I’ve found some validation for my earlier hypothesis that neighborhoods which opt out of the broader housing market will also opt out of speculative consequences including gentrification. From Alan Walks and Martine August, “The Factors Inhibiting Gentrification In Areas With Little Non-Market Housing: Policy Lessons From The Toronto Experience” [Urban Studies, 45(12), November 2008, downloadable from neighbourhoodchange.ca]:

“Perhaps the most important reason why the embeddedness of the Portuguese and Chinese communities factors large in inhibiting gentrification is their control over a significant proportion of the housing stock and dominance in the local real estate sector. In both cases, houses purchased within the community tended to stay in the community and were often converted for multifamily use using their own or bartered labour. In most cases, tenants were sought from within the community, as proficiency in English remained marginal at best (Teixeira, 1998, 2000; Chan, 2006)…

“As already noted, many immigrant communities, like the Portuguese in Brockton and the Chinese in South Riverdale, finance their housing purchases through family connections and their renovations via sweat equity (Murdie, 1986, 1991). This meant that the ethnic communities were able to raise capital during a period in which inner-city housing as a whole, and these neighbourhoods in particular, were devalued (and/or considered too risky to insure) by institutionalised finance capital. The influx of ethnic capital, and the conversion of many properties to multifamily use, had the positive effect of limiting devaluation and thus the rent gap in the face of de facto redlining, therefore reducing incentives for demolition and redevelopment (Smith, 1996). Much of the increase in rental in both neighbourhoods can be attributed to the conversion of properties to multifamily by ethnic owners and much of this housing was rented to tenants from within the community as many were uncomfortable having to deal with tenants in English (Teixeira, 2007; Chan, 2006). Likewise, ethnic contacts are often sought out first when properties are put up for sale (Murdie, 1991) and, considering that demand for housing from within both the ethnic communities remained strong well into the 1990s, this would have meant that a significant portion of the housing stock was effectively removed from the capitalist property market available to gentrifiers (although it would still have been available to ethnic speculators)…

“In both cases, the reliance on ethnic sources of housing finance capital and labour appears to have played a distinct role in maintaining a measure of ethnic control over a section of the housing stock, which acted as a complementary stabilising force for the community at a key time in its evolution. Thus, a third policy recommendation would be to support the usage of ethnic and/or non-market or non-profit sources of housing finance and/or non-market programmes that can match vacant properties to new residents, thus largely bypassing the traditional housing market and in turn reducing, if not preventing, speculative real estate activity and gentrifiers’ access to key properties. Such a policy need not be targeted at ethnic communities—embattled working-class communities could also benefit from such a system… Of course, the extent of the phenomenon (of ethnic housing finance) and its precise effects in obstructing gentrification in our two case studies remains somewhat of an unknown. This is an area that clearly warrants further empirical exploration by gentrification researchers.”

Similarly, City Council actions to inhibit speculation and move housing off the marketplace helped: “in South Riverdale the city council specifically adopted policies to prevent ‘white-painting’ in the neighbourhood and protect affordable housing. While short-lived (from 1974 until 1977), a municipal ‘speculation tax’ was implemented across the city and the City’s non-profit housing corporation (City Home) was instructed to acquire selected apartment units and houses as a complement to its stock of projects and limited equity co-operatives (City of Toronto Planning Board, 1977, pp. 22, 50). Although the number of houses acquired in this way in South Riverdale was small (55 units), it was a disproportionately high share compared with the rest of the inner city and sent an important signal to the development industry that the city intended to protect low-income housing in the area.”

Other factors that may have contributed include tackiness, with the appropriate “historic preservation” response from The Powers That Be: “Perhaps even more important was the way that a significant proportion of the housing stock was renovated by the incoming southern European communities in west-central Toronto… The extent of dislike for such mediterraneanised facades is revealed by gentrifiers’ attempts to ban the use of ‘angel brick’ under the rubric of heritage preservation (Caulfield, 1994, pp. 204–207).”

They also discuss how delayed deindustrialization allowed the neighborhoods to maintain their working-class character longer — after all, industrial activity (a) creates housing demand by the working class and (b) has environmental externalities that the gentrifying classes dislike/avoid.