Our weak savings performance is perhaps the only issue upon which economists watching Turkey agree. Among them, some consider low savings as driven by low growth, and state that when other factors limiting growth are eliminated, the savings ratio will improve on its own.[1] Others argue that low savings is the main cause of low growth, and thus improving savings is the precondition for improving growth. [2] I don’t know which one is my position yet, but I am ever more convinced that there is a causal relationship between the unplanned urbanization in Turkey and the savings rate problem, and thus, growth. Today I would like to state three observations on this causality.
The first observation is well known. Middle-income countries save 25-30 percent of their GDP while high-income countries save 20 percent. Over the last three decades, Turkey, as a middle-income county, has saved even less than the average among high-income countries. Below, Figure 1 illustrates the abnormality of Turkey’s saving performance. Turkey’s savings rate is in between that of high-income countries, including the US, and is way below that of Korea. Another striking point lies in the ups and downs in the rate. Upwards movements are quite rare, but a downturn occurs cyclically every 4-5 years. What causes this? I thought of election cycles, a usual suspect with this sort of thing. Indeed, a closer look at the figures reveals that the cycles are related to local elections. The drops in the savings rate are particularly remarkable in 1989, 1994 and 1999, each of which were local election years.
Figure 1 – Savings rate across the world (share in GDP, %) middle and high income countries, Korea, USA and Turkey, 1990-2010
Source: World Development Indicators (WDI), The World Bank
But let me stop torturing this limited data and ask a question: How can local administrations in Turkey influence saving behavior? If owning a home and the changes in housing prices affect saving behavior, which they do, [3] then Turkey’s local administrations have a magic wand that can influence savings. Remember that particularly after the 1980s, millions of rural migrants built gecekondus (squatters) in urban areas. Local administrations, desperate for votes, integrated these into the legal system during election periods, thus raising their value. How did this impact saving behavior? I think the impact was negative. But of course it is too soon to give a concrete answer, as there have been no studies that investigate the issue.
Before stating the second observation, let me ask another question. The latest statistics suggest that 63 percent of Turks live in their own homes. How do you think this rate compares? I don’t have an answer for that, but Figure 2 below demonstrates an interesting picture. Home ownership in Turkey is on par with that in the US, and is higher compared to France, Germany and Switzerland. In Korea, which has a higher per capita income than Turkey, only 54 percent of the people live in their own homes. The home ownership ratio in Spain, Italy and former soviet Bulgaria are higher than that in Turkey. So, it appears that the ratios are influenced by unique cultural factors, political dynamics, urbanization patterns and income/wealth distribution policies.
Figure 2 – Home ownership in selected countries (%)
Source: TURKSTAT 2009 Household Budget Surveys, US Census Bureau, KoreaStat, Wikipedia
Having assessed Figures 1 and 2 together, I think that instead of saving, Turkish people get a share of urban lands, with the help of local administrations and the toleration of the central government. In this unique system of ours, it is perfectly rational that people prefer spending their money on zoning privileges that earn high returns, rather than running after crumbs from their savings. Think about it: You purchase a 50-square meter piece land on the cheap. Then you obtain the title deed and eventually, the municipality permits the construction of a 3 to 5-floor building on your land. Returns on savings within the financial system are taxed, whereas those on zoning can be kept completely off of the record. Maybe this is why the Turks, who traditionally save gold, like to say that the streets of Istanbul are paved with it.
The change in home ownership rates is, of course, also important. Though the ratios in Turkey are high relative to many other countries, the value and the quality of homes are just as important. Unfortunately, there is not much recent data about these. The statistics in Figure 3 are compiled from TURKSTAT’s Household Budget Survey, the most reliable source in this regard. The figure shows how home ownership rates in urban Turkey changed between 2002 and 2009. The rates for the last couple of years are quite odd, assuming no calculation errors were made. The ratios of urban households living in their own homes decreased from 65 percent in 2003 to 55 percent in 2009. Despite the giant construction attempts by the Turkish Housing Development Administration and private companies, the ratio of tenants has been increasing.
Figure 3 – Rate of households living in own home, urban population, 2002-2009, (%)
Source: TURKSTAT Household Budget Surveys
Someone has to explain this trend and its correlation with savings. Hoping to initiate a strong research agenda on the issue, I would like to finish with a couple of questions:
- Is the squatting phase in Turkish cities really coming to an end? If yes, will urban transformation increase the ratio of tenants and reduce that of homeowners? What will be the repercussions of these new trends on savings and the current account deficit?
- How does the correlation between the hike in housing prices and the drop in the propensity to save apply to Turkey’s cities? For instance, what proportion of the urban population is inclined to sell homes that rise in value and buy cheaper ones instead? What is the share of property owners who recently have started to earn rental income on more than one home?
- How long will it be before the main duty of the municipalities is to create fictive housing price bubbles and make property owners richer? Try assessing this from the perspective of social welfare rather than of the welfare of the lucky few. Can’t anti-speculation policies help to raise municipal revenues – for accelerating subway investments for instance – prevent asset price bubbles, and enhance savings? The Korean government declared the entire residential area of Seoul to be an overheated speculation zone, increased taxes on property sales and designated the tax basis on the market value instead of the current value as set on the title deed. But how would Audi Q7 sales and political party finance be affected If Turkey’s cities adopt such an approach?
I want to stop here. The questions above and others like them have to be answered in order to better grasp the savings dynamics in Turkey. The statistics make Turks look like Americans who live in squatter settlements. I hope we break this chain.
[1] For example, please see Dani Rodrik, Saving Transitions, World Bank Economic Review 14, pp. 481-507 (2000).
[2] For example, please see Kemal Derviş: http://www.cnnturk.com/2010/ekonomi/genel/05/03/dervis.turkiye.tasarruf.oranini.artirmali/574638.0/index.html
[3] I highly recommend the TÜSİAD-EAF report by Caroline Van Rijckeghem and Murat Üçer for the savings problem in Turkey and a general literature review on the issue. “The Evolution and Determinants of the Turkish Private Saving Rate: What Lessons for Policy?” (ERF Research Report Series No. 09-01)
*Esen Çağlar, TEPAV Economic Policy Analyst, http://www.tepav.org.tr/tr/ekibimiz/s/25/Esen+Caglar
