Niseko: Isolating Age and Location as Property Price Determinants

George Maxwell & Leo Kitchell
George Maxwell & Leo Kitchell
11 min read

In our last edition of the Niseko brief, we outlined how newer developments are being put on the market at a higher price point indicating that Niseko’s real estate market is changing to accommodate a wealthier demographic. We concluded that Niseko is becoming more luxurious and will continue to be, based on the current and planned construction projects in the area. 


Niseko-Mount-Annapuri-Niseko-Japan-from-aboveFig.1: Niseko's property market fundamentals suggest it is becoming a more luxury holiday destination


While newer properties draw a premium, it is unclear how much of that premium is attributable to the age of the development rather than to superior location. Conversely, it is unclear to what extent location alone increases prices. To illustrate the isolated effects of each variable, we will break down the square meter price of the Alpen Ridge, from our previous article, into its component parts. In other words, how much of the Alpen Ridge’s $15,972/m2 listing price is due to its age and how much is due to its proximity to the lifts? 


Alpen-Ridge-Yotei-View-Niseko-JapanFig.2: It is unclear to what extent the price/m2 of Niseko's Alpen Ridge development, pictured, is attributable to location


It is no revelation that location, especially proximity to ski lifts, is a strong price determinant in resort towns. Holidaymakers will pay for convenient access to activities in addition to greater luxury, making properties in close proximity to ski lifts more desirable. Correctly interpreting the stand alone impact on price this proximity has on property, requires eliminating the confounding effects of other variables. For example, in Niseko, old developments in premier locations are being replaced by new luxury buildings, making the effects of each factor unclear when they are not specifically isolated. Put simply, it is important to isolate both age and proximity to lifts as price factors. Otherwise, the newest properties may just be being developed in the most expensive locations. This would yield an incorrect conclusion that the age of the property is impacting the price of the property when it is truly attributable to location. 


Fig.3: Total Listed Prices in Hirafu Village


This is neatly demonstrated by graphing our compiled data on median price per square meter according to property type. On first glance, it appears that condominiums and penthouses demand a significantly higher price per square meter than other types of property. However, using a multivariable linear regression to isolate the effect of property type on price per square meter, the type of property was found to be a negligible price determinant, except among houses. Houses were found to demand $2,200 more per square meter than other property types, with a 99 percent confidence level. For condominiums, penthouses, and apartments, we did not find correlation at a 95 percent confidence level, the standard in most statistical analysis. While a logarithmic trend is useful for describing the premium demanded by ski-in-ski-out properties and the diminishing effect of distance to the lift at large values, it overestimates the effect of proximity to the lifts at the greatest distances. We hypothesise this trend loses applicability for properties situated in Outer Hirafu*, given the necessity of driving and the wide variety of building quality.

* Note: We define Outer Hirafu as those properties located at least 500 meters from central Hirafu along the roads leading northeast and southwest out of the town, including Izumikyo and Kabayama. 


Fig.4: Properties in Niseko closer to lifts were found generally to demand a premium, except those in 'Luxury' resorts


Having established the importance of property type, we will now discuss the isolated significance of proximity and age. In Hirafu, we found, with a high level of confidence, that properties demand a premium the closer they are to a lift. For each 10 percent increase in distance of the development from a lift, the property loses approximately $250 per square meter.

We will use an analogy for clarity: take two buildings. Both are the exact same except for distance to the lift. Building A is 50 meters away and Building B is 55 meters away. According to our research, Building A would be $244 per square meter more expensive than Building B. 


Niseko-lifts-japan-hokkaido-snowFig.5; Proximity to lifts was found to be a significant price determinant for property in Niseko


As with any logarithmic trend, this $250 per 10 percent only holds for relatively small percentage changes. The $250 per 10 percent is 95 percent accurate, while $500 per 20 percent is 91 percent accurate, and $750 per 30 percent change is 87 percent accurate. 


To demonstrate this scaling, we will introduce a third comparison, Building C. Building C is located 100 meters from the nearest lift, 100 percent further than Building A. According to our research this property would sell for $1778 less per square meter than A—decidedly different from what the rule of thumb suggests.** 

** Note: Calculated using “difference in prices = (-2565.8 * ln(Distance A)) + (-2565.8 * ln(Distance C)).” 


map-of-niseko-united-illustrated-japan-hokkaidoFig.6: An illustrated map of the Niseko skiing resort with Hirafu village (bottom left) at the base of Mt. Niseko Annapuri 


Our regression equation was price per square meter = $33,884.7 + (Is Condominium*-613.3) + (IsHouse*2024.9) + (Is Penthouse*1667.4) + (-2565*ln(Distance From Lift in Meters)) + (-3789.4*ln(2022-Construction Age). It is important to note that the 'Is Condominium' and 'Is Penthouse' variables were not significant, and can be omitted for ease of use. Interpreting this equation, we can say that a hypothetical building will draw $33,885 per square meter before taking into account age, distance from lifts and property type.


Our regression’s  adjusted R2 was 0.803, meaning that 80.3 percent of the variation in price per square meter is explained by our model. The remaining variation is due to variables like standard of development, variety of amenities and quality of marketing materials, which are hard to quantify. 




USD/SqM Effect per 1% Increase

Statistically Significant?


Is a House

2,024 increase if property is a house



Distance to Nearest Lift




Age of Building (Since 2022)






While it’s no surprise that location is important, further analysis of our dataset reveals age is arguably a more significant price determinant for Niseko property. Our analysis indicates that a building 10 percent older will sell for $361 less per square meter. Because some of our data includes pre-listed properties that will sell in the future, our calculation for age is “2022 - year constructed”. Therefore, a building constructed in 2011 is calculated to be 10 percent older than one built in 2012. 

*** Note: we have chosen 2022 as the end year so readers can forecast expected prices, if desired.


Hiyuko-development-niseko-snow-hokkaidoFig.7: Newer developments in Niseko were also found to demand a premium


It is important to note that we are not implying that buildings lose $361 per square meter for each 10 percent they age. Instead they list for $361 more for each 10 percent newer they are. While this difference may seem contrived, it is because we posit that building age is more reflective of quality of development than building degradation. Therefore, as degradation only explains a small percentage of price changes, it is better to think of new buildings gaining value, as we don’t expect the current listings to depreciate at the $361 per 10 percent rate.


Returning to our example from our last article, let’s break down the Alpen Ridge’s $15,972/m2 listing price. The Alpen Ridge was built in 2009 and is located 30 meters away from the nearest lift. Our hypothesized price per square meter would then be $33,844.7−2565*ln(30)−3789.4*ln(2022−2009) = $15,400, an underestimate of approximately 3.5 percent****.

**** Note: This difference is due to the 19.7 percent variation not explained by our model, as mentioned above.


To further illustrate how location and age interact, we can use our regression to find an equivalent modern listing. Instead of using an old building with close proximity to a lift, we will now choose a set of values for a new, but distant development. We find that a building constructed in 2019 and located 262 meters away from the lifts will draw the same listing price as the Alpen Ridge. Although not a perfect comparison, the 2019 Haku Villas, located 338 meters from the lift is one such listing, priced at 388 less per square meter. 



Fig.8: Graph demonstrating that properties closer to lifts in Niseko demand a premium


Our regression is visualized in the graph above, in which properties closer to the lifts are dark blue. Examining this graphic, it becomes clear that the properties above the trend are located closer to the lifts, while those below it are further away. However, it is important to note that this trendline plots the effect of age without taking into account the other variables, as our regression did.  


Returning to the general trends of Hirafu, we conclude that both age and proximity are highly influential factors in determining the price of a property. Having discussed the broad based trends present across all Hirafu development, our next blog will break down the composition and price trends of Hirafu real estate. Check back for more soon!

Recent Articles

Sign up to receive the Propeterra's newsletter and exclusive property news and updates. You can unsubscribe at any time by clicking on the unsubscribe links in our emails.



posts by tag

See all

Market Cover_Emerging Markets-1


Market Cover_Frontier Markets-1


Market Cover_Special Situations-1-1


Market Cover_Developed Markets-1


Recent Articles

3 minutes read

Mongolia’s Active Diplomacy in a Post-COP World

In November, Mongolia celebrated sixty years as part of the United Nations (UN). For many countries, this means little, as such milestones, by definition, come and pass on a regular basis. It assumes greater importance for Mongolia, though, given the particular geopolitical framework it has to work with. By dint of geography, this vast country has an unusual position. In a strategically important location between Russia and China, it is relevant to not only its own prosperity but also the broader world. Since independence, Mongolia has tried to maintain constructive relations with those it shares borders, but also to project beyond North Asia. Its successful third neighbour policy, takes in partners from South Korea to Japan, India to the United States.


4 minutes read

Investing in Good or a Good Investment?

How can we judge the environmental impact of property development?
‘Sustainable’, ‘green’, ‘eco-friendly’, ‘ethical’: too frequently seen as - and
used as - zeitgeisty buzzwords to allay any discomfort investors may feel
over looking for the next most profitable venture. But anyone with even
half an eye on the news and weather reports can see that environmental
sustainability can not be merely an optional consideration when planning
developments or deciding where to invest – especially when it comes to
real estate, given that almost 40% of carbon emissions are from the built
environment. But in a world with such interconnected systems and
ecologies, how can we predict impacts that may be surprising and far

AidData, a research lab at the College of William and Mary, Virginia, has
been developing the technology and resources to do just that. Their team
includes economists, political scientists, geographers, developers,
program evaluators, policy analysts, and communications professionals,
and they work with 21 different countries in Asia, Latin America and
Africa, using data and hard evidence to improve policy and development
outcomes, helping policymakers to move towards the UN’s 17 Sustainable
Development Goals.

Using satellite, household survey, economic, health, and other spatial
data, along with machine learning analysis and super-accurate data on
geographical boundaries, AidData have created various tools to analyse
the impacts of various development programs, and to predict whether
similar programs would work in other locations. Their Geospatial Impact
Evaluation methods enable fast and rigorous analysis of outcomes for
localised development projects, replicating the standards of a randomised
controlled trial. Development agencies and governments are able to very
quickly discover whether some certain infrastructure project, for example
irrigation systems or village road improvement, had the desired effect on
outcomes like infant mortality, poverty, or deforestation. The highly
localised and granular data can also show whether outcomes are uneven
from region to region, allowing more targeted development in future.
Is there a way we can apply the same rigorous impact-analysis to our
small-scale investments that AidData is enabling international
development agencies and world governments to apply to their
investment in aid, infrastructure, health, and other initiatives?

Environmental, Social, and Corporate Governance are becoming ever
more important criteria when analysing an investment’s risks and
opportunities. In many emerging markets such as Malaysia and Indonesia,
the Islamic finance sector is growing more and more influential too, with
its strong alignment to ESG principles. Propeterra is looking at ways to
adapt the Geospatial Impact Evaluation methods created by AidData and apply them to affordable and social housing projects in emerging and

Social Impact

4 minutes read

Addressing the Global Affordable Housing Shortfall


Housing Crisis