I was recently researching the extent and measurement of the informal economy in Pakistan for a colleague and the econometric methods used to estimate the shadow economy are pretty interesting and intuitive. For every Rs 100 in taxes, the government receives only Rs 38 and the rest is eaten up by the tax payer, the collector and tax practitioner.
The official description of the shadow economy is ‘unmeasured and untaxed economic activity’ taking place in a country. And this is obviously poorly reflected in official measures of national income and output. Pakistan can very easily fall into the description of the ‘shadow state’ where decisions and actions are taken by an individual ruler and do not conform to a set of written laws and procedures, although these might be present. A history of our politics and the way constitutions and have been torn ad taped together with new paragraphs added here and there is a testament to this fact. In a shadow state rulers ‘manipulate external actors’ access to both formal and clandestine markets, by relying on the global recognition of sovereignty, and are thereby able to undermine formal government institutions’. This weakens bureaucratic structures and manipulates markets. These “informal commercially orientated networks” are created that operate alongside government bureaucracies (The Shadow State in Africa: A Discussion, Nikki Funke and Hussein Solomon, 2002).
A State Bank of Pakistan study last year (Arby, Malik and Hanif, 2010) makes use three different methodologies to measure the informal economy of Pakistan. The first is the monetary approach based on the idea that higher tax rates induce people to use currency for transactions to avoid tax reporting (known as the ARDL model). The assumptions of this approach are that informal economic activities are the direct consequence of high taxes. Since such transactions are mainly carried out by currency, the overall currency in circulation in the economy has two components: currency used for informal economic transactions and formal transactions. Thus the transaction velocity of money in both the informal and formal economies is the same.
The second method, the electricity consumption method is even more novel. Electric-power consumption is regarded as the single best physical indicator of economic activities in a country. Overall economic activity and electricity consumption can be observed and move together with GDP. By exploiting this relationship, one can have a proxy measurement for the overall economy and estimates of the hidden economy can be found by subtracting official GDP from the estimated overall GDP.
The last method they use is an economic model (MIMIC model) that whereby the informal economy is taken as a latent variable which on the one hand caused by a set of variables and affects other variables on the other. The model they select consists of three causes of the informal economy including tax/GDP ratio, M2/GDP and the regime durability as well as two indicators including currency in circulation (as ratio to M2) and growth in electricity consumption.
The ARDL results show that the informal economy has increased its share in the Pakistan’s economy until end of 1990s and has a declining trend since then. In the 1960s and 1970s it was below 30 percent and increased to 33 percent in the 1990s, and declined to 23 percent in current decade of 2000s. Results of MIMIC model show that the informal economy has been around 30% of the total economy in Pakistan. The growth path remains steady irrespective of the initial values. It is evident that ratio of informal economy to the recorded economy has been fairly stable in Pakistan. Thus we can conclude that the informal economy has grown with almost the same rate as the recorded economy. The electricity approach, on the other hand, shows that the extent of the unmeasured economy was less 5 percent during 1970s which then increased sharply until 1990s and then remained stagnant. However, this approach may not reflect the actual performance of the economy as official numbers of electricity consumption do not incorporate self- generation of electricity by economic agents in the mid 1990s onward due to crisis in official sector of power generation and distribution in Pakistan.
The results are generally close to those obtained by other studies on Pakistan but their remain some concerns as other studies show a rising trend up to the end of 1990s while some show a declining trend. M. Ali Kemal (Pakistan Institute of Development Economics, 2007), writes that if there was no tax evasion, budgets balance might have been zero and positive for some years and we would not have needed to borrow as much as we have.
The impact of the underground economy is significant to the movements of the formal economy, but the impact of formal economy is insignificant in explaining the movements in the underground economy. In the long run, underground economy and official economy are positively associated. Kemal (2007) estimates that the black economy ranged between Rs 2.91 trillion and Rs 3.34 trillion (54.6 percent of GDP to 62.8 percent of GDP respectively) in 2005 and tax evasion ranged between Rs 302 billion and Rs 347 billion (5.7 percent of GDP to 6.5 percent of GDP respectively) in 2005. Underground economy and tax evasion were increasing very rapidly in the early 1980s but the rate of increase accelerated in the 1990s. It declined in 1999, but reverted to an increasing trend until 2003. It declined again in 2004 and 2005. This supports Arby, Malik and Hanif’s ADRL model.
A study conducted by The Lahore University of Management Sciences in 2003 showed that the Rs 720 billion collected in tax in 2005-06 was only the 38 percent of the Rs 2 trillion that could have been collected. Two-thirds of the income earned in the shadow economy is estimated to flow into the official economy. The growth of the informal economy affects everyone.
The State Bank of Pakistan study can be found here.