![Sample our Geography journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5937/TOC-Subject-Sample-2021-Geography.jpg"})
Abstract
The 1990s saw a considerable fall in young adult homeownership rates in Britain. There is a concern that the future might hold further falls as a result of reforms to financing higher education. Using estimates from a housing tenure choice model, this paper conducts micro-simulation analyses to assess how this change could affect young adult homeownership transitions. The simulations reveal that increased student debt levels and the new repayment profile and their interaction with lender-imposed borrowing restrictions delay a first-time homeownership transition. The extent of the delay primarily depends upon the expected earnings profile, but lender criteria and general rises/falls in real house prices are also important. The analysis suggests that there will be much greater variation in the timing of house purchase between different groups of future graduates, brought about by increased graduate earnings heterogeneity, homeownership affordability schemes targeted toward specific types of households, and parental financial assistance.
Acknowledgements
The author gratefully acknowledges the financial support of the ESRC through research grant RES-000-22-0487. The British data used in this paper were made available through the Data Archive. The data were originally collected by the ESRC Research Centre on Micro-Social Change at the University of Essex. Neither the original collectors of the data nor the archive bear any responsibility for the analyses or interpretations presented here. The author also wishes to thank Geoff Meen, the previous Managing Editor Alex Marsh and the three anonymous referees for their constructive comments.
Notes
1 The 2003 figure is discounted using a HM Treasury discount factor of 3.5 per cent, which gives a present value in 2003 prices.
2 Although the HE reforms came in force in 2006, further changes were implemented in 2007. The changes mainly involved up front support for students. This will affect students starting university in 2008/09. The 2006/07 reforms are explicitly considered in Andrew (Citation2007).
3 Since 2006/7, the maximum maintenance grant available to non-mature full-time students has generally risen in line with prices from £2,700 to £2,906 in 2009/10.
4 The focus here is on a 3-year degree course as this is the type of degree the majority of young people will read.
5 Note that graduates from the highest-income families end up with the least debt because they are restricted in the amount that they can borrow.
6 Students in higher education tend to return to the parental home at the end of the academic year (Ermsich, Citation1999).
7 This user cost expression assumes that after tax cost and debt equity finance are the same.
8 Because of high transactions costs, this may result in obtaining predictions for households over consuming homeownership housings services.
9 Median FTB ratios are matched to renter households and those that purchased a home in that year and Median Repeat Purchaser ratios are matched to households who were homeowners in the previous year.
10 A problem with this set up concerns the treatment of the social rented sector. Although it should be modelled, access to it is rationed by local authorities and not available for renting to most young people. Given that the main switch away from homeownership is to private renting (Andrew, Citation2004), and as private renting is primarily a short-term housing tenure, the alternative option for social tenants would be owner occupation if they could afford it. Representing the alternatives to rent as one choice eases the estimation problems considerably.
11 The endogenous dummy variables in the household formation equation are the presence of a spouse and the presence of a child. In the housing tenure choice equation, they are being married and being income and wealth constrained. Time means of weakly exogenous variables are used as proxies for unobserved heterogeneity.
12 This result is available from the author on request.
13 Cross-section models by nature are not able to pick up the incidental parameters.
14 Median loan-to-value ratio in 2004.
15 As revealed later, the level of female earnings means that the loan value from the joint income multiple is greater than the single income multiple plus the second income.
16 IFS commentary report No. 98 and Dearden et al. (Citation2008).
17 This is in line with HM Treasury which uses a 2 per cent rate of long-term earnings growth and the analysis conducted by Dearden et al. (Citation2008). Ignoring real earnings growth from productivity gains exaggerates the effects of student debt, as Dearden et al. (Citation2008) argue.
18 Using the OECD reported average of 1700 hours per year.
19 Females with higher educational qualifications tend to delay having children.
20 These are in line with the IFS calculations.
21 The male graduate is assumed to have previously headed a household in the final year of study. Otherwise, the household formation probability will be lower.
22 The additional delays caused by increased student debt are very similar to the results presented previously in Andrew (Citation2007), where simulations were undertaken using the reported real earnings distributions without productivity gains (i.e. zero real earnings growth). The main difference concerns the time taken to overcome the credit constraints within each regime.
23 The empirical models account for such changes and 100 per cent mortgages through simple instrumentation. See Andrew (Citation2007) for more details.
24 Joint probability greater than or equal to 50 per cent or the highest figures if it is under 50 per cent.
25 In current prices, the income thresholds are £9000 and £15 000 for the 2003/04 and new regimes respectively. The thresholds undertaken in the simulations are expressed in 2006/07 prices.
26 Relative housing costs are assumed to be unchanged (stable). This means that housing in general (renting and owning) becomes more expensive. Due to the partial nature of the analysis, the joint probabilities fall over time.
27 Scenarios assuming long-term real house price growth at 4 per cent per annum were also examined. There is no difference in the delay caused by the introduction of the new regime but the ability of the representative households in each regime to meet lender requirements is delayed by a year. As expected, the joint probabilities decrease more dramatically over time. Note that as real house prices rise, the desired level of housing services is less than proportionately reduced which explains why the delay is not so long. If the standardised dwelling measure was used instead, then the delays would be more substantial. The results are available from the author on request.
28 Social HomeBuy is tied to properties owned by a local authority and registered social landlords. It has the added incentive of providing a discount to the purchase price. New Build HomeBuy is applicable only to newly built property.
29 Key workers refer to employees in jobs that benefit the local community such as police officers, nurses, teachers etc.