝑖𝑡 𝐸 0.91 (2.47) ES × ∆𝑃𝑖𝑡 𝐸 -5.31* (2.82) FI × ∆𝑃𝑖𝑡 𝐸 -5.33** (2.66) Special Article | 108 FR × ∆𝑃𝑖𝑡 𝐸 -1.98 (2.57) IE × ∆𝑃𝑖𝑡 𝐸 -6.47** (2.74) IT × ∆𝑃𝑖𝑡 𝐸 -4.27* (2.44) NL × ∆𝑃𝑖𝑡 𝐸 -2.40 (2.40) PT × ∆𝑃𝑖𝑡 𝐸 -5.37** (2.56) Constant 6.46*** 3.74*** (0.95) (1.00) Country fixed effect yes yes R-squared 0.40 0.64 No of obs. 108.00 108.00 Note: The table reports the regression estimates, where the dependent variable is house price growth (∆𝑃𝑖𝑡) and the key explanatory variable is expected house income growth (∆𝑃𝑖𝑡 𝐸). The interaction term refers to cross-country variation in the relationship between expected house price and house price growth. Heteroskedasticity robust standard errors are reported in parentheses. Asterisks (∗∗∗, ∗∗ and ∗) denote statistical significance at 1, 5 and 10 per cent levels. We find similar results in the case of real income as shown in Table 4. The coefficient on expected house price growth is 1.90 for the panel setup and -0.62 for where cross-country variation is allowed for. Moreover, the coefficient at the country level is such that the combined coefficient of the expected house price growth and the respective country’s expected house price growth is significantly different from 1. TABLE 4 RATIONALITY TEST: REAL INCOME Dependent variable: ∆𝑌𝑖𝑡. Variable (Combined panel) (Cross-country variation) ∆𝑌𝑖𝑡 𝐸 1.90*** -0.62*** (0.23) (0.00) Special Article | 109 BE × ∆𝑌𝑖𝑡 𝐸 6.84*** (0.00) DE × ∆𝑌𝑖𝑡 𝐸 3.27*** (0.00) ES × ∆𝑌𝑖𝑡 𝐸 4.76*** (0.00) FI × ∆𝑌𝑖𝑡 𝐸 2.44*** (0.00) FR × ∆𝑌𝑖𝑡 𝐸 3.05*** (0.00) IE × ∆𝑌𝑖𝑡 𝐸 1.24*** (0.00) IT × ∆𝑌𝑖𝑡 𝐸 2.62*** (0.00) NL × ∆𝑌𝑖𝑡 𝐸 3.91*** (0.00) PT × ∆𝑌𝑖𝑡 𝐸 2.12*** (0.00) Constant 4.19*** 7.11*** (1.17) (0.00) Country fixed effect yes yes R-squared 0.42 0.67 No of obs. 108.00 108.00 Note: The table reports the regression estimates, where the dependent variable is household real income growth (∆𝑌𝑖𝑡 ) and the key explanatory variable is expected household real income growth (∆𝑌𝑖𝑡 𝐸). The interaction term refers to cross-country variation in the relationship between expected household real income growth and household real income growth. Heteroskedasticity robust standard errors are reported in parentheses. Asterisks (∗∗∗, ∗∗, ∗) denote statistical significance at 1, 5 and 10 per cent levels. Lastly, drawing from Table 5, we reject the rationality hypothesis in the case of the real interest rate. Based on this, we consistently reject the rationality hypothesis across both model specifications for house prices, interest rates and income, as we Special Article | 110 find the coefficient of expected house prices, expected income and expected real interest rate is statistically significant and different from 1 on average and in the case of each individual country. TABLE 5 RATIONALITY TEST: REAL INTEREST RATE Dependent variable: ∆𝑅𝑖𝑡. Variable (Combined panel) (Cross-country variation) ∆𝑅𝑖𝑡 𝐸 1.67*** 0.63** (0.14) (0.26) BE × ∆𝑅𝑖𝑡 𝐸 2.83*** (0.47) DE × ∆𝑅𝑖𝑡 𝐸 1.59*** (0.49) ES × ∆𝑅𝑖𝑡 𝐸 3.01*** (0.54) FI × ∆𝑅𝑖𝑡 𝐸 0.51 (0.39) FR × ∆𝑅𝑖𝑡 𝐸 1.51** (0.62) IE × ∆𝑅𝑖𝑡 𝐸 0.41 (0.41) IT × ∆𝑅𝑖𝑡 𝐸 0.80** (0.34) NL × ∆𝑅𝑖𝑡 𝐸 2.83*** (0.46) PT × ∆𝑅𝑖𝑡 𝐸 0.54 (0.39) Constant -0.74** -0.42* Special Article | 111 (0.30) (0.22) Country fixed effect yes yes R-squared 0.60 0.81 No of obs. 102.00 102.00 Note: This table reports the regression estimates, where the dependent variable is real interest rate (∆𝑅𝑖𝑡) and the key explanatory variable is expected real interest rate (∆𝑅𝑖𝑡 𝐸). The interaction term refers to cross-country variation in the relationship between the expected real interest rate and real interest rate. Heteroskedasticity robust standard errors are reported in parentheses. Asterisks (∗∗∗, ∗∗ and ∗) denote statistical significance at 1, 5 and 10 per cent levels. Rejecting rationality for expectations of house price growth, real income and real interest rates is not unexpected, and indeed it correlates with the literature mentioned previously, which has tended to reject the hypothesis of rationality particularly in the context of house prices. The findings indicate that markets for housing, income expectations and interest rates may not be efficient, likely due to behavioural biases, information asymmetries and other market frictions. This inefficiency highlights the need for tailored interventions to stabilise housing markets and address speculative bubbles. These results also imply that models based on rational expectations may not accurately forecast future movements for house prices. To understand the non-rationality of house price growth, we next examine house price forecast error, and analyse the role of income forecast error and interest rate forecast error in explaining the variation in this. 4.4 House price forecast error and its determinants As we examine the role of household income and interest rate as key determinants of house price, we aim to understand the extent to which the forecast error in the former can contribute to the forecast error of the latter. As shown in columns (1) and (2) of Table 6, we find that forecast errors for the growth rate of income have a significant positive association with house price forecast errors, indicating that inaccuracies in income predictions lead to larger errors in house price forecasts. However, it appears to be inconclusive regarding the impact of interest rate forecast errors; while the coefficient is negative in one column, suggesting a decrease in the forecast error for house prices is associated with a higher rate of forecast error for interest rates. The relationship is not statistically significant. TABLE 6 HOUSE PRICE FORECAST ERROR, INTEREST RATE FORECAST ERROR AND INCOME FORECAST ERROR Dependent variable: ∆𝑃𝑖𝑡 𝐸𝐹 Variable (1) (2) ∆𝑌𝑖𝑡 𝐸𝐹 0.66*** 0.72*** (0.11) (0.22) Special Article | 112 ∆𝑅𝑖𝑡 𝐸𝐹 -0.13 (0.39) Constant 2.56*** 2.07*** (0.43) (0.77) Country fixed effect Yes yes R-squared 0.28 0.29 No of obs. 108.00 102.00 Note: This table reports the regression estimates, where the dependent variable is the house price forecast error (∆𝑃𝑖𝑡 𝐸𝐹), and the explanatory variables are the income forecast error (∆𝑌𝑖𝑡 𝐸𝐹) and the real interest rate forecast error (∆𝑌𝑖𝑡 𝐸𝐹). Heteroskedasticity robust standard errors are reported in parentheses. Asterisks (∗∗∗, ∗∗ and ∗) denote statistical significance at 1, 5 and 10 per cent levels. Overall, while forecast errors for income levels strongly influence the corresponding errors for house prices, the effect of interest rate forecast errors is not significant. The results suggest that, in the context of forecasting house prices, accurate predictions of income play a crucial role. When income forecasts are inaccurate, it leads to significant errors in predicting house prices. This finding aligns with the broader economic understanding that household income is a key determinant of housing demand and affordability. Therefore, any inaccuracies in income projections could have substantial implications for housing market dynamics, affecting areas such as housing affordability, demand–supply dynamics and, ultimately, overall market stability. On the other hand, the inconclusive relationship between interest rate forecast errors and house price forecast errors is somewhat surprising, given the pivotal role of interest rates in shaping borrowing costs and mortgage rates, which in turn influence housing demand and affordability. While expectations of real interest rates do appear to impact house price forecasts, the same relationship does not pertain for the forecast errors of both variables. 5. CONCLUDING THOUGHTS Studies of house price expectations have generally been somewhat limited by the absence of data on the issue. This is despite the fact that expectations themselves have been demonstrated to comprise an important factor in terms of impacting market developments. Therefore, the availability of the European Central Bank’s (ECB) Consumer Expectations Survey (CES) is particularly welcome, coming as it does at a time when house prices have started to increase following an increase in household savings, which has been evident since the COVID-19 pandemic. We believe our results in assessing house price expectations have a number of interesting implications. First of all, as noted by much of the literature that has Special Article | 113 assessed this issue, in the context of house prices, we fail to find evidence to support the rational expectations hypothesis. The tendency for households to have house price expectations that are different from rational expectations, which are often associated with efficient markets, can exacerbate the variability of house price movements. Periods of significant house price appreciation, which are maybe initially due to variations in fundamental variables in the housing market, can then be amplified by alternative house price expectations among consumers. Our estimates suggest that while actual movements in real interest rates do not appear to significantly impact changes in house prices, expected changes in real interest rates do have a significant effect on expectations of future house price movements. This underscores the importance of the signalling of monetary policy and, in particular, the growing body of literature that focuses on central bank communications (see Casiraghi and Pio Perez (2022) for more on this). It would appear this communications channel can have a significant impact on the housing market in terms of guiding consumers’ expectations. Finally, in terms of the impact on house prices, our results confirm overall the importance of consumers’ expectations regarding the general economy, given the significance of the household income variable. By changing households’ perceived potential affordability levels, expectations about the general economy is demonstrated to have the most pertinent impact on the housing market. This bears out the well-established relationship between the housing market and the general economy, and identifies the expectations channel as another means by which developments in the latter can have significant implications for the former. REFERENCES Abraham J.M. and P.H. Hendershott (1996). ‘Bubbles in metropolitan housing markets’, Journal of Housing Research, Vol. 7, No. 2, Special Issue: House Price Indices: Policy, Business, and Research Applications, pp. 191–207. Armona L., A. Fuster and B. Zafar (2019). ‘Home price expectations and behaviour: Evidence from a randomized information experiment’, The Review of Economic Studies, Vol. 86, No. 4, pp. 1371–1410. Blackey D.M. and J.R. Follain (1995). ‘In search of empirical evidence that links rents and user costs’, NBER Working Paper No. 5177, NBER. Bonham, C. and R. Cohen (2001). ‘To aggregate, pool, or neither: Testing the rational expectations hypothesis using survey data’, Journal of Business and Economic Statistics, Vol. 19, pp. 278–291. Browne F., T. Conefrey and G. Kennedy (2013). ‘Understanding Irish house price movements – A user cost of capital approach’, Central Bank of Ireland Research Technical Paper 4/RT/13. Special Article | 114 Brunnermeier, M. (2009). ‘Deciphering the liquidity and credit crunch 2007–2008’, Journal of Economic Perspectives, Vol. 23, No. 1, pp. 77–100. Campbell S.D., M.A. Davis, J. Gallin and R.F. Martin (2006). A trend and variance decomposition of the rent-price ratio in housing markets, Finance and Economics Discussion Series 2006–2029, Board of Governors of the Federal Reserve System (US). Cameron, G., J. Muellbauer and A. Murphy (2006). Was there a British house price bubble? Evidence from a regional panel (ERES), European Real Estate Society (ERES). Capozza D. and P. Seguin (1996). ‘Expectations, efficiency and euphoria in the housing market’, Regional Science and Urban Economics, Vol. 26, No. 3-4, pp. 369–386. Case K. and R. Shiller (1989). ‘The efficiency of the market for single-family homes’, American Economic Review, Vol. 79, No. 1, pp. 125–137. Case, K.E., J.M. Quigley and R.J. Shiller (2005). ‘Comparing wealth effects: The stock market versus the housing market’, Advances in Macroeconomics, Vol. 5, No. 1. Case K., R. Shiller and A. Thompson (2012). ‘What have they been thinking? Homebuyer behaviour in hot and cold markets’, Brookings Papers on Economic Activity, Vol. 43, No. 2, pp. 265–298. Casiraghi M. and L. Pio Perez (2022). Central bank communications. Monetary and Capital Markets Department Technical assistance handbook, International Monetary Fund, January. Clayton J. (1997). ‘Are housing price cycles driven by irrational expectations?’, Journal of Real Estate Finance and Economics, Vol. 14, No. 3, pp. 341–363. Cronin D. and K. McQuinn (2016). ‘Credit availability, macroprudential regulations and the house price to rent ratio‘, Journal of Policy Modeling, Vol. 38, pp. 971–984. Cronin, D. and K. McQuinn (2021a). ‘House prices and the credit driven household demand channel: The case of the Irish economy’, Credit and Capital Markets, Vol. 54, No. 2, pp. 199–221. Cronin, D. and K. McQuinn (2021b). ‘Fiscal policy and growth forecasts in the EU: Are official forecasters still misestimating fiscal multipliers?’, Review of World Economics, Vol. 157, Issue 3, pp. 453–462. Cronin, D. and K. McQuinn (2021c). ‘The (pro-) cyclicality of Government consumption in the EU and official expectations of future output growth: New evidence’, International Economics and Economic Policy, Vol. 18, pp. 331–345. Croushore, D. and S. van Norden (2018). ‘Fiscal forecasts at the FOMC: Evidence from the Greenbooks’, Review of Economic and Statistics, Vol. 100, No. 5, pp. 933–945. DeFusco, A., C. Nathanson and E. Zwick (2017). ‘Speculative dynamics of prices and volume’, NBER Working Paper 23449, NBER. De Stefani, A. (2021). ‘House price history, biased expectations and credit cycles,’ Real Estate Economics, Vol. 49, No. 4, Winter, pp. 1238–1266. Special Article | 115 Diaz, A. and M.J. Luengo Prado (2012). ‘Homeownership: Economic benefits’, in: S.J. Smith, M. Elsinga, L. Fox O’Mahony, O. Seow Eng, S. Wachter and R.K. Green (eds.), International encyclopaedia of housing and home, Vol 2, Oxford: Elsevier, pp. 439– 445. Duca, J., J. Muellbauer and A. Murphy (2010). ‘Housing markets and the financial crisis of 2007–2009: Lessons for the future,’ Journal of Financial Stability, Vol. 6, No. 4, pp. 203–217. Duca, J., J. Muellbauer and A. Murphy (2011). ‘House prices and credit constraints: Making sense of the US experience,’ Economic Journal, Vol. 121, No. 552, pp. 533–551. Duca, J.V., J. Muellbauer and A. Murphy. (2021). ’What drives house price cycles? International experience and policy issues’, Journal of Economic Literature, Vol. 59, No. 3, pp. 773–864. Fitzgerald J., S. Kenny and A. Lopez Cermeno (2021). ‘Household behaviour in Ireland, Sweden, the US and the UK under rationing’, Trinity Economics Paper tep1221, Trinity College Dublin, Department of Economics. Gallin J. (2008). ‘The long-term relationship between house prices and rents’, Real Estate Economics, Vol. 36, No. 4, winter, pp. 635–658. Glaeser E. L. and J. Gyourko (2007). Arbitrage in housing markets, NBER Working Papers 13704, National Bureau of Economic Research, Inc. Harmon, O.R. (1988). ‘The income elasticity of demand for single-family owner-occupied housing: An empirical reconciliation’, Journal of Urban Economics, Vol. 24, No. 2, pp. 173–185, https://doi.org/10.1016/00941190(88)90037-X. Keane, M. and D. Runkle (1990). ‘Testing the rationality of price forecasts: New evidence from panel data’, American Economic Review, Vol. 80, pp. 714–735. Kelly, R. and K. McQuinn (2014). ‘On the hook for impaired bank lending: Do sovereign-bank interlinkages affect the net cost of a fiscal stimulus?’, International Journal of Central Banking, Vol. 10, No. 3, pp. 95–128. Kuchler T., M. Piazzesi and J. Stroebel (2022). ‘Housing market expectations’, NBER Working Paper 29909, NBER. Leamer, E. (2007). ‘Housing is the business cycle’, paper presented at the Housing, Housing Finance and Monetary Policy Federal Reserve bank of Kansas City Symposium, Jackson Hole, WY, 31 August. Leamer, E. (2015). ‘Housing really is the business cycle: What survives the lessons of 200809’, Journal of Money Credit and Banking, Vol. 47, No. S1, pp.43–50. Liu, X. (2019). ‘The income elasticity of housing demand in New South Wales, Australia’, Regional Science and Urban Economics, Vol. 75, pp. 70–84, https://doi.org/10.1016/j.regsciurbeco.2019.01.007. Meen, G. (1996). ‘Ten propositions in UK housing macroeconomics: An overview of the 1980s and early 1990s’, Urban Studies, Vol. 33, No. 3, pp. 425–444. Meen, G. (2000). ‘Housing cycles and efficiency’, Scottish Journal of Political Economy, Vol. 47, No. 2, pp. 114–140. Special Article | 122 TABLE A2 DYNAMIC PANEL GMM ESTIMATION FOR HOUSE PRICE EXPECTATIONS AND ITS DETERMINANTS (1) (2) (3) Expected house price growth L1. Expected house price growth 0.511*** 0.345* 0.358*** (3.33) (1.87) (2.80) Expected income growth 0.512*** 0.990*** 0.918*** (7.86) (4.57) (4.28) Expected real interest rate -0.491** -0.472** (-2.40) (-2.32) House supply growth 0.003 (0.29) P value Hansen statistic 0.981 0.944 0.982 Observations 98 98 85 p value of AR(1) 0.233 0.382 0.212 p value of AR(2) 0.501 0.391 0.444 Note: GMM refers to generalised method of moments. Economic & Social Research Institute Whitaker Square Sir John Rogerson’s Quay Dublin 2 Telephone: +353 1 863 2000 Email: [email protected] Web: www.esri.ie ' & * An Institiúid um Thaighde Eacnamaíochta agus Sóisialta Cearnóg Whitaker Cé Sir John Rogerson Baile Átha Cliath 2 Teileafón: +353 1 863 2000 Ríomhphost: [email protected] Suíomh Gréasáin: www.esri.ie